Podcast: Download (Duration: 41:57 — 48.3MB)

The fastest way to launch an ecommerce product today is to build the audience first and the product second. On a recent My Wife Quit Her Job podcast episode, my co-host Toni and I broke down a seven-figure product launch from someone in the ECF forums who spent zero dollars on ads. She built a community for a year before she ever mentioned her brand, and when she finally launched, her people bought.

That is the entire playbook in one sentence. Meta ad costs have gotten so brutal that seasoned sellers are quitting the platform, and the old “build it and they will come” model only ever worked because ads were cheap. If you do not already have a jackpot to burn on paid traffic, community-first is the only launch strategy that still works at a small budget.

Below is the framework we walked through: why the shotgun launch fails now, how to build the audience in the months before your product ships, and the four content styles that work best if you hate being on camera.

Key takeaways

• Most sellers launch backwards: they finish the product, then look for customers. Reverse it. Build the community first, then launch to people who already trust you.
• Meta ad costs have risen to the point where forum sellers are pulling ads entirely. Paid launches now require a much larger budget than they did even two years ago.
• Your product’s production window (usually 3 to 4 months) is free runway. Start posting content the day you place the manufacturing order, not the day the boxes arrive.
• Four content styles work for ecommerce founders who are not natural performers: the founder journey, entertaining content that piggybacks on the product’s topic, voiceover-only “hands and pans” video, and short off-the-cuff phone clips.
• If you refuse to be on camera at all, you can still make it work, but your niche knowledge or your visual craft has to be a real outlier. Being on camera is the fast track for a reason.

Why does the “build it and they will come” launch fail in 2026?

The “build it and they will come” launch fails because paid traffic is no longer cheap enough to force sales on a store nobody knows. Meta ad costs have climbed for years, and sellers on the ECF forums (a private community that requires a million dollars in sales to join) are openly saying they cannot afford to advertise on Meta anymore.

There was a brief window when Temu and Shein got hit by tariffs and CPMs dropped for a bit. Costs have come right back. If you launch a product into that environment with no audience and no jackpot to spend, you land in a garage full of inventory wondering why the traffic never showed up.

The Field of Dreams model held together for a while on Amazon, but even Amazon has been squeezing sellers for the last ten years. There is no cheap acquisition channel left for a cold start.

What is the community-first launch playbook?

The community-first launch playbook is to spend the months leading up to your product creating organic content for the exact audience you eventually want to sell to. The person from the ECF forums did this for a full year, never mentioning her brand once, and hit a seven-figure launch with zero ad spend. Her audience was already loyal by the time the product existed.

That is the extreme version, and you do not need to wait a full year. Product production alone usually takes three to four months, and that window is free runway you are already burning.

Start posting the day you place the manufacturing order.

The community is the moat. If you build a group of people who care about the topic your product sits inside, converting them from a community member into a customer is a much shorter jump than converting a cold Meta impression.

How much lead time should I give an audience before launching a product?

Give your audience at least the length of your production window, which for most physical products is three to four months. That is the minimum viable runway. Anything less and you are running the same shotgun launch, just with a few TikToks stapled to the front.

The seven-figure example built for a full year. Christina, one of our course students who sells nail clippers with a built-in magnifier for dogs, built her Instagram and TikTok audience with pet content for a while before her product went live. Same pattern, different timeframe.

If you are debating whether to set up your Shopify store first or start creating content first, the answer is: do them in parallel. Product goes into production on one side, content starts on the other side the same week.

What content should I make to build an audience before launch?

The four content styles that reliably work for ecommerce founders are the founder journey, entertaining content that piggybacks on your product’s topic, voiceover video that shows the product or process without your face, and short off-the-cuff phone clips. Pick whichever fits your personality and your product, and stop trying to do all four.

The scrunchie founder I follow is a good example of the first style. She started by pulling out her phone before work every day and filming a one-minute unedited clip of her hand-sewing scrunchies. Just her, a $400 sewing machine, and her actual workday.

By the time she scaled up to three machines, her audience was invested enough that people who would never buy a scrunchie for themselves were rooting for her to succeed.

The founder journey works because it shows movement and progress. Here is my back room. Here is the first order.

Here is the fabric I picked and why. Here is what broke this week. People get addicted to watching a real business get built in real time.

What is founder-journey content and how do I make it?

Founder-journey content is short video of you doing the actual daily work of running your business, narrated in your own voice. You show the good and the bad, but you present the best genuine version of yourself, not the worst.

Being genuine does not mean showing the ugliest twenty seconds of your day on loop. You do not need to film yourself rethreading the sewing machine 22 times, and you do not need the meltdown when an order arrives wrong.

Show that an order arrived late and how you handled it. People need a reason to have a favorable view of your company, and you control which slice they see.

What if I am not entertaining enough for on-camera content?

If you are not personally entertaining, pick a product where the topic itself is entertaining, and let the topic carry the content. Pets, cooking, satisfying process video, and craft demos all pull views on their own without a charismatic presenter. Christina is actually very entertaining, but the pet content works even when the pets are the main character.

The gut check I recommend: before you decide you are the entertaining founder type, ask five friends and tell them to be brutally honest. Most people who think they are funny on camera are not, and the videos do not get engagement. That is a survivable problem if the subject matter does the heavy lifting.

Melissa on TikTok built her entire audience teaching people to fold towels and sheets. The video is folding laundry, and that is the whole thing.

Pick a topic viewers will watch and engage with regardless of who is on screen.

How do I make ecommerce content if I refuse to be on camera?

If you refuse to be on camera, the two best options are voiceover video (hands-and-pans style where the camera is on the product or the process) and hiring a brand ambassador or UGC creator to be the face for you. Both work, but you are giving up the fast track.

Kelly and Renee in our course both make crocheting content this way. Everything is filmed hands-and-pans, close on the work, without ever showing a face.

They write a script, use AI to help sharpen it, record the voiceover first, then cut in the visuals to match. It reads as founder-led without ever showing the founder.

The honest tradeoff: every platform now asks you to check a box confirming whether your video is AI-generated, and the trust signal of a real human on camera is going up as AI content floods the feed. If you go faceless, your niche knowledge or your visual craft needs to be an outlier for the algorithm and the audience to latch on.

How do I get past the mental blocks that stop me from filming?

Remove the specific hurdle that is stopping you, one hurdle at a time. Toni’s block was tech (something breaks every time she records, and she never trusts the setup) so she rents a studio for one day every three months, records 15 to 20 videos in a three-hour session, and hands the files to an editor. Zero setup, zero mic-not-plugged-in disasters.

My block was different. Setting up lights and audio used to take me 20 minutes, and any time I did not feel like doing 20 minutes of setup I did not record.

So I built a permanent recording rig where I hit one button and start. The barrier had to go to zero.

Our friend Jim Wang solved it a third way. He records directly in the TikTok app on his daily walk, one or two sentences at a time, deleting the clips he does not like and stitching the good ones together in-app.

The platform is the studio. He needs no editing software, no separate camera, and no permanent rig.

What is the cheapest ecommerce content setup that actually works?

The cheapest content setup that works is a phone you already own and a wireless lav mic from Amazon for around $25 to $30. Total investment is under $30 because you own the phone.

The four creators I follow for tech, AEO and AI content all shoot this way. Phone on a tripod, cheap lav mic, auto-generated captions, and that is it.

One of them is basically at his day job kicking back and recording 30-second clips about investing. Millions of followers, zero production.

More complicated equipment introduces more things that can fail on you at the worst moment. If you already know you are not a technology person, a phone plus a $25 mic is more reliable than a real camera rig you will fight with every session.

Do I need a teleprompter to record short-form video?

You do not need a teleprompter, but it is one of the highest-ROI hurdle-removers you can buy, and current models cost about $25 on Amazon. If your block is going blank or stumbling mid-sentence, a teleprompter fixes it immediately.

I use one because I lose my voice after 20 minutes of projecting, and I cannot do the “record two sentences, stumble, delete, retry” TikTok-style flow. With a teleprompter and a script I can knock out 15 videos in one take.

If your block is different (setup friction, camera-shyness, blanking, no script), buy the tool that removes your specific block. Do not buy a full studio kit before you know which piece is actually stopping you.

Why does a pre-launch community drive product sales without ads?

A pre-launch community drives product sales without ads because it hands you a warm audience that already trusts you and is already reachable, so you skip both the persuasion cost and the acquisition cost that would otherwise force you onto paid platforms. The community is a warm launch list you built for free over the months your product was in production.

The high-net-worth divorce lawyer on my TikTok feed is a great illustration outside of ecommerce. He records two-minute takes on one specific topic he is an obvious expert on (“why I would not want to be married to another lawyer”), and he is building authority with high-net-worth individuals who might one day need him. If he decided to sell a divorce-related product tomorrow, it would sell.

Same thing works in ecommerce. Shalene built a fitness audience early, pivoted with them as they aged, and now sells wellness products, memberships, and courses to the same people who followed her for workouts a decade ago.

Derek Halpern did it with Truebani by partnering with an influencer who already owned the community. In every case the community came first and the product decision came second.

Frequently asked questions

Do I need a big budget to launch an ecommerce product in 2026?

No, but only if you build a community first. Meta ad costs have risen to a point where seasoned sellers on the ECF forums are pulling ads entirely, so a paid cold-start launch now requires a much bigger budget than it did two years ago. The zero-budget path is to spend the months before your product ships creating organic content for the audience you eventually want to sell to.

How long before launch should I start creating content?

Start creating content at least three to four months before launch, which is the typical production window for a physical product. That window is free runway you are already burning while manufacturing. The seven-figure ECF launch we discussed built her audience for a full year before ever mentioning her brand.

Can I build an ecommerce audience without ever appearing on camera?

Yes, but it is the slower path. The two workable no-face formats are voiceover video (close-up “hands and pans” shots of the product or process) and hiring a brand ambassador or UGC creator to be the face. On-camera founder content is the fast track because a real human is the strongest trust signal on a feed increasingly filled with AI content.

What is the easiest content style for an ecommerce founder to start with?

The easiest style is short unedited founder-journey clips shot on your phone, showing the actual daily work of your business. The scrunchie founder I follow started with one-minute phone clips before her workday, and that alone was enough to build a loyal audience that later bought from her at scale.

What is the fastest way to get comfortable on camera?

The fastest way is to identify the one specific hurdle blocking you (tech setup, being on-camera, going blank, editing, or scheduling) and remove that one hurdle in isolation. For Toni it was tech, so she rents a studio every three months and hands editing to somebody else. For me it was setup time, so I built a permanent rig I can start recording from in one button-press.

How does building a community translate into product sales?

A community translates into product sales because you have already earned trust with people who care about the topic your product sits inside, and you know exactly where they are. When you launch, you are selling to a warm audience that watched you build the business, not to cold Meta impressions. In several cases we have seen (Shalene in fitness and wellness, Derek Halpern with Truebani, and the ECF forum launch) the community existed first and the specific product was chosen partly because the community pushed the founder toward it.

Ready To Get Serious About Starting An Online Business?

If you are really considering starting your own online business, then you have to check out my free mini course on How To Create A Niche Online Store In 5 Easy Steps.

In this 6 day mini course, I reveal the steps that my wife and I took to earn 100 thousand dollars in the span of just a year. Best of all, it's absolutely free!

Podcast: Download (Duration: 46:09 — 53.1MB)

The way to launch a niche product on Kickstarter is to build a pre-launch email list, hire a crowdfunding marketing agency to scale Meta ads to 4-5x ROAS, and treat the campaign as three phases with a “simmering pot” middle you actively manage. On this episode of the My Wife Quit Her Job podcast, I sat down with Miroki Tong, co-founder and CEO of StoryEngineDeck.com, who has done exactly this three times.

Miroki and her co-founder Peter sell decks of creative writing prompt cards, a product most people cannot describe until they hold it. Their first Kickstarter asked for $50,000 and raised $250,000, and their second raised over $1 million and became the world’s most crowdfunded worldbuilding tool.

Today they still hit 4-5x return on ad spend on Meta at an under $10 cost per acquisition, and they refuse to use AI for anything creative.

Below is the playbook: how the first Kickstarter took off without a large audience, how they structure a 30-day campaign around the dead middle, how they hit 4-5x Meta ROAS on a niche product, and the mistakes that sink most creative-product launches.

Key takeaways

• You can hit a million-dollar Kickstarter without a huge audience if the product is strong, the lead-gen starts weeks before day one, and the landing page and founder video are excellent. Story Engine had a small comic-convention audience when their first campaign 5x’d its $50K goal.
• Every crowdfunding campaign has a “simmering pot” middle. The first 72 hours and the last 72 hours drive most of the pledges. You have to actively scale ads down during the middle to avoid bleeding money, then scale them back up at the end.
• 4-5x Meta ROAS on a niche product is possible with fresh creative and by mining customer feedback (live-stream questions, ad comments, review comments) to write the next ad’s headline. Story Engine refreshes ads about monthly.
• Plan stretch goals BEFORE the campaign goes live. Peter invented stretch goals mid-campaign on both of Story Engine’s biggest launches and spent the next five years fulfilling them.
• Do not undervalue the ask. Budget every real cost (materials, marketing, artists, writers, shipping, fulfillment, and the 30% tariff surcharge on top) and price the campaign to pay all of it. Creators who eat surprise costs to keep backers happy go bankrupt.
• Long-form YouTube video is where a hard-to-explain product actually clicks. A single 15-minute how-to on one of Story Engine’s expansions is one of their best-performing videos.

Can you succeed on Kickstarter without a large existing audience?

You can succeed on Kickstarter with only a small existing audience, but the product, landing page, founder video, and pre-launch lead-gen all have to be excellent to compensate. Peter had a modest comic-convention following when he launched the original Story Engine deck for $50,000. It closed at $250,000 by day three or six, catching him so off-guard that he did not know how to fulfill it.

Miroki is direct about the trap: having an audience is not sufficient on its own. She knows creators with much larger followings who have run crowdfunding campaigns that raised nowhere near what Story Engine has raised.

The bar is a great product, a killer landing page, a strong founder story, proper lead-gen, and money spent on ads. Skip any one of those and even a big list will not save the campaign.

What is the structure of a successful 30-day Kickstarter campaign?

A successful 30-day Kickstarter follows a three-phase shape: a huge lift in the first 72 hours, a “simmering pot” middle where pledges naturally slow, and a second huge lift in the final 72 hours. This shape is normal, not a sign of failure, and the campaign needs to be built and managed around it.

The first 72 hours are what your pre-launch email list, launch-day ads, and lead-gen momentum are for. If you hit day one without immediate backers, prospects second-guess the campaign from the start.

The middle two weeks are the danger zone. Ads that were working on day one will start to underperform, and if you set-and-forget them, you will bleed money for ten straight days. Miroki’s team actively scales ads down during the middle, uses the time for creative pivots and course-corrections, and then scales ads back up for the final push.

How do you generate pre-launch leads for a Kickstarter campaign?

You generate pre-launch leads by running Meta and Google ads for lead capture in the weeks before launch, emailing every previous crowdfunding backer, working with a crowdfunding marketing partner like BackerKit, and building anticipation across your existing newsletter and social channels. Momentum has to be built before day one, not on day one.

Story Engine runs pre-launch ads that capture emails, then converts those leads into day-one pledges when the campaign opens. Because they have run multiple successful campaigns, they also message backers from previous projects on the Kickstarter platform (“Hey, I have a cool new project, come check it out”), which is one of crowdfunding’s compounding advantages.

They partner with BackerKit for a lot of the ad management. BackerKit started as a crowdfunding platform and grew a marketing arm; they take a percentage and their costs scale with success. If you are not already an ads expert, Miroki recommends hiring them over trying to run pre-launch ads yourself.

How do you achieve 4-5x ROAS on Meta ads for a niche product?

Story Engine hits 4-5x Meta ROAS on Meta at under a $10 cost per acquisition by letting Meta’s algorithm find their audience (no manual interest targeting), refreshing creative about once a month, and mining customer feedback for the next ad’s angle. They do not narrow-target; they let Meta optimize and put the effort into making better creatives.

The creative-refresh source is where most sellers under-invest. Miroki’s team reads three things constantly: direct reviews on the website, comments left on the ads themselves, and repeated questions in their live streams. Any question that keeps coming up is a signal they are failing to communicate something, and that failure becomes the headline of the next ad.

They also build the full creative arsenal for each launch: long-form video, short-form video, static images, catalog ads, long-copy text, short-copy text, and multiple call-to-action variants. Different formats hook different people at different moments; a viewer might be entertained by a short-form video first and then read a detailed static ad two days later before pledging.

How often should you refresh Meta ad creative?

Story Engine refreshes their Meta ad creative about once a month. That cadence works for them because their product is a physical deck sold on repeat launches; a store selling faster-moving or trend-driven products should refresh more often (I refresh weekly).

The refresh does not require inventing all-new concepts every time. It usually means new headlines, new hooks, new opening frames, and swapping in the questions that surfaced in the last month of customer feedback.

What are the biggest mistakes creators make on Kickstarter?

The two biggest mistakes are inventing stretch goals mid-campaign and undervaluing the funding ask. Both mistakes make the campaign “successful” on paper and then quietly kill the business over the following years.

Stretch goals are extra rewards you promise backers as the campaign blows past its funding target. Peter invented Story Engine’s stretch goals on the fly during both of the biggest launches (“I will make new lesson plans, I will design a cool new edition book, I will create this new guidebook”), and the team spent the next five years fulfilling them. Plan every stretch-goal tier before the campaign launches so you know exactly what you are committing to.

Undervaluing the ask is the other career-ender. Creators price the campaign to cover materials only and forget marketing, artists, writers, shipping, and fulfillment.

When tariffs hit in 2025 and 2026, many creators chose to eat the surprise 30% cost rather than pass it to backers. Miroki is direct about this: eating surprise costs to keep backers happy is not sustainable, and creators who do it go bankrupt.

What should be in the Kickstarter budget nobody remembers?

The costs most Kickstarter creators forget to budget for are marketing spend, artist and writer fees, shipping and fulfillment logistics, the “success budget” for what happens if the campaign 5x’s its goal, and any tariff or duty surcharge that could hit before the boat lands. Every one of these is real and every one of these has killed campaigns.

Build two budgets side by side. The lean budget is the bare minimum you need to succeed and ship the promised product; the success budget is what happens if you blow past the goal and now have to manufacture, package, and ship 6,000 units instead of 100.

More scale means more logistics, more coordination, more per-unit oversight, and more ways to accidentally ship late.

How does Story Engine sell a hard-to-explain product with content?

Story Engine sells a hard-to-explain product by combining short-form video (where Peter uses the deck live on camera to build a story mashup in 60 seconds) with long-form YouTube (where a single 15-minute how-to walks through one of the expansions in depth). The short form pulls people in; the long form gets them to actually understand the product and buy.

For a hard-to-explain product, the demo IS the pitch. Peter can whip out the deck on camera and generate a “tortured cowboy in a post-apocalyptic city where underground worms cause earthquakes” story in 60 seconds, and that demo does more marketing than any voice-over pitch could.

Their once-a-year in-person demo at GenCon (100,000+ attendees, North America’s largest tabletop RPG convention) works the same way: the second they demo the deck, it clicks.

The long-form video is what unlocks the harder sells. Their 15-minute deep-dive on the Culture Keyhole expansion is one of their best-performing videos precisely because it takes 15 minutes to properly show what the product does. Miroki thinks they should make one for every expansion.

Should you refuse to use AI to make your creative product?

Refusing to use AI for a creative product is a defensible marketing position in 2026, and it has become part of Story Engine’s brand. Every prompt is human-written, every image is licensed from a real artist or photographer, every marketing newsletter is written by a person, and they say so openly.

The reason it works as a position: AI-generated content is flooding every feed, and buyers are getting better at spotting it. In a market of AI-generated writing tools and AI-generated art, “made by humans” is a real differentiator, and buyers who care about creativity are willing to pay for it.

Miroki draws the Canva parallel: when Canva first arrived it democratized graphic design and helped every small business, then everyone got bored of the same soft-beige-arch aesthetic and the best graphic designers rose back to the top. The same cycle will hit AI content, and the creators who kept their craft intact will be waiting.

Frequently asked questions

Do you need an existing audience to run a successful Kickstarter?

No, but you need enough audience for a strong first 72 hours or the campaign will stall. Story Engine’s first campaign 5x’d its $50K goal with only a small comic-convention following, because the product, the landing page, the founder video, and the pre-launch lead-gen were all strong. Creators with much larger followings routinely run campaigns that raise less, because the fundamentals were weak.

What ROAS should I expect on Meta ads for a Kickstarter campaign?

Story Engine consistently runs 4-5x ROAS on Meta for their crowdfunding campaigns and their evergreen store, at a cost per acquisition under $10 per deck. That is exceptionally good for a niche product; more typical is 2-3x ROAS. Their edge is fresh creative refreshed monthly, no manual audience targeting (Meta finds the buyers), and mining customer feedback for every new ad’s angle.

How long is a typical Kickstarter campaign, and how does the funding flow?

A typical Kickstarter campaign runs about 30 days. Most of the pledges come in the first 72 hours and the last 72 hours; the middle two weeks are naturally slower, which is normal. Plan to actively scale ads down during the middle to avoid burning budget when conversion rates drop, then scale them back up for the final push.

Should I plan stretch goals before my Kickstarter launches?

Yes, plan every stretch-goal tier before launch. Making them up mid-campaign is how creators end up owing backers five years of extra fulfillment on rewards they never costed out. Story Engine did this on both of their biggest launches and it consumed years of team capacity.

Is it worth hiring a crowdfunding marketing agency like BackerKit?

Yes, if you are not already a paid-media expert. Agencies like BackerKit take a percentage of what they raise, which means their cost scales with your success rather than being a fixed spend you might not recover. They know the crowdfunding-specific patterns (three-phase funding shape, when to scale ads down, how to lead-gen before launch) that most general-purpose ad agencies do not.

What kind of video content works best for a hard-to-explain product?

Both short-form video (60-90 second live demos of the product in use) and long-form YouTube video (10-15 minute how-tos that walk through the product in depth) work, and the two complement each other. Short-form pulls new viewers in; long-form is what gets them to actually understand the product enough to buy. Story Engine’s best-performing YouTube video is a 15-minute deep-dive on a single expansion.

Ready To Get Serious About Starting An Online Business?

If you are really considering starting your own online business, then you have to check out my free mini course on How To Create A Niche Online Store In 5 Easy Steps.

In this 6 day mini course, I reveal the steps that my wife and I took to earn 100 thousand dollars in the span of just a year. Best of all, it's absolutely free!

Podcast: Download (Duration: 41:00 — 47.2MB)

Claude Code is the terminal-based version of Anthropic’s Claude that runs on your computer, reads and writes your files, and executes scripts for you. On this episode of the My Wife Quit Her Job podcast, my co-host Toni and I break down exactly how we are using it to save around 30 hours a week between our ecommerce and consulting businesses. She uses it heavily for data analysis, Klaviyo audits, and email drafting; I use it for content generation, custom scripts, and fixing broken WordPress plugins.

The name “Claude Code” is misleading, because it has almost nothing to do with coding. It is essentially Claude with access to your computer, and the best use cases for ecommerce sellers involve zero code writing at all.

If you are already using regular Claude or ChatGPT in a browser, Claude Code is the next step and it is where the real time savings happen.

Below is the full breakdown: the specific workflows we automated first, how to install it if you have never opened a terminal, the mistakes we made, and the limitations we hit that are worth knowing before you spend hours going down a dead end.

Key takeaways

• Claude Code is Claude with permission to read, write, and modify files on your computer. It runs in your terminal. Despite the name, most of the highest-value ecommerce use cases involve no code writing at all.
• The highest-ROI workflows to automate first: Klaviyo flow audits, editorial calendar folder creation, customer-support email drafting, review-mining for short-form video scripts, and repurposing one piece of content into 10 platform-specific variants.
• The skill that matters most is articulating what you want in plain English. Traditional coding skills are becoming less valuable; specification and prompting are becoming more valuable.
• Skip Claude Cowork. Everything Cowork does, Claude Code does better and with fewer limits.
• You do NOT need to know how to code or how to open a terminal. Ask Claude in a browser to walk you through installing Claude Code step by step; it will.
• Always double-check what it writes. Especially when it touches Klaviyo segments (terminology mismatch), gives permissions (“yes, yes, yes”), or edits files outside the project directory.

What is Claude Code and how is it different from Claude and Claude Cowork?

Claude Code is a terminal application that gives Claude direct access to read, write, and execute files on your computer. Regular Claude runs in a browser and can only see what you paste in; Claude Code can open your Klaviyo export from the folder you saved it in, analyze it, and write the summary chart to disk without you copy-pasting anything. Claude Cowork is a middle option that sits between them, and you should skip it entirely because Claude Code is more versatile.

The three tiers, in order of power and setup effort:

1. Claude (browser): paste text in, read text out. Fine for one-off analysis but you export from every tool by hand and paste it into the chat.
2. Claude Cowork: a browser-based intermediate step. Not worth learning; anything it does, Claude Code does with fewer limits.
3. Claude Code (terminal): Claude with file-system access. Runs on your Mac or PC via the terminal. This is where the 30-hour-a-week time savings actually live.

The name causes confusion. The word “Code” makes people who do not write code assume this tool is not for them, and it is not a coding tool.

Toni is not a coder and gets the highest-value use out of it in our team. If the tool were renamed “Claude Desktop” it would probably have 10x the users.

Do you need to know how to code to use Claude Code?

You do not need to know how to code, and you do not need to have ever opened a terminal before. Toni had to Google “where is the terminal on a Mac” the first time. The trick is to use regular Claude (in the browser) to walk you through the Claude Code install, step by step, and to keep asking dumb questions until you understand each step.

The pattern that works for non-technical users:

1. Open regular Claude in your browser (or ChatGPT if you are managing Claude API credits).
2. Ask: “I have a Mac (or Windows PC). Walk me through installing Claude Code step by step. Show me what to click and what to type.”
3. When you get stuck, paste the error message back and say “explain this to me like I have never used a terminal.”
4. Keep asking until you understand. It never tells you you are dumb. In fact it tends to inflate your ego, which is a separate problem worth double-checking against.

The classic “help me, my Mac died and I do not know where the terminal is” moment happens to everyone the first time. You are not behind if you have to Google that.

What ecommerce workflows should I automate with Claude Code first?

The highest-ROI ecommerce workflows to automate first are Klaviyo flow audits, editorial-calendar folder auto-creation, customer support triage, review-to-short-form-video-script generation, and one-piece-of-content-into-ten-platform-variants repurposing. Each of these takes a task that used to eat multiple hours per week and reduces it to a single “run this” command.

Ranked by how quickly a normal ecommerce store will see time savings:

1. Klaviyo flow audits. Any store with more than 10 flows should be auditing them monthly. Nobody does because the export is a mess. Claude Code pulls flows, sorts by activity, color-codes stale ones, and hands you a “delete these, fix these, ignore these” summary chart.
2. Customer-support email triage. A morning script scans Gorgias tickets, marks obvious sponsor/spam solicitations for review, and drafts replies to the real support questions. Saves an hour of inbox digging.
3. Editorial-calendar folder creation. Add a row to a Google Sheets editorial calendar, and Claude Code auto-creates the matching Google Drive folder with a marketing brief, launch plan, and graphics sub-folders using your naming convention. Saves the entire job that used to be a person’s task.
4. Review-to-short-form-video-scripts. Export your product reviews from a plugin like Judge.me, feed them to Claude Code, and get back 50+ short-form video scripts that handle the most common objections. 15 minutes of your time, weeks of content queued.
5. Content repurposing. One YouTube script becomes a LinkedIn post, a Twitter thread, a Facebook post, a TikTok script, and a full SEO-optimized blog post that embeds the video. All in Steve’s voice because Claude Code was trained on his prior writing.

How do you use Claude Code to audit a Klaviyo account?

You use Claude Code to audit a Klaviyo account by giving it access to your Klaviyo export folder and a prompt that spells out how to evaluate each flow (age, active subscribers, open rate, revenue per recipient). It reads the export, sorts and aggregates the data, and returns a color-coded chart telling you which flows to delete, fix, or ignore.

The workflow that Toni uses for a client with 187 Klaviyo flows:

1. Export flows from Klaviyo, aggregated by month.
2. Save the export into a folder Claude Code has access to.
3. Prompt: “Analyze these flows. For each one, tell me the age, how many subscribers are actively in it, and flag any flow that has had zero recipients in 6+ months. Color-code the output: green (healthy), yellow (needs review), red (delete or archive).”
4. Claude Code reads the sheet, produces the summary chart, and calls out flows that need immediate attention.

This is the workflow that turned monthly audits from “impossible with 187 flows” into “runs in three minutes.” Trying the same thing with browser Claude requires exporting, uploading, waiting for it to process, and dealing with truncation on the larger accounts.

Can Claude Code check whether contractors are actually doing their work?

Yes, and it is one of the more underrated uses. If you have hired someone to manage your Klaviyo or Meta ads, Claude Code can dump the account’s actual state and check it against what the contractor said they were going to deliver. “You said you would create 10 flows this month. Here are the 4 you actually created.”

This is legitimately useful for stores that outsource to agencies or freelancers. You do not need to know Klaviyo or Meta yourself; Claude Code reads the account state and translates it into “here is what actually happened this month vs. what was promised.”

The natural extension is having Claude Code do the work directly once you have trained it on the past patterns. Hire a contractor to build the first 10 flows, feed those into Claude Code as reference, and have it generate the next 10 following the same conventions. Then check the work before it goes out.

What are the biggest mistakes to avoid when starting with Claude Code?

The three biggest mistakes when starting with Claude Code are giving blanket permissions that let it modify files outside your project directory, not double-checking its output on tools with weird terminology (like Klaviyo segments), and expecting it to be right the first time on complex workflows. All three are recoverable, but each has burned real users badly.

The permission trap: Claude Code asks “can I do X?” many times per session, and it is tempting to click “yes, yes, don’t ask me again” and move on. I did this on a whim once and Claude Code started modifying files outside the project directory.

Nothing terrible happened in my case, but the news has recent stories about users who accidentally gave permission and Claude deleted an inbox. Do not blanket-permission the whole disk; give per-project permissions instead.

The terminology gap: on tools like Klaviyo, Claude Code can tell you exactly what segments you need to build, but sometimes uses terminology that does not match the tool’s actual UI. It will describe a segment condition perfectly and then you go to Klaviyo and cannot find where to click.

This deficiency will probably close over time. For now, expect to do a small amount of manual translation on Klaviyo segment builders.

The first-try expectation: complex workflows almost never work on the first prompt. Expect three rounds of “this is 85% there, now fix these five specific things.” That last 15% is why you double-check every output.

How do you keep Claude Code from touching files it shouldn’t?

You keep Claude Code from touching files it should not by scoping permissions per project rather than blanket-approving with “yes, don’t ask me again,” and by starting each project in its own dedicated folder so the permission scope is that folder only. If you are cautious, run it on a virtual private server (VPS) instead of your main machine so any accident is contained to a disposable environment.

The VPS pattern is worth considering for anything experimental. You install Claude Code on a $10/month cloud server, give it whatever permissions you want, and if something goes wrong the blast radius is that server, not your laptop with your family photos and your business records on it. Steve is planning to move team-shared Claude Code work onto a VPS for exactly this reason.

For most solo users, scoped-per-project permissions on your normal machine are fine. Just resist the muscle-memory habit of hitting “yes” without reading what the permission is for.

What are Claude Code’s current limitations for ecommerce work?

Claude Code’s current limitations for ecommerce work include some rough edges on Klaviyo segment building (right logic, wrong terminology), the fact that it is tied to your specific machine (not shared across a team by default), and the general AI risk of confident-sounding wrong answers on any complex tool it has not been trained deeply on. None of these are dealbreakers, and most will close within a year.

The machine-tied problem: my Claude Code project on my desktop is not automatically available on my laptop. If you swap between computers, you either sync the project folder manually, use git, or set up on a VPS that both machines connect to.

The over-confidence problem: Claude Code will produce a beautifully formatted answer that is subtly wrong. This is the classic AI failure mode. The countermeasure is to always double-check load-bearing outputs, especially anything that will get sent to customers, spent as ad budget, or executed as a segment.

The Klaviyo segment issue specifically: Toni hit this last week analyzing a 300,000-subscriber list. Claude Code correctly identified the buckets to segment (unengaged, problem children, etc.), but the segment conditions it suggested did not map cleanly to Klaviyo’s actual UI terminology. Expected in the current version; expect improvement.

Frequently asked questions

What is Claude Code and how is it different from regular Claude?

Claude Code is a terminal application that gives Claude direct read, write, and execute access to files on your computer, while regular Claude runs in a browser and only sees what you paste in. The name is misleading because most of the highest-value ecommerce use cases involve no code writing at all. If you are already using regular Claude for data analysis, Claude Code eliminates the export-upload-wait-download cycle.

Do I need to know how to code to use Claude Code?

No. Toni is not technical and gets the highest-value use of Claude Code on our team, because the core skill is articulating what you want in plain English. If you get stuck on setup, ask regular Claude in your browser to walk you through the install step by step, and keep asking dumb questions until each step makes sense.

Which Claude Code workflows save the most time for an ecommerce store?

The five highest-ROI workflows to start with are Klaviyo flow audits (especially if you have 30+ flows), customer-support email triage, editorial-calendar folder auto-creation, generating short-form video scripts from your product reviews, and repurposing one piece of long-form content into 10 platform-specific variants. Between those five, most stores will save 20-30 hours per month in the first month.

Should I use Claude Cowork instead of Claude Code?

Skip Claude Cowork. Everything Cowork does, Claude Code does better and with fewer limits. The extra intermediate step is not worth learning, especially since installing Claude Code takes maybe 20 minutes if you use browser Claude to walk you through it.

What is the biggest risk of using Claude Code?

The biggest risk is giving blanket permissions (“yes, yes, don’t ask me again”) that let Claude Code modify files outside your project directory. There have been real news stories about accidents where users approved a permission chain and Claude deleted their emails. Scope permissions per project, do not blanket-approve, and consider running experimental work on a VPS so any accident is contained to a disposable machine.

How much time can Claude Code realistically save an ecommerce operator?

Toni and I both estimate around 30 hours a week saved between our businesses once the initial workflows are set up. The setup cost is real; each workflow typically takes an hour or two of prompting to get to a state where you trust it. The payoff compounds because every subsequent run is essentially free time.

Ready To Get Serious About Starting An Online Business?

If you are really considering starting your own online business, then you have to check out my free mini course on How To Create A Niche Online Store In 5 Easy Steps.

In this 6 day mini course, I reveal the steps that my wife and I took to earn 100 thousand dollars in the span of just a year. Best of all, it's absolutely free!

Podcast: Download (Duration: 45:11 — 52.0MB)

The state of ecommerce in 2026, based on Andrew Youderian’s annual survey of 300 seven, eight, and nine-figure store owners in his eCommerceFuel community, is that manufacturing your own products is winning, Amazon is fading as a growth channel, owning your own warehouse cuts your growth in half, and deep financial literacy is the single biggest predictor of higher profit margins. The 300 stores in the survey represent about $3.5 billion in aggregate GMV.

Andrew came back on the My Wife Quit Her Job podcast (for the seventh time) to walk through the biggest takeaways from his 61-page report. Some of the numbers surprised both of us.

Manufacturing adoption jumped nearly 50% over three years while every other business model dropped or stayed flat. Amazon’s share of revenue has cratered back to 2017 levels even though more sellers than ever have listings there.

And in the single stat that blew up on Twitter after Andrew released it, store owners who own their warehouse are growing 80-90% slower than those who lease or outsource fulfillment.

Below is the full breakdown: which business models are winning, why Amazon is a demand-capture channel rather than a growth channel, what the warehousing data really means, why 75% of sellers reported “no meaningful alpha” from AI in 2025, and the financial literacy threshold that separates 9% profit margins from 14% profit margins.

Key takeaways

• Manufacturing your own product jumped almost 50% over three years. Every other business model (dropshipping, private/white label, reselling, hybrid) either dropped meaningfully or stayed flat. Dropshipping fell from 9% to 4%.
• More sellers than ever are ON Amazon (about two-thirds of the 300 stores), but Amazon’s share of their revenue has fallen back to 2017 levels (~20%). Sellers use Amazon as a demand-capture channel now, not the point of their spear.
• Owning your own warehouse correlates with growing 80-90% slower than sellers who lease or outsource fulfillment (4% growth vs. 34% for leasers). Correlation not causation, but the pattern is stark.
• The sweet spot for inventory turns is 5-6 turns per year (turning your inventory every ~2 months). Too little inventory means missed orders; too much ties up working capital. Both extremes hurt.
• 75% of surveyed sellers embraced AI in 2025 but the survey found no meaningful profit or growth alpha versus the 25% who didn’t. Andrew expects that to shift in 2026 as tools mature and sellers get more disciplined about what to build.
• The single biggest jump in profit margin comes at the top of the financial-literacy scale. Going from a self-rated 4/5 to 5/5 lifted average net margin from 9.7% to 14.3%, a 50% relative profit bump.
• Store owners rate their own DTC website 90% enjoyment. Amazon is around 20%. TikTok Shop is the lowest at 11-15%.

What business models are winning in ecommerce in 2026?

Manufacturing your own product is the only business model that grew in the last three years, jumping nearly 50% in adoption among Andrew’s 300 surveyed sellers, while every other model dropped or stayed flat.

Dropshipping fell from 9% to 4%. Private/white label fell from 18% to 11%.

Hybrid (reselling plus your own products) fell from 20% to 14%. Straight reselling stayed roughly flat at 11-12%.

The drivers are the same forces reshaping the whole retail landscape. The end of the de minimis exemption in May 2025 destroyed the economics of most dropshipping and light-touch reselling. Amazon competition on generic products is now a race to the bottom, so slapping your brand on an Alibaba SKU (what Andrew calls private label) rarely differentiates you long enough to build a durable business.

Manufacturing in this context means proprietary product you designed and had made to your spec, not just a rebrand of an existing item. That is the harder path with the higher upfront R&D cost. It is also the path that leaves you with defensible IP when the copycats arrive.

What is the difference between private label and manufacturing?

Private label (what Andrew calls white label) is when you find an existing product on Alibaba, slap your brand on it, and resell it as-is. Manufacturing is when you design a proprietary product from the ground up and have it made to your specification. The difference is how much unique value you are adding, and it is the difference between a product a competitor can copy in a week and a product that took you a year to develop.

Terminology here varies by author. Some people (including me) use “private label” to mean the proprietary/manufacturing version, and “white label” for the Alibaba-rebrand version.

Andrew uses “private label” for the rebrand version and “manufacturing” for the proprietary version. The definitions matter less than the strategic distinction: are you adding real IP, or are you renting somebody else’s product?

Is Amazon still worth selling on in 2026?

Amazon is still worth selling on as a demand-capture channel but not as a growth engine. About two-thirds of the 300 surveyed sellers have Amazon listings, an all-time high, but Amazon’s share of their aggregate revenue has fallen from a peak of ~30% in the early 2020s back down to ~20%, which is roughly the 2017 level. Sellers are keeping Amazon on because branded searches (“Bumblebee Linens” or “your brand name”) still convert, but they are building growth off-Amazon.

The reasons are structural. Amazon advertising costs have risen more than 20% year-over-year for several consecutive years and Amazon fees keep climbing.

Foreign competition floods the marketplace (Marketplace Pulse reports new US Amazon sellers at all-time lows). And Amazon has started blocking AI crawlers from scraping listings, which means the growing share of shopping searches happening in ChatGPT, Claude, and Perplexity increasingly returns non-Amazon results.

The mental cost is also real for many owners. When Amazon suspends a listing or holds funds, someone on the team spends a week on support chats and it kills morale. Multiple sellers in Andrew’s survey (and me personally) have de-emphasized Amazon over the last few years partly for that reason.

Which sales channels do store owners actually enjoy?

Store owners rate their own DTC website highest at 90% enjoyment. Amazon lands around 20%, and TikTok Shop lands lowest at 11-15%.

The gap between DTC and everything else is the sharpest opinion in the entire survey.

The TikTok Shop enjoyment number surprises people who assume the platform is a gold rush right now. TikTok deliberately makes selling harder than it used to be because in the platform’s early days sellers flooded it with spammy products and failed fulfillment, which damaged the platform’s reputation. TikTok Shop enjoyment is climbing now that Fulfilled by TikTok (their FBA equivalent) is available, but the platform still ranks below Amazon on operator satisfaction.

Should you own your ecommerce warehouse or use a 3PL?

You should use a 3PL for most product categories, because Andrew’s survey found that sellers who own their warehouse are growing 80-90% slower than sellers who lease or outsource fulfillment (4% growth vs. 34% for leasers, 22% for outsourcers), even after controlling for revenue. This is correlation, not causation, but the pattern is stark enough that it blew up on Twitter when Andrew released the chart.

There are legitimate reasons this correlation exists that are not “warehousing kills growth”:

• Owning a warehouse means less time working on product and marketing (the things that actually drive growth).
• Owners with their own warehouse often already maxed out their niche’s opportunity, so growth was slowing anyway and warehousing became the natural capital deployment.
• Deep owned inventory can be a moat that trades growth for durability, which is a valid strategic choice.

The categories where owning your warehouse still makes sense: personalization (embroidery, engraving, custom work) that no 3PL will touch cost-effectively, extremely high-touch handling, and products where you genuinely need eyes on every unit. If your business is straightforward SKUs with clean fulfillment, a 3PL is the better default in 2026.

When does it make sense to buy your own warehouse instead of leasing?

Buying your warehouse makes sense when the rent on your leased space is climbing 30%+ per year and you need peace of mind on your lease horizon. That was my trigger. Rent on my Bumblebee Linens warehouse was going up 30% annually, and buying eliminated that recurring shock plus removed the several moves we had already gone through when previous leases ended.

The upside is stability. The downside is exactly what the survey shows: your capital and attention shift toward the physical operation and away from the levers that drive growth. Only make this call if the warehouse operation is genuinely part of your business moat (as personalization is for Bumblebee Linens), or if the lease math is truly punishing.

What is the ideal inventory turn rate for an ecommerce store?

The ideal inventory turn rate for an ecommerce store is 5-6 turns per year, meaning you cycle through your entire inventory roughly every two months. Andrew’s survey found this range correlated with the highest revenue growth AND the highest net income growth. Turn faster than that and you risk stockouts and missed orders; turn slower and your working capital sits trapped in unsold inventory.

Inventory turn = annual revenue divided by average inventory value. If you do $500K per year and hold $200K in inventory on average, you turn 2.5 times per year, which is on the slow side and probably means capital is trapped.

Watch out for the “amazing deal” trap when a supplier goes under and offers you 50% off their remaining stock. I did this recently and now sit on roughly 18 months of inventory in a cramped warehouse.

The math still works long-term, but the short-term pain (cash flow, physical space, labor to move it around) is real. Andrew’s survey data suggests the sweet spot is discipline, not opportunism.

Did AI adoption actually improve ecommerce financial performance in 2025?

AI adoption did not produce meaningfully better financial performance for ecommerce stores in 2025, according to Andrew’s survey. About three-quarters of respondents said they had meaningfully embraced AI, and their profit margins, growth rates, and net income growth were statistically indistinguishable from the quarter that had not embraced it. Andrew expects that to shift in 2026 as tools mature and operators get more disciplined about what they build.

The reason is not that AI is useless. It is that a lot of the 2025 building was undisciplined, and when you can build anything, you have to be careful not to build everything.

Andrew admits he spent time and money on internal AI tools that were fun to have but did not move the needle. When “no coding required” turns every idea into a buildable app, prioritization becomes the new bottleneck.

There ARE individual stores getting real alpha from AI. On-site AI search, AI cross-sells, AI-generated content for social channels, automated inventory planning, and mini ERPs built in Claude Code have all shipped in Andrew’s eCommerceFuel community.

But at the aggregate level, adoption alone did not win in 2025. Discipline about WHAT to automate is where the 2026 winners will separate.

What AI tools are ecommerce operators actually using well?

The ecommerce operators winning with AI in 2025-2026 are using Claude Code (or similar terminal AI) for one-off internal tools that would previously have required hiring a developer, and using ChatGPT/Claude in the browser for data analysis on Klaviyo, Meta ads, and inventory forecasting. Community members have built mini ERPs, inventory planning systems, and demand-management systems in two weeks each.

The pattern is: AI as leverage on the tasks a small team could not otherwise afford (custom apps, one-off analytics dashboards, personalized email drafting), not AI as a replacement for the core work of understanding customers and making product decisions. The stores getting outsized value have both technical intuition AND deep customer knowledge; AI multiplies both.

What is the single biggest predictor of ecommerce profit margin?

The single biggest predictor of ecommerce profit margin is the operator’s own financial literacy, and the jump happens at the top of the scale. Andrew asked respondents to self-rate their financial knowledge 1-5.

Going from a 3 to a 4 barely moved the needle (9.0% to 9.7% average net margin), but going from a 4 to a 5 lifted average net margin from 9.7% to 14.3%. That is a 50% relative profit bump for the last step.

The implication is that “reasonably good” financial understanding is not enough. Deep financial competence, the kind that includes understanding your financial statements cold, knowing how to take money out of the business, understanding debt and risk, and running unit economics on every SKU, is where the outsized returns live.

Andrew put together a free eight-part Financial Mastery series on his eCommerceFuel podcast covering exactly this territory. Anyone whose self-rating is a 3 or 4 should work through it (ecommercefuel.com/mastery).

What is contribution margin and why does it matter?

Contribution margin is the profit you make on each individual sale after subtracting the expenses that vary with that sale. It matters because most operators eyeball their profitability at the gross margin level (“we do 60% gross margins so we’re fine”), which massively overstates real profit once you include fulfillment, packaging, advertising, and labor. Understanding contribution margin per SKU changes how you discount, how you price, and which SKUs you push hardest.

There are typically two or three tiers you can calculate:

• CM1: revenue minus COGS, shipping, packing materials, and payment processing. The “what does one unit cost me to fulfill?” number.
• CM2: CM1 minus advertising and marketing costs attributable to that sale.
• CM3: CM2 minus a share of fixed costs (warehouse rent, salaries, software).

I calculate CM1 for every Bumblebee Linens SKU and include the labor cost of embroidery, which drops the “90% gross margin” on an embroidered handkerchief down to something much lower. The habit of running this math on every SKU changes what you promote, what you discount, and what you quietly retire.

Frequently asked questions

What are the biggest ecommerce trends according to the 2026 seller survey?

The five biggest trends: manufacturing your own product is up almost 50% while all other business models dropped, Amazon revenue share fell to 2017 levels even as more sellers list there, owning your warehouse correlates with 80-90% slower growth than outsourcing, financial literacy at the top of the scale drives a 50% profit-margin bump, and AI adoption in 2025 produced no aggregate financial alpha. All from Andrew Youderian’s survey of 300 seven-figure and above ecommerce sellers.

Is dropshipping dead in 2026?

Dropshipping is not dead but has shrunk significantly. In Andrew’s 300-seller survey, adoption fell from 9% three years ago to 4% today, driven mainly by the end of the de minimis exemption in May 2025 and the intensified Amazon competition on generic products. The dropshippers still succeeding are almost all in specific niches with barriers to entry (heavy freight, regulated products, or supplier relationships that are hard to replicate).

Should I put my ecommerce store on Amazon in 2026?

Yes, if you already have a DTC brand and want to capture branded search demand on Amazon; no, if you are treating Amazon as your primary growth channel. Amazon’s share of revenue for the surveyed sellers has fallen back to 2017 levels because rising fees, ad costs, and foreign competition have made it a demand-capture channel rather than a growth channel. Use Amazon so your existing customers can find you there, but build growth elsewhere.

Should I own my warehouse or outsource to a 3PL?

Outsource to a 3PL as your default in 2026, because Andrew’s survey found sellers who own their warehouse are growing 80-90% slower than those who lease or outsource, even after controlling for revenue. The exceptions where owning still makes sense: personalization (embroidery, engraving), extremely high-touch handling, high SKU counts with complex fulfillment, or when the warehouse operation is genuinely part of your business moat.

What inventory turn rate should I aim for?

Aim for 5-6 inventory turns per year, meaning you cycle through your entire stock roughly every two months. Andrew’s survey found this range correlated with both the highest revenue growth and the highest net income growth across 300 seven-figure stores. Slower and your capital sits trapped; faster and you risk stockouts.

Is AI worth investing in for a small ecommerce business?

AI is worth investing in specifically as leverage on tasks a small team could not otherwise afford (custom apps, one-off analytics dashboards, personalized email drafting, content repurposing), but Andrew’s survey found no aggregate financial alpha for AI adopters in 2025. The winners are disciplined about what they automate. Start with 1-2 high-ROI use cases and measure the outcome before building anything else.

How much can improving financial literacy boost my profit margin?

Improving financial literacy from a self-rated 4 out of 5 to 5 out of 5 lifted average net profit margin from 9.7% to 14.3% in Andrew’s survey of 300 seven-figure ecommerce sellers, a roughly 50% relative bump. Deep understanding of financial statements, unit economics per SKU (contribution margin), debt, and risk is the single biggest predictor of ecommerce profit margin in the data.

Ready To Get Serious About Starting An Online Business?

If you are really considering starting your own online business, then you have to check out my free mini course on How To Create A Niche Online Store In 5 Easy Steps.

In this 6 day mini course, I reveal the steps that my wife and I took to earn 100 thousand dollars in the span of just a year. Best of all, it's absolutely free!

Podcast: Download (Duration: 47:05 — 54.2MB)

The way to launch on TikTok Shop with a $5,000 budget is to send out roughly 500 samples in month one, layer in a small retainer program (roughly $200 for 15 videos per creator), skip the top 0.4% of high-GMV creators, and instead build a community of hungry mid-tier creators who convert into long-term ambassadors over 6 to 8 months. On this episode of the My Wife Quit Her Job podcast, I sat down with Sohan Sanka, head of growth marketing at Reacher, who has helped brands drive over half a billion dollars in TikTok Shop revenue.

The math is unforgiving if you approach TikTok Shop like Amazon. Of 3.3 million TikTok Shop creators, only 0.4% drive over $10K in GMV in a rolling 30-day window, and only 0.09% drive over $50K.

The obvious move is to chase that top 0.4%, but they are already booked with the Goalies and the Nellows, and poaching them costs more than nurturing your own community from scratch. The winning play is to find hungry mid-tier creators who will work twice as hard at a fifth of the cost.

Below is the full playbook: the product margins TikTok Shop actually requires, the affiliate outreach template that converts, the retainer math that gets you 375 pieces of content for $5,000, and why your first affiliates should be doing awareness content, not selling.

Key takeaways

• Only 0.4% of TikTok Shop creators drive $10K+ per month. Only 0.09% drive $50K+. The top creators are already booked with big brands, so your play is to find hungry mid-tier creators and build a loyal community over 6-8 months.
• TikTok Shop products need at least 4-5x gross margin to survive the stack: 6% TikTok platform fee, 20% average creator commission, 15-20% promo discount, GMV Max ad blend, and your landed cost. Sub-30% net margins on the platform are normal.
• The sweet-spot price range has widened from $30-$50 to include higher-AOV bundles ($200 ice cream makers, live-selling giveaways, and consumable subscriptions).
• Your $5,000 first-month budget breaks down to roughly 500 samples at $10 landed cost each, producing ~250 posts, of which ~15% (about 38) generate any ROI. A hybrid retainer approach ($200 for 15 videos per creator) can get you 375 pieces of content for the same money.
• Your first affiliates should NOT be selling your product. They should be creating top-of-funnel awareness content (the “conspiracies of dehydration” hook for an electrolyte brand), because your product does not have the reviews or sales credibility to convert yet.
• The outreach message that converts leads with the offer in the preview line (“Retainer opportunity…”), shows credibility second, and adds a personal signature last. Cold-start response rate is 1-3%, sample-request rate is ~50% of those, and post rate is 70% of that.

Is TikTok Shop worth selling on in 2026?

TikTok Shop is worth selling on in 2026 as a customer acquisition and awareness channel that amplifies your DTC and Amazon business, but not as a standalone profit engine. Amazon is still the preferred checkout destination, and DTC is where you build the CAC-to-LTV math that drives real profitability. TikTok Shop’s role is to test organic creative fast, build a creator community, and generate the halo effect that lifts brand searches everywhere else.

The halo effect is the underrated win. When a product goes viral on TikTok, brand searches on Google, Amazon branded searches, and DTC direct traffic all lift together. Sohan has watched brands take a break-even TikTok Shop launch and turn it into a 3-4x Amazon lift on the same product.

If you are already on Amazon and squeezed by rising ad costs, TikTok Shop is the most legitimate alternative acquisition channel in 2026. Just do the unit economics with your eyes open before you launch.

What kind of product works best on TikTok Shop?

The products that work best on TikTok Shop are physical products in the $30-$200 range with 4-5x gross margins, a genuine differentiator, and an angle that creators are already talking about organically. Consumables and repeat-purchase products do well because they compound in your community. Premium eco-friendly products with no unique story typically struggle because there is no hook creators can build content around.

The traditional sweet spot 18 months ago was $30-$50, driven by the free-shipping subsidy threshold at $30, and that has widened significantly. Live selling, bundles, and higher-AOV products like $200 ice cream makers are now working because creators can build entire demo streams around them.

A product where the demo IS the pitch (the ice cream maker turning cream into gelato in 90 seconds) is easier to sell than an abstract benefit.

The products that struggle: undifferentiated commodities where the story is “we are eco-friendly,” premium-only positioning with no viral hook, and anything that requires the creator to explain complex benefits in a 60-second video.

What margins do you need for a product to work on TikTok Shop?

You need at least 4-5x gross margin to make TikTok Shop economics work sustainably. The stack of fees and payouts leaves you with much less than the sticker price suggests. On a $50 product with 4x gross margin ($37.50 gross profit), a typical launch month looks like this after fees:

• TikTok platform fee: 6% of $50 = $3
• Creator commission (average 20%): $10
• Coupon/promo (15-20% off): $7.50-$10
• Landed cost (COGS + shipping): $12.50
• GMV Max ad blend (varies): $3-$6

That is roughly $36-$40 in costs on a $50 sale, leaving 20-30% net margin at best. To get there, you either need 4-5x gross margin AND disciplined creator commissions, or a strategic decision to run break-even on TikTok Shop and capture profit in the Amazon and DTC halo.

How do you find affiliates when you are a brand new TikTok Shop seller?

You find affiliates as a brand-new TikTok Shop seller by first getting over the $2,000 GMV threshold that unlocks creator outreach, then messaging mid-tier creators (not the top 0.4%) with a retainer offer that leads the message. Use a tool like Reacher, Cellico Launchpad, join.brands, or TikTok Shop Blasts to hit the initial GMV threshold, then start outreach.

The initial GMV bootstrap options:

• Employee purchases of your own listing to hit the $2K threshold
• Cellico Launchpad, join.brands, or similar seed services that get real buyers to your listing
• TikTok Shop Blasts

Once you can message creators, filter your outreach by relevance to your niche first, then by post rate (aim for 50-70%), then by GMV. Relevance matters more than raw creator size at your stage because a highly relevant mid-tier creator will out-convert a top-tier creator posting about your product as a favor.

What is the outreach message that actually gets responses from TikTok Shop creators?

The outreach message that gets responses leads with the offer in the preview line (before the message truncates in the creator’s inbox), then adds credibility, then closes with a personal signature. “Retainer opportunity: $XXX for 15 videos” opens better than “Hi, we’re a brand and we’d love to work with you.” Cold-start response rates are 1-3%.

The template Sohan recommends:

1. Hook (visible in inbox preview): “Retainer opportunity” or “We gave $70K in cash and prizes to our community last month, and our top affiliate made $10K.”
2. Offer: Retainer amount, commission structure, and what the creator gets (free product plus cash).
3. Credibility: “We just crossed $10M in sales” or “growing 50% year-over-year.” Only if true.
4. Product proof: Affiliates can see the sales count on your product card, so make sure your listing already looks like a dark horse before you message.
5. Personal signature: “Sohan, founder of BrandGummies.” Not a corporate footer.

The funnel math from 100 outreaches: 1-3 respond, ~50% of those request a sample, ~70% of those actually post, and 10-15% of posted videos generate any ROI. That is roughly one ROI-generating video per 100 outreaches when you cold-start.

Should you chase the top TikTok Shop creators or nurture mid-tier ones?

Nurture mid-tier creators over 6-8 months rather than chasing the top 0.4% of GMV creators, because the top creators are already saturated with big-brand offers and cost more to poach than a whole community of hungry mid-tier creators costs to build. The top 0.4% of TikTok Shop creators drive $10K+ per month; the top 0.09% drive $50K+.

Everyone tries to sign them, and you will not win that auction as a new brand.

The alternative math is more forgiving. There are thousands of neglected mid-tier creators in every niche who will work twice as hard for a fifth of the cost of a top-tier creator, IF you turn them into loyalists.

Reward awareness content, not just closed sales. Run monthly cash-prize challenges, and invite the top performers into a Discord where they get access to samples, briefs, and early product drops.

Over 6-8 months, that community becomes a self-sustaining engine.

The math of poaching a top creator: retainer fees in the thousands, competition with 10+ other brands offering the same, and once you land them they are one creator in your program. The math of nurturing 100 mid-tier creators: $5-10K spread across the community, and after 6-8 months you have an army posting 1,000+ videos a month.

Why should your first TikTok Shop affiliates make awareness content instead of selling?

Your first TikTok Shop affiliates should make top-of-funnel awareness content instead of direct-sell content because your listing is too new to carry the reviews, sales history, and credibility that a hard sell needs to convert. Awareness content warms the audience up over 2-3 videos so that when you eventually pitch, viewers already recognize the brand. Direct-selling on a cold listing is how you get 0.06% ROI on your first batch.

The pattern Sohan uses with new brands (electrolyte-powder example):

1. Video 1 (awareness): “The conspiracies of dehydration” – broad, entertaining, no product mention.
2. Video 2 (angle): “Dehydration for breastfeeding women” – narrower niche, hint at the problem your product solves.
3. Video 3 (pitch): “Here is what is in our electrolyte powder that replenishes those specific nutrients” – the sales pitch, now landing on a warmed audience.

This is the opposite of how most brands intuit the sequence. They assume “just sell hard on video one because we know our audience converts on Amazon.” TikTok is not a search platform. People are scrolling to be entertained; you have to earn the sale.

How do you budget a $5,000 TikTok Shop launch month?

A $5,000 first-month TikTok Shop budget breaks down two ways: pure seeding (500 samples at $10 landed cost = 250 posted videos = ~38 ROI-generating videos) or hybrid retainer (fewer samples plus retainer contracts at ~$200 per 15 videos = 375 total pieces of content). The hybrid retainer approach gets you more testing angles per dollar because retainer creators are contractually obligated to post multiple times.

The pure-seeding path:

• 500 samples × $10 landed = $5,000
• 50% post rate = 250 posted videos
• 10-15% ROI-generating = 25-38 videos that pay back
• Cost per ROI video = ~$200

The hybrid retainer path (Sohan’s recommendation):

• Some samples for reach (say $2,000)
• $3,000 in retainer contracts: 15 creators × $200 for 15 videos = 225 videos guaranteed posted
• Combined with seeded content, total pieces per month = ~375
• More angles tested = higher chance of landing a video that GMV Max can scale to 3.5x ROI

What is GMV Max and how does it work with your creator content?

GMV Max is TikTok’s automated ad system that pulls in every creator video with a TikTok Shop link (assuming the creator has ads authorization on) and tests around 500+ creatives per week automatically. You set an ROI target before launch; TikTok’s algorithm figures out which creatives to scale. Videos that generate at least $1 in sales get more budget; videos that flop get cut.

Realistic ROI targets:

• Launch phase (1.8-2.0 ROI): Essentially breaking even after in-platform fees and sampling, but scaling volume for credibility.
• Scale phase (3.0-3.5 ROI): Profitable, with commissions and GMV Max blended into unit economics.

The point where the model finally clicks: your community is big enough that you send one sample per creator, they post 15 videos on average, and 10% convert. At that scale, every $1 you spend on community returns $5-$7.

That is the target 6-8 months out. Before then, expect to run at break-even and rely on the halo effect.

How do you automate TikTok Shop affiliate management at scale?

You automate TikTok Shop affiliate management by using a platform like Reacher to handle outreach (10,000+ messages per day), personalized creative briefs (auto-generated per SKU when a sample is approved), reminder messages (auto-sent when a sample arrives), and segmentation (auto-inviting anyone who generates $100+ to your Discord). The marketer’s time then shifts from message-sending to analyzing which creatives and creators are actually working.

The automation stack Reacher provides:

• Outreach: Bulk messages with filters (niche, post rate, GMV) and AI-personalized hooks.
• Brief delivery: Personalized per SKU, auto-triggered when a sample is approved in Seller Center.
• Reminders: Automated post nudges when a sample arrives, including bonus incentives if the creator has not posted within the 13-day window.
• Community segmentation: Auto-invite top GMV creators to Discord, collect spark codes, get ads authorization, capture phone/email.
• Response handling: AI chatbot with a knowledge base handles ~80% of creator replies; a human handles the remaining 20%.

Without automation, this workflow requires several VAs for outreach alone. With automation, one marketer manages a 1,000+ creator community and spends their real time on the strategic analysis (which top 10% of hooks are working, which creators to graduate into the Discord, which creatives to push into paid).

Frequently asked questions

How much does it cost to launch on TikTok Shop with an affiliate strategy?

A realistic first-month TikTok Shop launch budget is $5,000, which covers roughly 500 samples at $10 landed cost each, or a hybrid mix of samples plus retainer contracts with mid-tier creators. Expect 250 posted videos from pure seeding or 375 pieces of content from the hybrid approach. Of those, 10-15% will generate ROI; the rest fund the community-building and awareness that pays back in months 3-8.

Why is only 0.4% of TikTok Shop creators driving most of the sales?

Only 0.4% of TikTok Shop creators drive $10K+ per month because the platform’s algorithm concentrates rewards on the creators who consistently produce content that converts. Of 3.3 million analyzed creators, 16% make over $1 in sales, 4.6% make over $100, 2.3% make over $500, and only 0.4% cross $10K. The top creators are already booked with major brands, which is why new brands should nurture mid-tier creators into loyalists rather than compete for the elite.

What margins do I need to make TikTok Shop profitable?

You need at least 4-5x gross margin to make TikTok Shop economics work sustainably, because the fee stack (6% platform fee, 20% average creator commission, 15-20% promo discount, plus GMV Max ad spend) leaves 20-30% net margin at best. Below 4x gross margin you will either lose money or need to treat TikTok Shop as a pure break-even acquisition channel that pays back through Amazon and DTC halo lift.

Should my first TikTok Shop affiliates try to sell my product directly?

No. Your first affiliates should create top-of-funnel awareness content that entertains and educates the audience about the problem your product solves, without pitching the product, because a new listing lacks the reviews and sales credibility a direct pitch needs to convert. The pattern that works is 2-3 videos per creator, moving from broad awareness to narrower niche to actual product pitch on video three.

What is a good response rate on cold TikTok Shop creator outreach?

A realistic cold-start response rate is 1-3% on TikTok Shop creator outreach. From 100 messages, 1-3 will respond, roughly 50% of those will request a sample, roughly 70% of sample recipients will actually post a video, and 10-15% of posted videos will generate ROI. That means about one ROI-producing video per 100 cold outreaches when you are just starting.

How long does it take to build a profitable TikTok Shop creator community?

Building a profitable TikTok Shop creator community typically takes 6 to 8 months of consistent investment. During that window, you are running at break-even or a controlled loss while you seed samples, run retainer contracts, and nurture mid-tier creators into loyalists. Once the community reaches the scale where one sample generates 15 videos on average with ~10% converting, every $1 spent on community returns $5-$7.

Ready To Get Serious About Starting An Online Business?

If you are really considering starting your own online business, then you have to check out my free mini course on How To Create A Niche Online Store In 5 Easy Steps.

In this 6 day mini course, I reveal the steps that my wife and I took to earn 100 thousand dollars in the span of just a year. Best of all, it's absolutely free!

Podcast: Download (Duration: 41:40 — 48.0MB)

For a brand-new ecommerce blog in 2026, put it on Shopify’s native blog, not WordPress. My co-host Toni and I spent the last decade recommending WordPress as the gold standard for ecommerce content, and we have both changed our minds for most new-store situations. The tracking, AI-search behavior, user-experience, and consolidated-platform math has all flipped in Shopify’s favor.

That does NOT mean migrating an existing 12-year-old WordPress blog with real authority. Redirects still hurt in AI search, and I would leave a proven WordPress blog exactly where it is. But for any brand-new content going forward, and for any store that has a small or stagnant WordPress blog, Shopify’s native blog is now the better call.

Below is the full breakdown: what changed since we first recommended WordPress, why Shopify’s tracking-and-attribution problem has become WordPress’s problem instead, when NOT to migrate an existing WordPress blog, and the state of blogging in 2026 (short version: dead as a standalone business, alive as an AI-mention engine for stores).

Key takeaways

• Standalone informational blogs with nothing to sell are dead as a business in 2026. Blogging only makes sense today if your site actually sells something.
• The purpose of an ecommerce blog in 2026 is to get mentioned in AI answers (ChatGPT, Claude, Perplexity, Google AI Overviews), not to rank blue links. Google now favors blogs attached to real transaction businesses.
• My Bumblebee Linens blog (attached to the store) is up ~40% in traffic since December. My standalone My Wife Quit Her Job blog is DOWN ~40% year-to-date. Both use the same tactics; the difference is one is attached to a store.
• Shopify’s native blog has closed most of the historical gap with WordPress. The remaining WordPress advantages (a mature plugin library, more layout control) rarely outweigh Shopify’s tracking, attribution, and AI-friendly consolidation.
• Do NOT migrate a 12-year-old WordPress blog with existing authority. Redirects hurt in AI search. Leave the legacy content where it is and write NEW content on Shopify.
• If your existing WordPress blog is small or stagnant (no traffic, weak domain), just move everything to Shopify and consolidate.
• Advertorial-style posts (long-form content that answers a real question and links to your products) still work extremely well. Whether they live on WordPress or Shopify matters less than whether they exist.

Is blogging still worth it for an ecommerce business in 2026?

Blogging is still worth it for an ecommerce business in 2026 IF the blog exists to get you mentioned in AI answers and to drive people into your store, not to build a standalone content business. The blogs attached to real transaction sites are up in traffic; the standalone informational blogs are down. My Bumblebee Linens blog is up ~40% since December while My Wife Quit Her Job (no store attached) is down ~40% over the same window.

Google is actively favoring content on sites that sell something. The signal is: “when someone searches this term, do we send them to a site that can actually solve the transaction, or a site that will just talk about it?” AI engines follow a similar logic when deciding what to cite. A blog post from a store selling handkerchiefs will get cited on “handkerchief gift ideas” more often than a blog post from a pure content site on the same topic.

The tactical implication: if you are running an ecommerce store, blog. Just do it inside the store. If you are running a pure content business with no products or services to sell, the business model is not viable anymore.

Why did we switch from WordPress to Shopify for ecommerce blogging?

We switched from recommending WordPress to recommending Shopify’s native blog because the four historical advantages of WordPress (superior editor, plugin library, better SEO, and same-domain hosting) have either closed or reversed. Meanwhile, four Shopify advantages (unified tracking, AI-search preference for consolidated sites, better ad attribution, and cleaner user experience) have grown into decisive edges.

The historical WordPress advantages, ranked by how much they still matter:

• Plugin library: still bigger than Shopify’s, but modern Shopify covers most of what a blog needs (SEO, table of contents, alt text). Advantage: WordPress, but narrower.
• SEO: WordPress used to outrank Shopify on identical content. That gap is gone. Both platforms rank fine now.
• Editor flexibility: WordPress used to blow Shopify away. Shopify has closed enough of the gap that it no longer matters for most posts.
• Same-domain hosting: the killer WordPress advantage of putting your blog at yoursite.com/blog instead of blog.yoursite.com. Shopify has fixed this in 2025-2026 with subfolder support.

The Shopify advantages that have grown decisive:

• Unified tracking: one Google Analytics account, one Meta pixel context, one attribution model. No cross-subdomain gymnastics.
• AI-search consolidation preference: AI engines appear to favor sites where everything (content and products) lives in one ecosystem.
• Ad attribution: when you send Meta or Google traffic to a blog post that lives on Shopify, the ad platform can track the full purchase funnel. On a WordPress subdomain, that funnel is broken.
• User experience: landing on a Shopify blog post feels like being in a store. Landing on a WordPress blog feels like leaving one. Add-to-cart on a Shopify blog is one click; on a WordPress subdomain, it is a jump.

Did the WordPress-WP Engine drama affect this decision?

The WordPress-WP Engine drama in late 2024 did not directly affect our recommendation, but it did leave a bad taste with a lot of developers and made “is WordPress really open source?” a legitimate question. Matt Mullenweg (CEO of Automattic, the company behind WordPress) blocked WP Engine from accessing the WordPress plugin repository over a personal beef and sued them in court. The whole thing looked like an abuse of the “open source” positioning that made WordPress popular in the first place.

The developer revolt led to alternative platforms getting traction, though none has broken through enough to name here. For most ecommerce store owners the drama is a footnote. But if you were on the fence about WordPress before, this made the decision easier.

Should you migrate an existing WordPress blog to Shopify?

You should NOT migrate an existing WordPress blog to Shopify if the blog has real authority, traffic, and years of accumulated backlinks. Redirects hurt in AI search (and in Google), and moving 100+ posts creates a redirect chain that will bleed authority for months or years. Leave the legacy content where it is and start writing NEW content on Shopify going forward.

The migration decision matrix:

• Existing WordPress blog with 5+ years of authority and real traffic: do NOT migrate. Keep the WordPress blog running for legacy content, write new content on Shopify.
• Existing WordPress blog with weak domain, no traffic, or under 1 year old: migrate to Shopify. The redirect pain is small; the consolidation win is large.
• No existing blog, starting from zero: Shopify’s native blog, no question.

Toni’s client fits the first category: 12+ year old WordPress blog, still drives revenue, hundreds of ranking posts. Migrating that risks losing everything that already works. Instead, all NEW posts go on Shopify, and the WordPress blog continues to serve its existing rankings.

Why do redirects hurt in AI search?

Redirects hurt in AI search because AI crawlers appear to deprioritize URLs that redirect elsewhere, and because the citation graph AI models build associates authority with specific URLs. When you redirect from oldsite.com/post to newsite.com/post, the new URL has to re-earn its citations from scratch, and the AI models that had ingested the old URL do not automatically update their associations.

Google’s traditional 301 redirect logic (which passes ~99% of link authority to the new URL) does not fully translate to how AI engines track content. This is still an evolving area, but the safe assumption is: every redirect costs you something in AI search visibility, and mass redirects during a migration cost you a lot.

What kind of blog content still works for ecommerce in 2026?

The blog content that still works for ecommerce in 2026 is problem-solving content (“beef stroganoff without mushrooms,” “how to get rid of ants”) and gift guides (“second-year wedding anniversary gifts,” “gifts for the groom”). Both types answer a specific question and then link the reader to products that solve it. Personal-essay-style blogging is dead; procedural problem-solving content is thriving.

Two Bumblebee Linens examples that are driving the ~40% traffic lift:

• Gift guides: “second year wedding anniversary gift” (cotton is the traditional gift). The post ranks, drives traffic, and every product mentioned is a Bumblebee Linens SKU with a direct add-to-cart link.
• Wedding season content: “gifts for the groom,” “handkerchief use for weddings.” Seasonal traffic peaks that convert well because the reader has commercial intent.

The pattern that makes these work: the reader has a specific problem, the post answers it directly, and the products that solve it are one click away. No personal narrative or brand storytelling gets in the way; just the answer and the products.

What content should ecommerce stores stop writing in 2026?

Ecommerce stores should stop writing personal-essay content, generic industry commentary, and “a day in the life” posts. None of that ranks, none of it gets cited by AI, and none of it drives store revenue. The kind of blog post that took a full Sunday morning at a coffee shop to write is not the format that wins in 2026.

The winning format is question-answer-question-answer, because AI engines chunk and cite content this way and human readers with commercial intent scan this way. If your posts require a reader to invest 5 minutes in your personal story before they get to the useful content, they will bounce.

How do advertorials fit into the ecommerce blog strategy?

Advertorials are long-form blog posts that answer a specific reader question in depth and then link naturally to the products that solve the underlying problem. They work extremely well in 2026 because they combine the problem-solving format that AI cites with the commercial intent that ranks. Whether the advertorial lives on WordPress or Shopify matters less than whether it exists at all.

The examples that have been working:

• Male hair-fullness product with an advertorial that discusses factors in hair thinning and closes with an add-to-cart. Simon at Seller Summit shared this pattern.
• “How to teach my kid to read” from a homeschool-authority brand, with 16 steps and product recommendations woven in.
• Any long-form guide from a founder with credentials (teacher, degree, publishing history) writing on their area of expertise.

The differentiator on Shopify is that the add-to-cart button can literally live inside the post. On WordPress, you need to link to the Shopify product page as a separate hop, which loses conversions. For advertorials specifically, Shopify’s native blog has a decisive UX edge.

What are the tracking and attribution problems with a WordPress blog on a subdomain?

The tracking and attribution problems with a WordPress blog on a subdomain are that Google Analytics, Meta ads, and most attribution tools treat blog.yoursite.com and yoursite.com as separate properties, breaking the full funnel view. When you run Meta ads driving traffic to a blog post that leads to a product page on the main domain, Meta cannot attribute the conversion cleanly. When you look at analytics, you have to switch between two accounts and manually reconcile the data.

The specific pain points:

• Google Analytics: two properties, two dashboards, two attribution windows. Cross-domain tracking is possible but fragile.
• Meta pixel: the pixel fires on the blog subdomain and on the store, but Meta’s optimization struggles to connect blog-post-view to product-purchase because they are technically different origins.
• Klaviyo / email attribution: abandoned-cart flows, browser-abandonment, and post-purchase flows are all optimized for a single-domain experience. Cross-subdomain breaks parts of this.
• Ad platform optimization: Meta’s algorithm needs clean signal to optimize toward purchase. Broken subdomain tracking gives it dirty signal, which hurts ROAS.

Frequently asked questions

Should a new ecommerce store use WordPress or Shopify for its blog in 2026?

Use Shopify’s native blog for any new ecommerce store in 2026, because WordPress’s historical advantages (better editor, bigger plugin library, subdomain SEO parity) have all either closed or reversed. Shopify’s unified tracking, AI-search consolidation preference, cleaner ad attribution, and better user experience now decisively outweigh what WordPress still does slightly better. The old default of “always WordPress for blogging” no longer applies.

Should I migrate my existing WordPress blog to Shopify?

Only migrate if your existing WordPress blog has weak authority, low traffic, or is under a year old. If your WordPress blog has 5+ years of authority, real ranking posts, and drives meaningful revenue, do NOT migrate, because redirects hurt in AI search and Google and mass redirects can bleed authority for months. Instead, keep the legacy WordPress blog for existing content and write all NEW content on Shopify.

Is blogging still worth doing for an ecommerce store in 2026?

Yes, if you use the blog to get mentioned in AI answers and drive people to your products rather than to build a standalone content business. Blogs attached to real transaction sites are gaining traffic in 2026 (Bumblebee Linens is up ~40% since December), while standalone informational blogs with nothing to sell are losing traffic (my personal blog is down ~40% year-to-date). Google and AI both favor content on sites that sell something.

What kind of blog posts should an ecommerce store write?

Write problem-solving posts (“beef stroganoff without mushrooms”) and gift guides (“second-year wedding anniversary gifts”). Both answer a specific question and link directly to products that solve the underlying problem. Skip personal essays, day-in-the-life posts, and general brand storytelling; the winning format is question-answer-question-answer, which is how AI engines chunk and cite content.

What is an advertorial and does it still work in 2026?

An advertorial is a long-form blog post that answers a specific reader question in depth and closes with links (or add-to-cart buttons) for the products that solve the underlying problem. Advertorials work extremely well in 2026, especially on Shopify where the add-to-cart can live inside the post. The format is winning because it combines the problem-solving content that AI cites with the commercial intent that ranks.

Why did Google’s Helpful Content Update hurt so many blogs?

Google’s Helpful Content Update decimated a lot of standalone informational blogs, including ours and most of the ecommerce-adjacent blogs in our peer network, because it started favoring content on sites with real business behind them (products, services, transactions). Standalone blogs got downranked in favor of content from stores, brands, and service businesses. This is one of the biggest reasons the “ecommerce store + blog on the same domain” model is now winning while pure content sites are losing.

Ready To Get Serious About Starting An Online Business?

If you are really considering starting your own online business, then you have to check out my free mini course on How To Create A Niche Online Store In 5 Easy Steps.

In this 6 day mini course, I reveal the steps that my wife and I took to earn 100 thousand dollars in the span of just a year. Best of all, it's absolutely free!

Podcast: Download (Duration: 42:13 — 48.6MB)

Agentic shopping is the emerging model where AI agents (ChatGPT, Claude, Perplexity, and platform-native agents like Google’s Universal Commerce Protocol and Shopify’s ChatGPT integration) do the product research, comparison, and cart-filling on behalf of consumers, with a human only clicking the final “buy” button. On this episode of the My Wife Quit Her Job podcast, my co-host Toni and I broke down what this actually means for ecommerce store owners, why the ad channels we all depend on (Meta, Google, TikTok) are about to change fundamentally, and what to invest in now to survive it.

I am in a full-blown midlife ecommerce crisis about this. My business currently runs on paid ads (Meta and Google), word of mouth, email, and SMS, and if a meaningful percentage of purchases in the next 2-3 years happen through AI agents that never see an ad, my paid channels stop working the way they do today.

Meanwhile, Amazon is scraping ecommerce sites and listing products without permission, Google Search is being replaced by AI Overviews, and social media is being flooded with AI-generated content that erodes the trust that made it a marketing channel in the first place.

Below is the full breakdown: what agentic shopping actually is, why Meta and Google ads will not disappear but will fundamentally change, why AI-mention ranking is now the single most important SEO investment, and the four practical things to do this quarter regardless of how the timing plays out.

Key takeaways

• Agentic shopping means AI agents research products, compare options, and fill carts on behalf of consumers. Humans typically just approve the final purchase. If this becomes mainstream, ad-driven customer acquisition breaks because agents do not click ads.
• Google is more at risk than Meta. Meta ads work by interruption in a feed; Google ads work at the moment of search intent. AI agents replace search behavior first.
• Amazon lost roughly 10 percentage points of “product discovery starts” over the last few years. People increasingly start on TikTok (for younger buyers) or in AI chat (for research-heavy purchases).
• Amazon’s Buy For Me program scrapes ecommerce sites and lists their products on Amazon without permission. Product data goes stale, orders arrive wrong, and the brand takes the customer-service hit.
• The four things to invest in now: AI-search ranking (product descriptions with quantified claims, statistics, and real specifics), brand loyalty (branded searches AI agents will honor), Reddit and PR mentions (where AI models cite), and content that gets indexed.
• Product descriptions do not need to be beautifully written for humans. They need to be data-rich for AI. Real statistics, specific specs, and named differentiators are what AI cites.
• Brand loyalty is the moat. When a shopper tells an AI agent “only show me Kate Spade dresses” or “only show me Marriott hotels,” the agent respects that. Building brand loyalty is the anti-agentic-shopping strategy.

What is agentic shopping and why does it matter for ecommerce?

Agentic shopping is when an AI agent (ChatGPT, Claude, Perplexity, or a purpose-built shopping agent) performs the entire product research and comparison process for a consumer, presents a curated set of options, and increasingly executes the transaction directly. The consumer only approves the final purchase, and sometimes not even that. This matters for ecommerce because AI agents do not see ads, do not fall for retargeting, and do not respond to social proof the way human shoppers do.

The mechanics are already live in early forms. Google announced UCP (Universal Commerce Protocol), which lets AI agents buy from participating stores without loading a browser page.

Shopify has an agentic storefront that integrates directly with ChatGPT, and Perplexity has been executing shopping tasks for months. The platforms are racing to build the plumbing.

The consumer behavior is not universal yet. Most people still shop with a human in the loop for high-stakes purchases (travel, luxury, engagement rings). But for low-stakes repeat purchases (household supplies, generic clothing, gifts) the trust barrier is much lower and AI agents are already getting used.

Will AI shopping agents kill Meta and Google ads?

AI shopping agents will not kill Meta and Google ads outright but will fundamentally change how they work, and Google is at higher risk than Meta. Google’s ad business depends on search-intent traffic that AI agents are already replacing. Meta’s ad business depends on interrupting scrolling behavior that humans will still do even if AI does their transactional shopping.

The distinction matters:

• Meta ads (interruption): Even in an agentic-shopping world, humans still scroll Instagram and TikTok to be entertained. Meta ads that create awareness during scrolling still land on human eyeballs. The click-to-purchase step may bypass Meta, but the awareness half of the funnel survives.
• Google ads (intent): Google’s core value proposition is capturing the exact moment someone searches “best noise-cancelling headphones under $200.” If the moment now happens inside ChatGPT instead of Google, that ad revenue moves too. Google is aggressively building AI Overviews and shopping-in-AI features to defend this.

The realistic timeline: Google search traffic starts dropping meaningfully in 2026-2027 as AI shopping matures. Meta’s core business is more durable because it captures attention, not intent. But both platforms will look different in three years, and the specific bidding, targeting, and creative strategies that work today will need to evolve.

What is Amazon’s Buy For Me program and why is it a problem for ecommerce sellers?

Amazon’s Buy For Me is a program where Amazon scrapes external ecommerce sites, lists their products on Amazon, and executes purchases on behalf of customers without the seller’s permission. When a customer buys the listing, Amazon places the order on the seller’s actual website (or against their scraped data), and any mismatch between Amazon’s listing and the seller’s real product becomes the seller’s customer-service problem.

The specific failure modes:

• Amazon’s scraped product data is stale, so buyers see outdated prices, colors, or specs.
• Orders get placed against variants that no longer exist, and customers receive the wrong item.
• The seller takes the blame and gets negative reviews on Trustpilot, Google Reviews, and their own site.
• Customer service workload spikes because the seller now handles complaints for problems Amazon created.

For solo operators or small teams, this is a real capacity problem. You cannot handle the increased customer service load caused by Amazon’s misinformation, and there is currently no official opt-out for Buy For Me.

The industry response has ranged from robots.txt blocking to legal review, but the situation is very much in flux.

How is customer product discovery changing in 2026?

Customer product discovery in 2026 has fragmented dramatically. Amazon used to be the default starting point for over 60% of product searches; that share has dropped by roughly 10 percentage points over the last few years.

AI chat interfaces have picked up a growing share for research-heavy purchases, and TikTok has become the discovery layer for younger consumers with the actual purchase often happening on Amazon or the brand’s DTC site.

Where different segments start their discovery today:

• Younger consumers (Gen Z, young Millennials): TikTok first, both for scrolling discovery and as an active search engine. Purchase often lands on Amazon.
• Research-heavy purchases (electronics, travel, high-consideration items): increasingly starting in ChatGPT or Claude to gather options and comparisons before any browser visit.
• Traditional brand-loyal purchases: still starting on Amazon or the brand’s site directly, though this share is shrinking.
• Impulse and gift purchases: Meta and Instagram feeds still drive significant impulse discovery even as the purchase channel diversifies.

The tactical implication for ecommerce sellers: you need to be visible in every one of these starting points, because you cannot predict which one a specific customer will use. Amazon-only strategies were already fragile in 2024. In 2026 they are actively risky.

What should ecommerce sellers invest in to prepare for agentic shopping?

The four highest-ROI things ecommerce sellers should invest in to prepare for agentic shopping are: AI-optimized product listings (data-rich descriptions with real statistics), brand loyalty (so shoppers tell AI agents “only show me my brand”), off-site AI-citation building (Reddit, PR mentions, review sites), and content marketing that gets indexed by AI crawlers. These four work regardless of exactly when agentic shopping becomes mainstream.

Ranked by leverage:

1. AI-optimized product listings. Rewrite every product description to include real statistics, specific specs, and named differentiators. AI models cite specifics, not vibes. This is a one-time investment that pays off across every AI touchpoint.
2. Brand loyalty. A shopper who tells their agent “only show me Kate Spade” or “only book Marriott hotels” bypasses generic AI ranking. Building brand affinity is the single strongest anti-agentic-shopping strategy.
3. Off-site AI-citation building. AI models are trained on and cite from Reddit, industry publications, Wikipedia, and specialty review sites. PR blitz, Reddit engagement, and getting mentioned in publications all matter more in 2026 than they did in 2023.
4. Content that gets indexed. A well-written blog post attached to your store gets cited when AI answers questions in your niche. This is the “get mentioned in AI answers” strategy we discussed in episode 643.

How do you write product descriptions that AI shopping agents will cite?

You write product descriptions that AI shopping agents will cite by leading with quantified claims, real statistics, and specific differentiators rather than beautiful prose. AI models pick up “91% of customers prefer this over the alternative” or “ships within 24 hours 98.3% of the time” but ignore “our handcrafted attention to detail speaks for itself.” The writing does not need to be lyrical; it needs to be data-rich.

The practical process I ran on Bumblebee Linens for over 1,000 products:

1. Have AI generate first-draft descriptions from your existing product data (specs, materials, dimensions, uses).
2. READ each generated description carefully. AI hallucinates specs constantly.
3. Add real statistics you actually track (shipping-on-time rate, return rate, customer preference from your own surveys).
4. Include specific differentiators as concrete claims, not vague brand language.
5. Publish and let AI crawlers index the new descriptions.

This took me weeks of reading. It felt like re-reading a book before publication because you cannot afford a single wrong stat in a description. But it is a one-time investment that keeps paying off as more AI systems cite content.

Why is brand loyalty the anti-agentic-shopping strategy?

Brand loyalty is the anti-agentic-shopping strategy because when a shopper tells an AI agent “only show me Yeti coolers” or “only book Marriott hotels” or “only Delta flights,” the agent honors that constraint and stops generic comparison-shopping. The brand becomes the filter the agent respects, which bypasses the whole “AI ranks generic products” problem.

The pattern is already visible in travel booking, where I have watched Toni use AI shopping heavily. She tells ChatGPT “I only want Marriott” and “show me Delta flights first” up front, and the AI does not question the loyalty; it just filters.

Every shopper who has strong brand preferences essentially opts their favorite brands out of generic AI ranking.

The implication for sellers: the brand-building investments that felt optional in 2015-2020 are strategic imperatives in 2026. Behind-the-scenes content, founder-journey video, factory tours, brand ambassadors, and community building all serve one purpose in an agentic-shopping world: making shoppers name YOU when they talk to their agent.

Will social media survive if AI generates most of the content?

Social media will survive but change significantly as AI-generated content floods the feeds and bot activity rises. Human users will spend less time on platforms where they cannot tell what is real, which erodes the attention Meta and TikTok sell to advertisers. The platforms are already responding by requiring AI-generation disclosures and giving verified-human accounts more distribution.

The specific concerns:

• Bot ratios: if bots on a platform outnumber humans by 10:1, the human attention Meta sells becomes proportionally more scarce and expensive.
• Trust erosion: when users cannot tell if an image is real, a review is real, or a person is real, they engage less deeply and click less often.
• Younger-generation defection: the older cohort’s social-media use is already declining. Whether Gen Z stays or moves to yet another platform is a live question.

The counter-argument: my kids’ entire social lives happen on social media, and they cannot imagine a world without it. For them, the platforms are utility infrastructure, not entertainment products they might opt out of.

So the demographic mix shifts, but the platforms themselves probably do not collapse.

What is happening to physical retail as ecommerce fragments?

Physical retail is unexpectedly rebounding in 2026, especially malls and category-defining stores. Younger consumers are treating malls as social spaces AND as try-before-you-buy showrooms, doing product research on their phones and then going to the mall specifically to try clothes on. The pandemic-era prediction that malls would die was wrong.

The behavior pattern I see in my own household: my daughter researches every clothing purchase on her phone (Instagram, TikTok, brand sites), goes to the mall to try items on in person, and then often buys online later for better prices or sizing options. The mall is not a purchase channel; it is a validation channel.

For DTC brands this means the pop-up-store and showroom-store strategy that has worked for Warby Parker, Allbirds, and others is going to matter more. Pure-digital brands that ignore physical presence entirely may be leaving conversion on the table for exactly the tactile validation the mall provides.

Frequently asked questions

What is agentic shopping?

Agentic shopping is when an AI agent (ChatGPT, Claude, Perplexity, or a platform-specific agent like Google’s Universal Commerce Protocol) does the product research, comparison shopping, and often the purchase itself on behalf of a consumer. The human typically only approves the final transaction, and increasingly the transaction happens without any human click at all. It matters for ecommerce because AI agents do not click ads, do not fall for retargeting, and do not respond to social proof the way humans do.

Will AI agents replace Meta and Google ads?

AI agents will not replace Meta and Google ads outright, but they will fundamentally change how both work, and Google is at higher risk than Meta. Google’s business depends on capturing search-intent traffic that AI agents are already replacing, while Meta’s business depends on interrupting scrolling behavior humans will still do even in an agentic-shopping world. The specific bidding, targeting, and creative strategies that work today will need to evolve in the next 2-3 years.

How do I get my products recommended by AI shopping agents?

To get products recommended by AI shopping agents, invest in four things: data-rich product descriptions with real statistics and specific differentiators, brand loyalty so shoppers name you when they talk to their agent, off-site AI-citation building through Reddit and PR mentions, and content marketing that gets indexed by AI crawlers. AI models cite specifics, not vibes, so rewrite product descriptions to include quantified claims like “91% of customers prefer this” or “ships within 24 hours 98% of the time.”

Is Amazon still the starting point for most product searches?

No. Amazon lost roughly 10 percentage points of “product discovery starts” over the last few years, dropping from its former dominant share as discovery fragmented across TikTok (younger consumers), AI chat (research-heavy purchases), and the brand’s own site (loyal customers). Amazon-only strategies were already fragile in 2024; in 2026 they are actively risky.

What is Amazon Buy For Me and can I opt out?

Amazon Buy For Me is a program where Amazon scrapes external ecommerce sites, lists those products on Amazon without the seller’s permission, and executes purchases on behalf of buyers, with no official opt-out mechanism currently available. Sellers have reported stale product data, mismatched orders, and negative reviews on their own sites caused by Amazon’s misinformation. Some sellers have started blocking Amazon’s crawler at robots.txt level, but that is a partial solution.

Should I be worried that social media will die because of AI content?

You should adjust rather than panic. Social media will keep functioning even as AI-generated content floods the feeds, but the platforms are changing as human attention becomes more scarce, verified-human content gets prioritized, and AI-disclosure requirements roll out. Continue investing in genuine human-created content (founder journeys, behind-the-scenes video, real customer stories); AI-generated content will get demoted as platforms fight the flood.

Ready To Get Serious About Starting An Online Business?

If you are really considering starting your own online business, then you have to check out my free mini course on How To Create A Niche Online Store In 5 Easy Steps.

In this 6 day mini course, I reveal the steps that my wife and I took to earn 100 thousand dollars in the span of just a year. Best of all, it's absolutely free!

Podcast: Download (Duration: 53:45 — 61.8MB)

Reddit is the single largest source of AI-search citations (around 20% on average, having dropped from a peak of 80% last September), which is why ecommerce brands who want ChatGPT, Claude, and Perplexity to recommend them need a real Reddit strategy. On this episode of the My Wife Quit Her Job podcast, I sat down with Danny Kirk, founder of ReddiReach, who has helped nearly 600 brands and startups build white-hat Reddit presence over the last decade.

The mechanics that make Reddit work for AI recommendations are simple to state and hard to execute. Reddit signed partnerships with Google, OpenAI, and other AI companies so their scraped data trains the models, and AI models cite Reddit heavily because it is the largest source of real human opinion on the internet.

If your brand gets mentioned by real Reddit users in relevant subreddits, your brand shows up in AI answers.

Below is the full playbook: the branded vs. non-branded Reddit strategies, how to build karma from scratch in 12 weeks, what will get you banned instantly, how to mine your customer list for existing Reddit users, and the four AI-search-optimization pillars (Reddit is one of them).

Key takeaways

• Reddit is the largest single source of AI citations, currently ~20% on average (peaked at ~80% in September 2024). Brands that get mentioned in relevant subreddits get recommended by ChatGPT, Claude, and Perplexity.
• Two strategies exist: branded (your official account acts as a helpful “fairy godmother” in relevant subreddits) and non-branded (mining your customer list for existing Reddit users who share their story authentically).
• Building karma legitimately takes 3-6 months of daily 15-minute effort. There is no shortcut that will not get you banned.
• Do 20 comments for every 1 post. Posts are inherently promotional; comments are helpful. Reddit’s algorithm and moderators heavily favor helpful commenters.
• The best Reddit case study Danny cites is user @YourBizBroker: 30,000 comment karma, no call-to-action in comments, no schedule-a-call links, just three-paragraph helpful answers on every “should I buy/sell/value a business” thread. He generates leads constantly.
• Never buy upvotes, never share tracking links, never post fake testimonials, and never automate. All four get you banned instantly and now cascade across every Reddit account you own.
• Reddit is one of four AI-search pillars along with content AI likes, on-site optimization, and broad brand mentions across the internet. All four matter.

Why does Reddit matter so much for AI search recommendations?

Reddit matters so much for AI search recommendations because Reddit signed data-licensing partnerships with Google, OpenAI, and other AI companies, and Reddit is the largest source of real human opinion on the internet. AI models are heavily trained on Reddit data and cite Reddit heavily in their answers. Studies from Visibility Labs and Jeff Oxford at 180marketing show Reddit currently accounts for around 20% of all AI-search citations on average (peaked at ~80% in September 2024).

The business model context makes the partnerships durable. Reddit was unprofitable for close to two decades before finally landing on data-licensing as its viable revenue model. AI companies are Reddit’s primary paying customers now, which means the AI citation flow will continue and probably grow.

For an ecommerce brand, this is the single largest lever available for AI-recommendation visibility. Getting mentioned in relevant subreddits by real Reddit users translates directly into AI answers naming your brand when someone asks the model “what should I buy for X problem.”

What is the difference between branded and non-branded Reddit strategies?

The branded Reddit strategy is running an official account under your brand name that acts as a helpful “fairy godmother” in relevant subreddits, answering questions and providing resources without selling. The non-branded strategy is mining your existing customer list for people who are already active Reddit users and asking them to share their authentic story on Reddit. Both work; the non-branded strategy carries less account risk and better AI-citation math.

• Branded strategy: Founder-led or brand-led account. Optimized profile, banner image, links to your website. Long-game play (6-12 months) to build karma and community trust. High reward when it works. Full risk concentrated on one account.
• Non-branded strategy: Real customers who are already active Reddit users authentically mention your brand in relevant threads. Distributed risk across many accounts. Highest AI-citation value because each mention comes from a different high-trust account.

Danny’s favorite branded-strategy example is Reddit user @YourBizBroker, a real business broker with ~30,000 comment karma. He answers every “should I buy/sell/value a business” question with a three-paragraph helpful comment, includes no call-to-action, and generates leads through people finding his profile after being impressed.

He has been doing this for years. This is what the long game looks like.

How do you build Reddit karma from scratch as an ecommerce brand?

You build Reddit karma from scratch by spending 15 minutes a day for 12 weeks doing 20 helpful comments for every 1 post, participating genuinely in subreddits you actually care about, and getting community upvotes rather than trying to shortcut the system. There is no karma-farming shortcut that will not get you banned in 2026. The white-hat path is the only durable path.

Danny’s 12-week starter plan (available as a free doc from ReddiReach):

1. Choose a good username. You cannot change it later. “founder-yourbrand” or similar works well. Cooler than a LinkedIn handle; not corporate.
2. Optimize the profile. Banner image, links to your website, bio section. This is the billboard when people click your name after seeing a helpful comment.
3. Join relevant subreddits. Where your customers actually hang out. Skim the rules of each; some have karma minimums to post.
4. Comment first, do not post. Answer real questions in your area of expertise for the first month. Aim for 5 great comments a day.
5. Farm karma in genuine ways. Ask helpful questions in AskToronto, AskNewYork, or your city’s sub. Answer questions in hobby subs you actually care about.
6. Wait until 1,000 karma before mentioning your brand. That is roughly the threshold where you have social proof and immunity from casual bans.

What tools help you find the right subreddits and posts to comment on?

The tools that speed up Reddit strategy without breaking the “always be a human” rule are Subreddit Signals (finds relevant subreddits for your niche and surfaces daily posts worth commenting on) and Mochi Social (identifies subreddits, reads their rules, and drafts posts that respect those rules). Both save hours a day of manual searching while keeping you in the human-only zone Reddit actively enforces.

Reddit is now aggressively banning automation tools. LinkedIn is doing the same, having recently banned multiple automation tools and even kicked the founders of those tools off the platform. The tools worth using are ones that help you find WHERE to be helpful, not ones that automate the helpfulness itself.

How many Reddit comments should you post per day as an ecommerce brand?

Post about 5 great comments per day and 1 post per week as an ecommerce brand on Reddit, at a 20:1 comment-to-post ratio. Comments are inherently helpful; posts are inherently promotional. Reddit’s algorithm and moderators both favor high-comment-ratio accounts because those accounts look like real community members rather than promotional bots.

If you want to move faster, you can do more comments, but the 20:1 ratio is the important part. An account with 500 posts and 50 comments looks promotional; an account with 50 posts and 1,000 comments looks like a real community member.

What will get your Reddit account banned in 2026?

The Reddit tactics that get you banned in 2026 are buying upvotes or downvotes, sharing tracking links, posting fake testimonials, automating any part of the account, posting the same message across many subreddits, and being promotional in subreddits that explicitly prohibit it. Reddit’s enforcement has gotten dramatically stricter over the last two years and now cascades across every account you own.

• Buying upvotes/downvotes. Karma is Reddit’s ranking currency. Manipulating it is the fastest ban path.
• Sharing tracking links. UTM parameters, affiliate links, shortened URLs. All get flagged. Danny recommends never sharing links at all.
• Fake testimonials or false use claims. “I’ve been using X for years!” from an account that just registered gets caught and banned.
• Automation of any kind. Reddit’s business model depends on being the last place with real humans. They actively hunt automation.
• Cross-account IP flags. Log the same brand account into different devices from different cities and Reddit’s fraud detection kicks in.
• Shadow bans. Unlike most platforms, Reddit does not tell you you have been banned. Your posts just stop showing up to others. Check in incognito mode to detect.

What happens if one of your Reddit accounts gets banned?

When one Reddit account gets banned, every other Reddit account tied to the same devices, IP addresses, or verification methods typically gets banned too. This cascading-ban behavior is why single-account strategies (branded) are riskier than distributed strategies (non-branded, using real customer accounts).

This is also why sharing your brand account’s password with a social media manager is dangerous. Logging in from a new city or new device flags the account. Best practice is to keep the account on one or two consistent devices and never share credentials.

How do you mine your customer list for real Reddit users?

You mine your customer list for real Reddit users by adding a “Are you active on Reddit? Would you share your experience with our product there?” question to your post-purchase email flow, distributed over the course of a year rather than blasted to your entire list at once. Only ask customers who are already active Redditors with karma and real posting history; forcing occasional users to post about your product looks fake and gets them banned.

The math is honest: probably 1-2% of your customer list will be active enough Redditors to be useful. That is fine. For a store with 20,000 customers, that is still 200-400 potential authentic mentions.

The critical rule is to not ask them all at once. Twenty simultaneous positive Reddit posts about your brand looks exactly like brigading, gets flagged, and gets everyone banned. Trickle the ask over 12 months.

How do you measure whether Reddit marketing is actually driving results?

You measure Reddit marketing results with a post-sales survey asking “How did you hear about us?” with Reddit as a specific option, plus AI-visibility tools like AI Peekaboo, Peak AI, or Profound that track how often your brand gets mentioned in AI answers. Direct sales attribution from Reddit is easier than AI-attribution because AI users often research on ChatGPT and then Google your brand, which shows up as “direct traffic” in analytics.

The AI-attribution tools are all imperfect in 2026 because the space is new. Danny’s take: “the data is as bad as it is ever going to be today, and it gets better from here.” Track something now, even if imperfect, so you have a baseline as the tools mature.

For a brand that started Reddit work six months ago, Danny has seen results ranging from moderate lift to 900% increases in AI-search traffic. The variance depends heavily on the niche depth (more passionate niches convert better) and the total addressable market of the relevant subreddits.

What are the four pillars of AI search optimization for ecommerce?

The four pillars of AI search optimization are Reddit presence, on-site content that AI likes to read (structured, question-and-answer format), technical on-site optimization (schema, clean HTML, fast load), and broad brand mentions across the wider internet (PR, industry publications, review sites, Wikipedia). Reddit is the largest single pillar for most brands, but all four matter and none can be skipped.

Ranked by weight for a typical ecommerce brand:

1. Reddit (approximately 20% of citations): highest single lever.
2. Broad brand mentions: highest correlation with “AI recommended brand” in cross-brand studies. Press coverage, podcast appearances, industry publication mentions.
3. On-site content AI likes: BLUF-style opening paragraphs, keyword-rich H2s, question-based headings, structured FAQ sections with schema.
4. On-site technical optimization: proper schema (Product, Article, FAQPage), fast load speeds, clean semantic HTML.

Should you create your own subreddit for your brand?

Creating your own subreddit is possible and gives you full control (you set the rules, you can be promotional), but it is a long-game bet and probably underweighted for AI-recommendation purposes because AI values human signal (comments, upvotes, engagement) more than promotional posts. A subreddit with 100 members that you fully control is worth less to AI than 10 real mentions in a subreddit with 100,000 members.

The exception is if you already have a large customer base you can migrate. A brand with a 250,000-member Facebook group can plausibly convert 5-10% of those to a branded subreddit, giving you a real community from day one. Without that seed audience, building a subreddit from scratch is much harder than getting mentioned in existing ones.

Frequently asked questions

Does Reddit marketing actually help get my brand recommended by AI?

Yes. Reddit is currently the single largest source of citations for AI-search results (around 20% on average, peaked at 80% in September 2024). Reddit has data-licensing deals with Google, OpenAI, and other major AI companies, so brand mentions in relevant subreddits translate directly into AI answers naming your brand. Danny has seen client brands lift AI-search traffic by up to 900% using this approach.

How long does it take to build a Reddit presence that actually works?

Building a Reddit presence that gets meaningful AI-recommendation lift typically takes 3-6 months of daily 15-minute effort. The first month is pure community participation with no brand mentions to build karma. Months 2-3 introduce your first branded content once you cross ~1,000 karma. Months 4-6 are where the compounding starts and AI recommendation impact shows up.

What will get my Reddit account banned instantly?

The fastest ways to get banned are buying upvotes or downvotes, sharing tracking links, posting fake testimonials or false use claims, automating account activity, cross-posting the same message across many subreddits, and being promotional in subreddits that explicitly ban it. Reddit’s enforcement now cascades across every account tied to your devices or IP addresses, so one ban usually means all your accounts.

Should I run a branded Reddit account or use my customers instead?

Do both if possible, but if you only have bandwidth for one, mine your customer list for real Reddit users. Non-branded customer mentions carry less account risk (distributed across many accounts) and higher AI-citation value (each mention comes from a different high-trust account). The branded account is a longer game with concentrated risk, so it should be the second pillar rather than the only one.

What is the ratio of comments to posts on Reddit for a brand?

The right ratio is 20 comments for every 1 post. Comments are helpful by nature; posts are promotional by nature. Reddit’s algorithm and moderators both favor accounts with high comment-to-post ratios because those accounts look like real community members. An account with 500 posts and 50 comments looks like a promotional bot; an account with 50 posts and 1,000 comments looks like a genuine contributor.

How much does Reddit marketing cost to hire out?

Reddit marketing agencies like ReddiReach typically charge in the low four figures per month for full-service branded-plus-non-branded strategy, including subreddit identification, content creation, community engagement, and reporting. The specific price varies by scope (how many subreddits, how many posts per week, whether AI-search-optimization is bundled). For solo operators, the DIY approach costs only your time (roughly 15 minutes per day).

Ready To Get Serious About Starting An Online Business?

If you are really considering starting your own online business, then you have to check out my free mini course on How To Create A Niche Online Store In 5 Easy Steps.

In this 6 day mini course, I reveal the steps that my wife and I took to earn 100 thousand dollars in the span of just a year. Best of all, it's absolutely free!

Podcast: Download (Duration: 40:41 — 46.9MB)

What is actually working in ecommerce right now, in one line: unpolished UGC beats produced ads, a real brand story ties every channel together, differentiated products with a moat outperform generic Amazon listings, and FAQ schema on collection pages is quietly winning the AI-search game. That is the through-line from the Shopify HQ Merchant Mastery event I just spoke at with my co-host Toni for the My Wife Quit Her Job podcast, recorded live from Soho after a day of sessions and unfiltered founder conversations.

Almost none of the merchants in the room were relying on Amazon as their main channel. They were Shopify-first brand builders, and the tactics they said were actually moving revenue right now were consistent across the room.

Below is the full breakdown: the UGC pattern that flipped for me, the brand-story exercise Scott Cunningham runs, the AI-search FAQ approach I use on my own store, the shipping-cost trap most sellers fall into, and my updated criteria for choosing which ecommerce events are worth your time and money.

Key takeaways

• Unpolished, phone-shot UGC is outperforming produced ads. The best-performing ads at the event were the ones sellers almost did not run because they thought the quality was too low.
• A real brand story is now table stakes. The same story feeds ad copy, email, sales pages, and social so every message reinforces the others.
• Generic products are dying. Buyers under 35 want differentiated, story-driven products they can hunt down; tick-tockable products win.
• Two AI-search strategies work right now: one FAQ per blog post, or every FAQ stacked on the category/collection page with FAQ schema markup. I use the collection-page version and it is working.
• Always price ecommerce shipping EXW and let your own freight forwarder handle it. FOB and DDP quotes from your supplier often pad the shipping cost and hide your real cost of goods.
• Renegotiate FedEx and UPS every 3 months, not once a year. Rates keep climbing and carriers will move if pushed.
• The best events are small, the attendees are open about numbers, and there is a live community around the event before you ever show up in person.

What is actually working in ecommerce right now

The four moves working across the Shopify-first brands at Merchant Mastery were unpolished UGC, a codified brand story, differentiated products with a real moat, and a Q&A content layer built for AI search. Every seller I talked to was leaning on some combination of those four, not on generic Amazon-style listings or hyper-produced brand videos.

The common thread is authenticity that scales. Buyers can now spot AI-produced polish in a second, so raw, real, and specific is what converts.

Everything below is a deeper look at each of those four, plus the shipping and event tactics that came up in the same conversations.

Why unpolished UGC is outperforming produced ads

Unpolished, phone-shot UGC is beating produced ads right now because AI has flooded feeds with over-polished content and authentic content stands out by contrast. About half the attendees at the event brought their products and spent the afternoon filming rough UGC for each other with lapel mics and phones, no lighting rigs and no editing suites.

The clearest example came from Mary, one of the attendees. Her group on the pre-event Zoom calls kept pushing her to run one specific ad she hated and refused to publish. She finally caved, ran it, and it became her single best-performing ad.

The pattern repeats across the room. The video a seller almost did not ship is often the one that hits.

Toni and I ran into the same wall when we launched our Profitable Audience podcast. I re-shot the first three videos at least five times because they were filmed in a hotel and I thought they looked unacceptable. We never used any of them, and the ones that eventually performed were the least produced.

If you are sitting on content you think is not “good enough,” publish it. The less produced it looks, the more likely it is to work.

How to build a brand story that ties every channel together

A brand story is a written document that captures your value propositions and origin, then feeds every piece of marketing copy you produce so the message stays consistent across ads, email, sales pages, and social. Scott Cunningham opened the Merchant Mastery day by walking the room through a GPT he built that extracts the story out of you with a structured questionnaire.

I teach the same exercise to students in my class. The first time I ask a founder what is special about their product, the answer is almost always “I do not know.” The exercise is really about extracting information the founder already has but has never written down.

Here is how I use it inside a real ecommerce business.

Before we run any launch or promotion with one of my clients, we write a marketing brief for the product. Inside the brief is the specific product story. Then every touchpoint for that campaign, ad copy, email copy, social captions, sales-page copy, and follow-up emails, pulls from the same brief so the message compounds instead of contradicting itself.

Once you have the raw brand story down, you can hand it to Claude or ChatGPT and generate the campaign assets in minutes instead of days. The hard part is the initial brain-dump. Everything after that is automation.

Why generic products are dying and what to sell instead

Generic products are losing because a younger buyer would rather hunt down a differentiated product with a story than one-click a generic version on Amazon. Almost every product at Merchant Mastery was either genuinely unique, personalizable, or tick-tockable, and it is the reason those brands are outperforming the alphabet-soup Amazon sellers.

I saw the buyer shift firsthand in the comments on a recent YouTube video I posted that hit 1.3 million views in three days. The dominant sentiment across thousands of comments: shoppers are tired of the identical products from Chinese factories that fill Amazon search results under a rotating cast of random-letter brand names.

That signal changes what “product research” should look like today. Before you commit to a SKU, ask two extra questions.

Is the product tick-tockable? Meaning: can a real customer hold it up on camera and instantly show why it is different or fun? A shot ski, a magnetic weekly pill holder, an old-fashioned cocktail tea bag, and a story-building card deck were all in the room; every one of them films well without a script.

Does it have a moat? At Sellers Summit this year, Chuck from Quiet Light gave a quiz on which business had the highest sale multiple, and the correct answer was always the one with the strongest moat when everything else was factored in.

Personalization, a proprietary design, a genuine brand, or a physical process no one else runs are all moats. A generic reseller listing is not.

How to rank in AI search: FAQ schema on collection pages vs one FAQ per blog post

Two strategies are working for ranking in AI search right now: one FAQ per dedicated blog post, and every FAQ stacked on the collection or category page with FAQ schema markup. The Merchant Mastery session on AI search used the blog-post-per-question approach. I use the collection-page approach on Bumblebee Linens.

Here is the exact workflow I use on a collection page like Bumblebee’s baptism handkerchiefs page.

Step 1: pull the questions. I ask Claude or ChatGPT to give me every question people ask about the parent topic (in this case, baptism gifts, because “baptism handkerchiefs” alone has little search volume).

Step 2: build the FAQ block. I take those questions, write short answers, and add them to the collection page with FAQ schema markup so search engines and AI models can extract the Q&A cleanly.

Step 3: put the FAQ near the bottom. Products stay above the fold so you do not push your buy buttons under a wall of text. The FAQ sits below and does its job for search.

The trade-off between the two approaches is real. The blog-post-per-question route is more classic SEO and gives each question a dedicated URL, which can rank for more query variations. The collection-page route sends AI-search clicks straight to a page full of products instead of a blog post where you have to bounce the reader over to your store.

If you have the resources, do both. If you have to pick one, I lean toward the collection-page version because the click already lands on a page that can convert.

How to price ecommerce shipping: EXW vs FOB vs DDP

The safest way to price ecommerce shipping from an overseas supplier is EXW, which means the supplier hands the goods off at the factory and your own freight forwarder handles everything from there. FOB (supplier ships to the port, you cover it from the port) and DDP (supplier handles the whole delivery end to end) are both convenient, but suppliers routinely pad the shipping line to make margin on the freight and hide the real cost of goods.

Barack, who runs the freight forwarder Forceget, walked the room through this at Merchant Mastery, and it lined up with what I have seen in my own supply chain.

Two extra shipping moves that came out of the same session.

Always keep two active suppliers on any core SKU. Once a supplier knows switching costs are high, they will raise prices every year, and unless you have a real alternative you cannot push back. A second supplier is your leverage.

Renegotiate FedEx and UPS every 3 months, not once a year. I have been doing it once a year, and Barack does it quarterly. Carriers have been raising prices aggressively (a recent USPS fuel surcharge plus roughly 6% base increases from UPS and FedEx), and quarterly renegotiation is now the baseline if you ship real volume.

My criteria for which ecommerce events are worth attending

An ecommerce event is worth your time and money if it is small enough to actually meet everyone, the attendees are open about their real numbers, and it is built around a live community you can plug into before and after the event, not just a firehose of sessions. Merchant Mastery hit all three, which is why the conversations were as valuable as anything on stage.

Here are the three filters I use before I book a flight now.

Small and intimate. If the event is too big, you never get past small talk. At Merchant Mastery I sat down at a table and attendees started volunteering their revenue and profit numbers without being asked, because the group was small enough that they already trusted the room.

Open attendees. This is the hardest one to screen for from the outside. Ask around: do people at this event actually share numbers, or does everyone play their cards close to the chest?

The former is worth the ticket. The latter is not.

A real community around the event. Scott’s attendees had been on twice-weekly Zoom calls for months before flying in, so they walked in as friends. The Sellers Summit event I run has a Discord community that runs year round for the same reason.

If an event has no community layer, join whatever community exists (a Facebook group, Discord, Circle space) and be active in it for at least a month before you show up. Skipping that step wastes most of your ticket cost.

The pendulum is swinging back toward real human connection precisely because AI has flooded every channel with content, so events and communities are getting more valuable, not less.

Frequently asked questions

What is actually working in ecommerce in 2026?

Right now the winning combination is unpolished UGC ads, a written brand story that feeds every channel, differentiated products with a real moat, and an FAQ layer on your key pages built for AI search. Generic Amazon-style listings, over-produced brand videos, and one-off blog posts with no distribution are all underperforming.

Why is unpolished UGC outperforming produced ads?

Because AI has flooded feeds with polished content, so raw phone-shot UGC now stands out and reads as authentic. The clearest sign it is working: sellers keep reporting that their best-performing ads are the ones they almost refused to publish because they thought the video quality was too low.

How do I create a brand story for my ecommerce business?

Start with a structured questionnaire (or a GPT built for the exercise) that pulls out your origin, your value propositions, and the specific reasons your product exists. Then write a short marketing brief for each product and reuse the same story across ad copy, email, sales-page copy, and social so the message compounds across every channel.

Should I use FAQ schema on collection pages or write a blog post per question?

Both work for AI search. Collection-page FAQ with schema markup sends clicks straight to a page full of products, which converts better; a blog-post-per-question strategy gives each question its own URL and can rank for more query variations. If you can only do one, put the FAQ on the collection page and put it near the bottom so it does not push your products below the fold.

What shipping term should I use with a Chinese supplier: EXW, FOB, or DDP?

EXW is the safest default because it makes your supplier’s true cost of goods and your true shipping cost visible as separate line items. FOB and DDP quotes routinely pad the shipping leg, so you end up overpaying without knowing it, and DDP also blocks you from claiming tariff refunds.

How often should I renegotiate FedEx and UPS rates?

Every 3 months if you ship real volume, and at minimum once a year. Carriers have been raising rates aggressively, so quarterly renegotiation is now the working baseline for high-volume ecommerce sellers.

What makes an ecommerce event worth attending?

Three things: the event is small enough that you can actually meet everyone, attendees are open about their real revenue and profit numbers, and there is an active community around the event you can plug into before and after. If the event has no community layer, it is mostly a session firehose and the ROI drops fast.

Ready To Get Serious About Starting An Online Business?

If you are really considering starting your own online business, then you have to check out my free mini course on How To Create A Niche Online Store In 5 Easy Steps.

In this 6 day mini course, I reveal the steps that my wife and I took to earn 100 thousand dollars in the span of just a year. Best of all, it's absolutely free!

Podcast: Download (Duration: 46:05 — 53.0MB)

The wholesale outreach strategy that closes at 50%, in one sentence: send a free physical product sample to a targeted independent retailer, then follow up every two weeks by phone and email using a CRM until they either place an order or ask you to stop. That is the exact playbook Dawn LaFontaine of Cat in the Box uses to close half of every prospect she samples, and she shared the full breakdown on this episode of the My Wife Quit Her Job podcast.

Dawn is a former stay-at-home mom who could not get hired anywhere after 27 years out of the workforce, so she bootstrapped a pet product company that is now sold in over 100 independent stores across the US and Europe. Along the way she also flipped her whole product line, from cardboard cat houses that got press but lost money on shipping to a range of wool cat toys handmade by a Fair Trade craft guild in Nepal.

Below is the full playbook: how she noticed the accidental Instagram signal that led to her hit product, why she pivoted away from her original press-magnet product, how she sourced ethical wool toys from Nepal, and her exact repeatable wholesale outreach process that batting-averages a 50% close rate without a single ad.

Key takeaways

• Wholesale outreach closes at 50% when you send a real product sample, then follow up every 2 weeks by phone and email until the retailer buys or tells you to stop. Persistence, not pitching, is the whole game.
• Use a CRM (Dawn uses HubSpot) to schedule the every-two-weeks follow-ups. A spreadsheet does not work; you will forget.
• Send the sample with a short paper letter that mentions the store employee you spoke to by name, includes the MSRP, the wholesale price, and a QR code to your Faire page.
• Pivot away from a product when it feels like pushing a boulder uphill: hard to sell direct, hard to place in stores, hard to ship profitably. Everything you learned still transfers to the next product.
• The best product-selection filter is still “fits in a shoe box.” Small, light, and unbreakable makes every downstream margin (shipping, packaging, storage) work.
• Retailer-buyer leads live in your industry’s independent trade magazine. Mine the “letters to the editor” and vendor call-outs for stores worth pitching.
• Nepal is a serious wool sourcing option. Higher unit cost than China, Fair Trade certified, azo-free dyes, and small minimum order quantities make it viable for niche pet and gift brands.

The wholesale outreach strategy that closes at 50%

Dawn’s wholesale outreach strategy closes at 50% because it is built on a physical sample plus disciplined every-two-weeks follow-up, not on ads or cold pitches. Half of every independent retailer she sends a sample to eventually places a wholesale order, and once they buy, they buy again with almost no additional selling.

She calls this her “batting 500” rate. Out of every 100 stores she samples, roughly 50 become paying wholesale customers, most of them recurring.

The rest of this post walks through the exact steps she uses, plus the product and sourcing decisions that made the strategy work in the first place.

Step 1: Build a targeted list of independent retailers worth pitching

Build your prospect list from your industry’s independent trade magazine, not from generic Google searches or business directories. The pet industry has a specialized magazine that surfaces the most cutting-edge independent pet stores in the country, the ones actually competing against Chewy and Amazon on differentiated selection instead of price.

Dawn’s list-building routine is boring on purpose. Every time a new issue arrives, she reads through it and writes down the name of every independent retailer that appears in letters to the editor, vendor spotlights, or store profiles. Those are the shops actively hunting for unique product.

That running list is her outbound pipeline. She works from the top down whenever she has time to make calls.

Step 2: Make the first call and offer a free product sample

Call the store, ask for the buyer by name, and offer to send a free physical sample of your product. Do not pitch on this call. The goal is a single yes: permission to send the sample.

Here is roughly what Dawn says when the clerk picks up.

“Hi, is [Buyer’s name] in? I’m Dawn LaFontaine from Cat in the Box, and I have a product sample I’d love to send to [Buyer]. Would she be interested in receiving a free sample?”

If the buyer is out, the clerk tells her whether email or a mailed sample is preferred. Either way, she gets both the buyer’s name and permission for the first touch. Roughly 25% of the time the buyer is actually on the phone, which turns into a real conversation.

Step 3: Send the sample with a physical letter, MSRP, and a QR code to Faire

Ship the sample inside a small package with a short printed letter, the MSRP, the wholesale price, and a QR code that points directly to your Faire page so the retailer can place a first order in one scan. The letter is on letterhead, three short paragraphs long, and mentions the store employee she spoke to by name.

The opening line does a lot of work. Something like: “I spoke to Chris in the store today and she said you wouldn’t mind receiving a product sample.”

That single sentence proves she actually called the store, took a real interest in the business, and did the human work before shipping the box. In an era of automated LinkedIn spam, a physical letter that references a real conversation with a real employee is a differentiator.

The Faire QR code is deliberate. Sending the buyer to Faire means you pay 0% commission on any order from a customer you referred yourself, since Faire only takes commission on the leads it generated. That is a meaningful margin difference at scale.

Step 4: Follow up every 2 weeks using HubSpot until they buy or say stop

Schedule a follow-up in HubSpot (or any CRM) for exactly 2 weeks after the sample ships, and repeat every 2 weeks with alternating calls and emails until the buyer either places an order or asks you to stop. A spreadsheet does not work here; you will forget who is due, and the whole system collapses.

Dawn had one prospect who took a full year of polite every-two-weeks follow-up before placing the first order. That one customer now buys from her forever, on autopilot.

Rotate the follow-up angle so you never sound like a recording.

• Call one: “Just circling back to make sure the sample arrived.”
• Email two weeks later: “Did any of the store cats get a chance to play with it? Any feedback?”
• Call two weeks after that: “Any questions on pricing or shipping?”
• Email two weeks after that: “Happy to send another sample if the first got lost in the shuffle.”

The message is polite persistence, not pressure. As long as the retailer has not said “please stop,” they are still in the pipeline.

Why the wholesale channel outperforms Amazon and paid ads for niche brands

Wholesale outperforms Amazon and paid ads for a lot of niche brands because wholesale buyers are almost zero-maintenance once acquired: they reorder on their own, they pay shipping, they do not return product, and they do not demand coupons. Wedding and event planners on Dawn’s list keep reordering without ever asking for a discount because they simply need the product for the next event.

Cat in the Box’s revenue now splits roughly a third own website, a third Amazon, and a third wholesale. Dawn wants wholesale to keep growing because it is the most profitable and least stressful of the three channels.

The trade-off is that wholesale is founder-led work. Dawn is a self-described “raging introvert” who avoided cold-calling for years, tried postcards and email-only outreach and rewrites of both, and none of it worked. The only thing that finally moved orders was the phone plus the physical sample plus the CRM-driven follow-up.

How one Instagram unfollow led to a 6-figure wool cat toy line

Dawn’s best-selling product line came from noticing who unfollowed her on Instagram, clicking through, and discovering that the unfollower sold a wool cat toy. She had never considered wool as a category, ordered 200 units each of two designs she designed herself (a red wool mouse with a 6-foot tail and a rainbow-tailed mouse), and sold out of all 400 units almost immediately.

The design and selling process worked in three moves.

She sourced from a Fair Trade craft guild in Nepal instead of China. The wool comes from New Zealand, the felt artisans are low-income women who get a living wage, health and retirement benefits, and incentives for keeping their daughters in school.

The dyes are also azo-free, which mattered to her because cats chew the toys. China quotes she got refused to confirm the dye chemistry, so she walked.

She designed the products herself over video. Finding the factories took heavy digging (a travel-guide article about ethical Nepal manufacturers was the entry point), then multiple rounds of $3 mailed samples until the octopus, mouse, and other designs looked right. One design took about a year of back and forth.

She launched with 200 units per design, listed on her website, emailed her list of ~5,500 subscribers, and offered the same product wholesale on Faire. The influencer moment came later: a TikTok creator with 11 million followers posted about the wool mouse organically, and inventory sold out across every channel.

How to know when to pivot your product

Pivot away from a product when direct sales feel like pushing a boulder uphill, retailers will not stock it, and the shipping economics do not work no matter how you redesign. Everything you learned building the first product still transfers to the next one, so a pivot is not lost effort.

Dawn’s original product was a line of designer cardboard cat houses that got her featured in Parade, NBC News, The Boston Globe, and US News and World Report. Press did not translate to profit. Three problems compounded.

Shipping ate the margin. The compact designs still cost around $12 to ship, and the larger, assembled designs could cost $30 to $40 to ship to California. She had to keep raising prices to cover free shipping, which suppressed conversion.

Independent pet stores would not stock them. Even her most compact cardboard design took up too much shelf space for small store footprints, and the margin was too thin for the retailer.

Minimum order quantities were brutal. Even for a seasonal SKU like her haunted-house cat design, she had to order 1,500 units at a time, then store and re-order them for the following year.

The signals to watch for are consistent across categories: hard to sell direct, hard to place wholesale, and unit economics that only work at volumes you cannot hit yet. Fix what you can, and if the boulder is still uphill, try something else.

The best product-selection filter: “fits in a shoe box”

The single best product-selection filter for a new ecommerce business is whether the product fits in a shoe box, because small and light drives every downstream margin: cheaper shipping, cheaper packaging, cheaper storage, and lower minimum-order quantities. Dawn credits ignoring that filter with the two years she spent trying to make cardboard cat houses profitable.

Wool cat toys pass the shoe-box test on every dimension. They cost around $5-$6 to produce, they are cheap to store, cheap to package, cheap to ship, they retail for $17-$25, and buyers reorder them for multiple cats or as gifts.

If you are still in the product-selection phase, run every candidate through the shoe-box test before you order the first sample.

Frequently asked questions

How do I get wholesale orders for my product without ads?

Build a targeted list of independent retailers from your industry’s trade magazine, call each store and offer to send a free physical product sample, then follow up every 2 weeks by alternating phone and email until they buy or ask you to stop. Dawn LaFontaine of Cat in the Box closes about 50% of the retailers she samples using exactly this loop.

How often should I follow up with wholesale prospects?

Every 2 weeks, using a CRM to schedule the reminders, until the retailer places an order or explicitly asks you to stop. Some buyers take a year of polite every-two-weeks follow-up before the first order, and once they buy they typically become recurring customers with almost no additional selling.

What is the best CRM for tracking wholesale outreach as a small brand?

Any CRM that lets you set a recurring follow-up reminder will beat a spreadsheet. Dawn uses HubSpot’s free tier for her wholesale outreach because the reminders never let a prospect slip out of the pipeline, and spreadsheets she had used before consistently failed her.

Should I use Faire for wholesale ecommerce?

Yes, but drive your own retailers to Faire rather than waiting for Faire to find them for you. When you refer a retailer yourself, you pay 0% commission on their orders, which meaningfully protects margin at scale. Faire’s built-in discovery is getting more crowded, so treat it as your checkout and re-order infrastructure, not your lead source.

What should I include in a wholesale sample package?

Include the product sample, a short printed letter on letterhead (2-3 paragraphs, referencing the employee you spoke to by name), the MSRP, the wholesale price, and a QR code to your Faire page so the retailer can place a first order in one scan. Small, personal, and easy to act on beats a generic catalog every time.

Is it cheaper to manufacture wool products in Nepal or China?

Nepal is typically more expensive per unit than China, but the trade-offs (Fair Trade certification, azo-free dyes, small minimum order quantities, and full transparency on materials and labor) can be worth the premium for niche pet, gift, and lifestyle brands. Dawn walked away from China because she could not get clear answers on which dyes were used on toys her customers’ cats would chew.

When should I pivot my ecommerce product?

Pivot when direct-to-consumer sales feel like pushing a boulder uphill, retailers will not stock the product, and the shipping economics do not improve after multiple redesigns. Everything you learned on the first product (sourcing, packaging, retailer relationships, ad testing) transfers to the next one, so pivoting is compounding effort, not wasted effort.

What product criteria should I use when starting an ecommerce business?

Start with the “fits in a shoe box” filter: small, light, and unbreakable makes shipping, packaging, storage, and minimum-order economics all work in your favor. Dawn’s cardboard cat houses failed the test and lost money on shipping; her wool cat toys pass on every dimension and now drive the majority of her growth.

Ready To Get Serious About Starting An Online Business?

If you are really considering starting your own online business, then you have to check out my free mini course on How To Create A Niche Online Store In 5 Easy Steps.

In this 6 day mini course, I reveal the steps that my wife and I took to earn 100 thousand dollars in the span of just a year. Best of all, it's absolutely free!

Podcast: Download (Duration: 46:54 — 54.0MB)

The ecommerce strategies worth stealing from Sellers Summit 2026, in one paragraph: rank in Reddit so ChatGPT and Perplexity cite your brand, launch on TikTok Shop with a realistic budget (not the fairy-tale version), renegotiate supplier and shipping costs instead of trying to fully leave China, sell your business as a “fixer-upper” that qualifies for an SBA loan, vibe-code the 60% of Shopify apps you are overpaying for, clone winning meta ads with an n8n automation, and turn customer reviews into short AI-generated ad videos in under 30 minutes. That is the full 2-day recap from this year’s event in Fort Lauderdale, walked through session by session with my co-host Toni on the My Wife Quit Her Job podcast.

We ran Sellers Summit 2026 single track this year with 30-minute talks, which turned out to be the format change that made the biggest difference: every speaker had a packed room, every attendee walked out with one or two things to implement instead of 122 slides they could not act on.

Below is the entire recap: what each speaker taught, the tactics that got the loudest reaction in the room, and the one piece of life advice from Hydrofighter’s Mike Barnhill that I have been carrying with me since the closing keynote.

Key takeaways

• Reddit is now a top ranking lever for AI search. ChatGPT, Gemini, and Perplexity all lean heavily on Reddit content, so a non-spammy Reddit presence is how you get cited in AI answers.
• Amazon is losing product discovery. Buyers now discover on TikTok and Reddit, research on AI, and buy on Amazon. Your funnel needs to sit upstream of Amazon, not on it.
• TikTok Shop is a paid channel, not free money. A realistic launch spend for a serious brand runs close to $100K when you count creators, ads, and infrastructure.
• Sourcing did not actually leave China in 2025 despite the tariff panic. Renegotiating packaging, shipping terms, and MOQs with your existing supplier beats a full-blown re-source for most sellers.
• Sell your business as a fixer-upper, not a finished house. Buyers want visible upside and an SBA-loan-eligible P&L.
• About 60% of the Shopify apps you pay monthly for can be vibe-coded in an afternoon. Kill the top 3 by monthly cost first.
• You can clone any Facebook ad from the ads library and drop your product into it using an n8n automation. Watch for the AI-goes-wrong outputs before you spend on media.
• Customer reviews plus a script generator plus Remotion equals a short AI-generated ad video in about 30 minutes.
• Wake up every day thinking this could be the best day of your life. Mike Barnhill’s line was the single most-quoted piece of advice from the whole event.

Why Sellers Summit is single track with 30-minute talks this year

We made Sellers Summit 2026 single track with 30-minute talks because the format change forces every speaker to distill their talk down to one or two actionable takeaways, which is exactly what attendees need. It is easily one of our top three events out of the 10 we have run, and the format is a big reason why.

A speaker like Brett Curry can fill an hour with 122 slides of pure signal, but no attendee walks out of a 122-slide hour and actually implements anything. Cap the same speaker at 30 minutes and they focus on one or two ideas, and the whole audience implements those ideas the next week.

Single track also solves a scheduling problem I could never win with two tracks. Sourcing sessions apply equally to Amazon and Shopify sellers, so putting them “on the Amazon track” always frustrated half the room. And attendee favorites like Tiffany Ivanovsky pulled the whole audience regardless of topic, which starved whichever talk we scheduled opposite her.

How to rank in Reddit for AI search (Jeff Skolnick’s session)

Jeff Skolnick’s Reddit session got more attendee reactions than any other talk at Sellers Summit 2026 because ranking in Reddit is now a core requirement for showing up in AI search answers. ChatGPT, Gemini, and Perplexity all rely heavily on Reddit content, so brands with a real Reddit presence are the ones AI engines cite when a shopper asks for a recommendation.

The core mechanic is simple: fewer people click the 10 blue links, more people read the AI overview, and the AI overview leans on Reddit. If real Reddit users mention your brand in a relevant subreddit, you show up in the AI answer.

Jeff’s talk walked through the exact non-spammy steps: how to build karma, which subreddits to participate in, and what will get you shadowbanned instantly. Reddit shadowbans without notification, so mistakes are silent and expensive.

If you sell ecommerce and you have not touched Reddit yet, this is one of the highest-leverage moves you can make in the next 90 days.

The 5 sourcing and shipping cost cuts every ecommerce seller should ask for (JK’s session)

The 5 sourcing and shipping tactics JK covered are all no-brainer negotiation asks that most sellers never make: shrink your packaging dimensions, renegotiate FedEx and UPS rates, ask suppliers for better terms on repeat orders, keep a second supplier active for leverage, and rethink shipping incoterms.

What made the session hit was context. During my keynote I asked how many attendees actually moved their sourcing out of China after the 2025 tariff scare. Only two hands went up in the whole room.

China still makes almost everything, and the vast majority of sellers who tried to re-source ended up staying. Given that, the practical play is to squeeze better terms out of your existing China supplier and shipping chain, not to force a full re-source that fails.

Packaging alone is a fast win. Shrinking a carton by one inch can drop shipping cost per unit noticeably at scale, and it costs you nothing but a redesign.

How to launch on TikTok Shop with a realistic budget (Ian Page’s session)

Ian Page from Bullseye Sellers gave the most honest TikTok Shop talk of the event: launching seriously on TikTok Shop costs close to $100K for the highest tier when you add up creators, ad spend, agency fees, and infrastructure. The lower tiers still cost real money.

The reason Ian’s approach is different is that most agency-run talks make the process sound complicated enough that you feel forced to hire them. Ian did the opposite. He broke down what you would spend to run it yourself, what you would spend to hire an agency, and the trade-offs of each, then said out loud that plenty of brands should do it themselves.

TikTok Shop matters right now because Amazon is losing the top of the ecommerce funnel. Discovery is moving to TikTok and Reddit, research is happening in ChatGPT and Perplexity, and only the checkout is still happening on Amazon. If your brand is not on TikTok Shop and your category fits, you are handing that discovery layer to competitors.

The category caveat: Ian was explicit that some categories are natural TikTok Shop fits and others are not. Ask him or someone honest before you commit budget.

How to sell your ecommerce business for the best multiple (Chuck Mullins’ session)

The best-multiple ecommerce business is the one that reads as a fixer-upper with visible upside and qualifies for an SBA loan, not the one that has been polished to look perfect. Chuck Mullins from Quiet Light walked the audience through three example businesses and asked the room to guess which sold for the highest multiple. The obvious “cleanest” business consistently sold for the lowest.

Chuck’s core analogy: too many sellers prep their business the way they prep their house for sale, painting walls, staging flowers, and hiding the flaws. That is the wrong instinct. Buyers of ecommerce businesses want to see the levers they can pull to grow the company after acquisition.

Two concrete moves that lift your multiple.

Make your business SBA-loan-eligible. Clean books, proper corporate structure, and financials that read cleanly to an underwriter widen your buyer pool and speed up close, which lifts multiple.

Show visible upside, not perfection. Under-optimized ad accounts, an unlaunched product line, or a wholesale channel you have not tapped are all features to a strategic buyer, not bugs.

If exit is on your 3-year roadmap, get a valuation now (Quiet Light offers a free one) so you know which fixes matter before you start “improving” the business into a lower multiple.

The brand storytelling framework that ties every ad and email together (Scott Cunningham’s session)

Scott Cunningham’s brand storytelling framework is a structured StoryBrand-style narrative that you write once for your brand, then reuse across meta ads, email flows, and sales-page copy so every touchpoint reinforces the same story. Scott also released a free AI tool that walks you through the framework and outputs the story assets, and he gave the QR code to every attendee in the room.

The talk hit hard enough that the Q&A line looked like a rock concert meet-and-greet. Storytelling is one of those topics every seller nods at and almost no seller actually does, and Scott’s tool removes the excuse.

The freebie is also included in the Sellers Summit 2026 virtual pass, so recordings buyers get the same access.

How to turn customer reviews into AI-generated ad videos in 30 minutes (Liz and Dana’s session)

Liz Saunders and Dana Jaunzemis showed a workflow that pulls customer reviews off your product page, feeds them into a script generator, and outputs short AI-generated ad videos in about 30 minutes using Remotion. I ran the workflow the day after the event and it worked the first time.

The best part of the workflow is the director loop. You produce a video, tell the tool in plain English what you do not like about it (“this cut is too fast,” “swap the second shot,” “change the voice”), and it re-renders on the fly. The output is good enough for meta ad creative, short social posts, and on-site promo videos.

The talk also used real attendee products in the live demo (Kelly Dream’s brand, and one of mine), which is what pushed it from theoretical to obviously doable. Half the appeal of Liz and Dana as speakers is that they demo on real businesses, not fake case studies.

How to vibe-code Shopify apps and stop paying monthly for 60% of them (my session)

Roughly 60% of Shopify apps in the app store can be vibe-coded yourself in an afternoon, which means most stores are overpaying by hundreds of dollars a month for functionality they could build. That was the argument I made in my session, and I demoed the process live on stage without touching the code panel.

The remaining 30-40% of apps (things with complex integrations, real infrastructure, or ongoing maintenance) are legitimately worth paying for. But the apps most sellers pay $19, $29, or $49 a month for are mostly the vibe-codeable category.

When I asked the room how many people were using the specific Shopify apps I flagged as rip-offs, a large chunk of the audience raised their hands. The point is not that those apps are bad. The point is that once you code your first replacement, the second one is easier, and you compound the savings every month.

Start with your 3 highest-cost monthly apps. Kill the most expensive one first.

How to clone winning Facebook ads with an n8n automation (Leah Segovia’s session)

Leah Segovia demoed an n8n automation that pulls any winning ad from the Facebook Ads Library and generates a new ad with your product dropped into it. He handed the entire n8n workflow to the audience for free, which is unusual for a session of this complexity.

The demo included a working deodorant example, a weight-loss example, and, memorably, a baby carriage example that no one in the room will forget.

Alongside the wins, he showed the AI-goes-wrong outputs so nobody left thinking this is a magic button. When AI goes wrong here, it goes very wrong, so review every generation before you spend media on it.

The output quality is also improving fast. What looked “good, not great” this year would have been unusable six months ago, and will be unrecognizable six months from now.

The product research framework that predicts Amazon launch success (Isabella’s session)

Isabella Ritz shared a product-research framework and an accompanying AI app that takes parameters from standard Amazon research tools (Helium 10, Jungle Scout) and outputs a numeric probability that a product launch will succeed. Her track record with this framework is strong enough that following it makes a flop-free launch close to expected.

The live demo was the moment the talk landed. She had populated the inputs ahead of time and clicked go on stage, and the app returned a probability score the whole room could see and interrogate.

If you are launching on Amazon or picking your first product, this is the kind of framework that pays for the whole conference ticket by itself.

The one-person content machine (Chris Schaeffer’s session)

Chris Schaeffer runs the content operation for 6 ecommerce companies as a one-person team using AI, and his session showed the exact stack that lets a single operator produce team-level content output. Everything he taught, he actively runs in his own agency, which is why it always translates.

Chris also released his content-generation freebies with no email gate. Click and download.

His session tied cleanly into Liz and Dana’s Remotion workflow. We did not plan the overlap, and it reinforced the message across day two: AI turned a full content team into a one-person job in the last 18 months.

What to do when your audience turns on you (Dave Bryant’s session)

Dave Bryant walked through the knitting.com saga at Ecomcrew and how they survived their audience turning on them, and the talk was the second most laugh-heavy of the whole event. In an era where a single social-media post can end a company overnight, “audience turnaround crisis management” is a core skill.

Dave’s playbook boils down to three moves: acknowledge quickly, correct what is fixable, and outbuild the outrage cycle by focusing on new customers and new value. Silence and defensiveness both make the crisis worse.

The halo effect: why your brand needs to be everywhere (Tiffany Ivanovsky’s closing session)

Tiffany Ivanovsky’s session on the halo effect tied together every day-two theme by showing why a brand needs to be everywhere at once: buyers discover on TikTok, research on AI, and purchase on Amazon, and every channel reinforces the others. She did a live TikTok recording on stage to demonstrate that messing up on camera is fine and posting anyway is what matters.

Her real lesson is compounding video reps. She showed her own mistake reel (including the time she swallowed a bug on a live) as proof that mistakes drive engagement, not away from it. Liz Saunders is another example in the same arc: she was already a good speaker 18 months ago, and now, after weekly webinars, weekly lives, and office hours, she is on paid-speaker level.

If you are avoiding video because you might mess up, that is the wrong instinct. Reps are what turn “pretty good” into “phenomenal.”

Closing keynote: wake up every day thinking this could be the best day of your life

The single piece of advice I have carried with me since Sellers Summit 2026 is from Hydrofighter’s Mike Barnhill: wake up every day thinking this could be the best day of your life. Toni delivered the closing keynote, and Mike’s line ended up being the through-line she landed on.

The context: business creates once-in-a-lifetime opportunities, and most of us are so beaten down by the day-to-day that we do not take them. The mindset flip is not toxic positivity. It is a decision to treat the day as if a great thing could happen, so when the opportunity shows up, you are actually in a position to say yes.

Toni also credited Danna Jaunzemis (our 2023 closing keynote) with the pre-talk mindset that killed her nerves: everyone in the room wants you to succeed, because nobody wants to sit through a bad talk.

Frequently asked questions

What are the biggest ecommerce trends from Sellers Summit 2026?

The biggest trends were Reddit as a top ranking signal for AI search, TikTok Shop and Reddit taking over product discovery from Amazon, AI-generated ad video that is finally good enough to run, and buyers valuing ecommerce businesses higher when they look like fixer-uppers with visible upside instead of over-optimized “clean” listings.

How do I rank in Reddit for AI search?

Build a real, non-spammy Reddit presence in the subreddits your buyers already use. Engage as a helpful human, avoid link-drops that trigger shadowbans (Reddit bans without notification), and prioritize being mentioned by other real users in your category. ChatGPT, Gemini, and Perplexity all lean heavily on Reddit content when generating AI answers.

How much does it cost to launch on TikTok Shop?

A realistic launch spend for a serious brand runs close to $100K at the top tier when you factor in creators, ad spend, agency fees, and infrastructure, per Ian Page of Bullseye Sellers. Lower tiers cost less but still require real budget. TikTok Shop is a paid channel, not free money.

Should I move my ecommerce sourcing out of China in 2026?

Most sellers who tried in 2025 ended up staying with their China supplier because the tariff picture kept shifting and China still makes almost everything. A better default is to renegotiate packaging dimensions, shipping incoterms, MOQs, and freight rates with your existing supplier, and keep a second supplier active only for leverage.

What makes an ecommerce business sell for the best multiple?

Visible upside plus SBA-loan eligibility. Buyers want to see the levers they can pull post-acquisition (an under-optimized ad account, an untapped wholesale channel, an unlaunched product line), and they want financials clean enough for SBA underwriting so more buyers can compete for the deal. Over-polishing your business into “perfection” removes the upside story and shrinks your multiple.

Which Shopify apps should I replace with my own code?

Start with the 3 apps you pay the most for each month and evaluate whether the functionality is genuinely complex (real integrations, ongoing maintenance) or something a mid-range LLM can vibe-code in an afternoon. Roughly 60% of Shopify apps fall into the second category, and killing them adds up to real monthly savings.

Can I really turn customer reviews into ad videos with AI?

Yes, in about 30 minutes end to end. The workflow Liz and Dana demoed pulls reviews from a product page, generates a script, produces a short video in Remotion, and lets you iterate in plain English (“cut the second shot,” “change the voice”) until the video is ad-ready. The output is good enough for meta creative, short social posts, and on-site promo.

Ready To Get Serious About Starting An Online Business?

If you are really considering starting your own online business, then you have to check out my free mini course on How To Create A Niche Online Store In 5 Easy Steps.

In this 6 day mini course, I reveal the steps that my wife and I took to earn 100 thousand dollars in the span of just a year. Best of all, it's absolutely free!

Podcast: Download (Duration: 50:26 — 58.0MB)

The AI automation stack behind the fastest-growing ecommerce brands right now, in one paragraph: an AI-scored influencer outreach agent (email plus TikTok DMs) that filters real creators on 100+ data points, an n8n workflow chaining Gemini plus NanoBanana plus Suno plus VO3 plus FFmpeg to clone any winning meta ad with your product dropped in, and a meta-ad-to-landing-page-to-Amazon funnel that lifts conversion and slashes click cost. That is the exact stack Leo Sgovio used to sell out 5,000 units of a Q4 game before Christmas without giveaways, and he walked through it end to end on this episode of the My Wife Quit Her Job podcast.

Leo runs Spliced.io, an influencer and affiliate platform for TikTok Shop and Instagram, and Combomat, a marketing automation tool for sellers. He is also a decade-plus Amazon operator who was head of innovation at Viral Launch, so his stack is what he actually runs on his own products, not theory.

Below is the full breakdown: the influencer outreach system that beats spam bots, the AI ad-cloning workflow anyone can copy, TikTok Shop launch economics you can actually trust, why AI clones for product claims just got banned, and the meta-to-Amazon funnel that is quietly outperforming direct-to-listing traffic.

Key takeaways

• Leo sold out 5,000 units of a new Amazon game in Q4 with no giveaways, using only creator content and Amazon PPC.
• Most influencer outreach tools spam creators. The differentiator is scoring each creator on 100+ data points before sending an invite so your message reaches people likely to say yes.
• You can clone any winning meta ad with an n8n workflow: Gemini analyzes the video, Suno generates music, NanoBanana generates frames, VO3 generates 8-second video segments, FFmpeg stitches everything together.
• Launching on TikTok Shop plus Amazon at the same time creates a halo effect that lifts Amazon sales dramatically. One of Leo’s competitors sold 70,000 units in Q4 running this exact play.
• TikTok Shop banned AI clones from making product claims (skincare, physical goods) in late 2025. AI is still fine for informational content and general social proof.
• The winning funnel is meta ad -> landing page -> Amazon, not meta ad -> Amazon direct. The landing page warms the buyer so the Amazon click converts instead of getting poached by cheaper competitors on the product page.
• Google Ads to a landing page runs at 15-20 cent clicks. The same traffic sent to an Amazon listing costs roughly triple.
• Amazon reported its lowest number of new sellers this year and just started charging developers $1,400/year to use its APIs. Fewer sellers means less competition for those willing to stay profitable.

The AI automation stack behind the fastest-growing ecommerce brands

The AI automation stack that fastest-growing ecommerce brands run in 2026 has three layers: AI-scored influencer outreach for TikTok Shop and Instagram, an n8n-based ad-cloning workflow that produces meta-ready video from any winning competitor ad, and a meta-to-landing-page-to-Amazon funnel that lifts conversion while cutting click cost. Together, the stack lets one operator do the work of four or five marketers.

Leo’s own results are the case study. His Q4 game launch sold 5,000 units and sold out before Christmas without giveaways, driven almost entirely by creator content and AI-produced ad video. The AI automation freed him to run marketing at that scale without hiring a team.

The rest of this post walks through each layer of the stack in order.

How to run influencer outreach at scale without spamming creators

Run influencer outreach at scale by scoring every prospective creator on 100+ data points (GMV, number of active brand deals, average video earnings, audience match, timing signals) before sending an invite, so your messages reach creators who are likely to engage instead of ignoring one more spam DM. Leo’s Spliced.io platform uses this scoring layer as its main differentiator against outreach tools that just blast creator databases.

The volume math is real. You can send about 1,000 messages per day on TikTok Shop or via Instagram email, and the tools that will let you send that many are the same tools that get creator inboxes ignored. Creators receive so many templated pitches that a generic message is dead on arrival.

The fix is to filter the API results yourself. When you query TikTok Shop or Instagram creator databases, you get back a list that “matches” your criteria (region, category, GMV band). That list still includes creators who are wrong for your product: a family-games creator whose audience is toddlers when your game is 5-12, or a $10K/month affiliate whose economics will not accept your 20% commission.

Score the returned list before you invite. That is the entire moat.

How much to offer TikTok Shop and Instagram affiliates

The right offer for a TikTok Shop or Instagram affiliate is one that matches or beats their current per-video earnings on comparable products, because they will not replace something that is already making them money with an offer that pays less. Leo’s rule: if a creator averages $10 commission per product video today, your offer needs to be in that neighborhood or above, especially when your storefront has no conversion history yet.

The cold-start problem on TikTok Shop mirrors the cold-start problem on Amazon a decade ago. Established affiliates want to see that your shop converts before they will spend a video slot on your product. Without a track record, you are competing against products with proven per-click earnings.

Leo runs a 3-phase creator strategy based on brand size:

• Phase 1 (small brand): micro-influencers only. Goal is to unlock the first $2,000 GMV milestone that lifts your weekly message cap on TikTok Shop.
• Phase 2 (past $2K GMV): open your message cap to 5,000/week and start hunting mid-tier affiliates. Use “open collaborations” so creators apply to you instead of you DMing every one.
• Phase 3 (past $50K GMV): 50,000 messages/week unlocked. Go after the top affiliates who now see you as an established shop.

To speed the cold start, some sellers use refundable-sample purchases: the creator buys the product, you refund them, the purchase still counts toward your GMV milestone. Combined with friends-and-family purchases (with more than one unit each), you clear the $2K threshold faster.

The initial cold-start phase can last a couple of months. That is normal.

How to clone any winning meta ad with n8n, Gemini, NanoBanana, Suno, and VO3

The n8n workflow to clone a winning meta ad chains 5 tools: Gemini analyzes the source video and generates a script, NanoBanana renders your product into the first and last frames, Suno generates on-theme music, VO3 generates the 8-second video segments the script calls for, and FFmpeg stitches all segments together into the final ad. Everything runs inside n8n with an Airtable interface for inputs, and the final video lands in Airtable ready to review.

Leo demoed the workflow live on stage at Sellers Summit with two cloned ads that were arguably better than the originals.

Example 1: he took a Secret deodorant ad (with the disco-ball raspberry finish), swapped in his own deodorant (an old Amazon product he sold in Canada), rendered a 3D product image through NanoBanana, and asked the workflow to keep the choreography but re-theme the color to match his brand’s green and flowers. The output was cleaner than Secret’s original.

Example 2: he took an Alo Yoga leggings ad (two girls eating popcorn in a mountain cottage), swapped in a different leggings brand’s model and logo on the popcorn box, and the workflow produced a near-identical ad end to end. No script writer, no video editor.

Here is the flow in plain steps:

1. Feed the source video URL and your product image into Airtable.
2. Gemini analyzes the source video, extracts the script beats and pacing, and outputs a scene-by-scene script.
3. Suno generates on-theme background music matching the video’s mood and length.
4. NanoBanana takes each scene beat plus your product image and renders the first and last frame of each 8-second segment.
5. VO3 generates each 8-second video segment using those frames as anchors.
6. FFmpeg stitches segments and the music track into the final MP4.
7. Airtable surfaces the final video for review.

Where this matters most: sellers currently paying scriptwriters and video editors can produce ready-to-run meta creative in a single pass, at a cost that is a rounding error against media spend. Leo has been giving the workflows away free on LinkedIn.

What TikTok Shop’s new AI-clone ban means for your ads

TikTok Shop banned AI clones from making product claims in late 2025, which means an AI avatar (even one built from your own likeness) is no longer allowed to say “these glasses are comfortable” or “this cream cleared my skin.” AI-generated informational content and general social proof are still allowed; AI-generated product endorsements are not.

The rule closes what was, until recently, the loudest AI ecommerce loophole: mass-producing “creators” who sell your product. Because AI outputs are now nearly impossible to distinguish from real people, TikTok classified the practice as deceptive.

Two things you can still legally do with AI on TikTok:

• Informational or educational content in your brand voice: “how to build a skincare routine,” “how to pick a family game.” No product claim, no direct sell.
• General social proof on your owned website: AI-generated testimonials as marketing copy, understanding that other platforms and jurisdictions have their own rules on advertorials and disclosure.

Meta has not (as of publication) enforced the same restriction on AI product claims, so ad-cloning workflows like Leo’s (which use no avatar, only product plus animation) remain within policy. Meta requires an AI disclosure flag when creative is AI-generated; check that setting before you spend.

The meta ad to landing page to Amazon funnel that outperforms direct-to-listing

The funnel that consistently outperforms direct-to-Amazon is meta ad to landing page to Amazon, because the landing page warms the buyer before they hit a product page full of competing ads, cheaper alternatives, and distractions. Cold traffic sent straight to an Amazon listing frequently gets poached by a competitor’s sponsored ad on your own detail page.

Leo also runs Google Ads on the same landing-page pattern. The click math is dramatic: Google Ads to a landing page costs roughly 15-20 cents per click, and the same Google traffic sent directly to an Amazon listing costs about triple.

The landing page functions as an advertorial. It explains why the product is right for this buyer, sets expectations on price, and produces a click to Amazon that is already convinced. When that pre-warmed click hits the listing, they buy instead of clicking away to the sponsored competitor above the fold.

The reason not to just send the sale to your own Shopify checkout is category-dependent. If you sell exclusively on Amazon, external traffic to a warmed listing helps ranking and lifts organic sales. If you have a Shopify store, sending the sale to your own site captures the email and the customer relationship, which is almost always the better long-term play.

Why launching on TikTok Shop and Amazon together beats either channel alone

Launching on TikTok Shop and Amazon at the same time creates a halo effect: TikTok Shop drives discovery, buyers research on AI or Google, and then most of them actually buy on Amazon because the checkout is familiar. One of Leo’s competitors sold 70,000 units in Q4 running exactly this dual-launch strategy, most of the revenue landing on Amazon even if the TikTok Shop side ran near-break-even.

The attribution problem is real. Most brands claim they see the halo effect and try to prove it with Amazon’s attribution API, but the API misses the most common shopping pattern: buyer sees the product on TikTok, opens Google, finds the Amazon listing, clicks the ad, buys. That path has no attribution trail.

Practical implication: do not try to make TikTok Shop economics pencil in isolation. Measure the combined lift on Amazon during the launch window and evaluate the two channels together.

Where Amazon is headed in 2026 (and why it might be a better opportunity, not a worse one)

Amazon reported its lowest number of new sellers this year, which counterintuitively makes Amazon a better opportunity for operators willing to stay profitable, because less competition means more room to rank and less bidding pressure on ads. Fees are still rising (advertising, FBA, AWD, and a new $1,400/year developer API fee), so profitability is harder, but the seller pool is shrinking.

Two forces are pulling sellers away. Guru marketing has swung heavily toward TikTok Shop, which is drawing new sellers there instead. And OpenAI and Google’s commerce protocols are trying to move product transactions inside their own AI answer engines, which will eventually pull some purchase intent out of Amazon entirely.

Amazon’s response so far is defensive. They blocked most AI crawlers from indexing Amazon and are pushing Rufus (their own AI shopping assistant) to keep the shopping conversation on-platform. That is not a growth strategy for sellers; it is a moat strategy for Amazon.

The developer-fee move is the more worrying signal. Tools like Helium 10 and Jungle Scout are what taught the last generation of sellers how to research products, and choking those partners with API fees slows the entire ecosystem’s ability to innovate on top of Amazon.

How Leo launched a new Amazon product with only creators and PPC (no giveaways)

Leo’s Q4 game launch used only creator content and Amazon PPC, no giveaways, and sold out 5,000 units before Christmas. The strategy leans on 3 factors that stack in Q4: existing category demand for gifts, external traffic from TikTok and Instagram creators, and semi-broad Amazon PPC campaigns that catch the resulting search volume.

Why no giveaways: giveaways used to be a ranking hack, but Amazon’s algorithm now weights genuine relevance and external traffic more than the pure velocity that a giveaway produced. Creator content produces both external clicks and genuine reviews, which is what the algorithm actually rewards.

For non-seasonal products, the same 3-factor stack works but the “demand tailwind” comes from you: paid meta plus creator content plus AI-generated variations rather than Q4 gift-buying pulling the market up.

Frequently asked questions

What is an AI automation stack for ecommerce?

An AI automation stack for ecommerce is the set of AI-powered tools and workflows a brand chains together to run marketing, creator outreach, ad production, and fulfillment with a smaller team. The current 2026 core is scored influencer outreach (Spliced.io or similar), an n8n-based meta-ad-cloning workflow (Gemini + Suno + NanoBanana + VO3 + FFmpeg), and a meta-to-landing-page-to-Amazon funnel.

How do I do influencer outreach on TikTok Shop without getting ignored?

Score every prospective creator on multiple data points (GMV band, active brand deals, audience match, per-video earnings, timing) before you send an invite, and only message creators whose economics will actually accept your offer. Generic mass-DM outreach is dead because creators receive too many templated pitches; targeted, scored outreach is what still works.

Can I clone a Facebook ad with AI and drop in my own product?

Yes, using an n8n workflow that chains Gemini (analyzes the source video), Suno (music), NanoBanana (frame rendering with your product), VO3 (8-second video generation), and FFmpeg (stitching). Leo Sgovio built and demoed this exact workflow at Sellers Summit and gives it away free on LinkedIn.

Are AI-generated ads still allowed on TikTok Shop and Meta?

TikTok Shop banned AI clones from making product claims in late 2025, including your own AI clone endorsing a product you have used. Informational AI content is still allowed. Meta allows AI ad creative with an AI disclosure flag; ad-cloning workflows that use no avatar (only product plus animation) currently remain within policy.

Should I send meta ads directly to my Amazon listing?

No. Send meta traffic to a warming landing page first, then to Amazon, because the landing page reads as an advertorial that pre-sells the buyer so they check out on Amazon instead of getting poached by a cheaper sponsored competitor on the detail page. Google Ads clicks also cost roughly one-third as much when they land on a landing page instead of an Amazon listing.

Is TikTok Shop or Amazon better for a new product launch?

Launch on both at the same time. TikTok Shop drives discovery and creator-led social proof; Amazon captures the actual purchase because most buyers still check out there. One of Leo’s competitors sold 70,000 units in Q4 running this dual-launch playbook.

Is Amazon still worth it as an ecommerce channel in 2026?

Yes for operators willing to prioritize profitability. Amazon reported its lowest new-seller count this year, which means less competition to rank, but fees are climbing across the board (advertising, FBA, AWD, and a new $1,400/year developer API fee). Amazon plus a discovery channel (TikTok Shop, meta, or Google) still outperforms most other single-channel setups.

Ready To Get Serious About Starting An Online Business?

If you are really considering starting your own online business, then you have to check out my free mini course on How To Create A Niche Online Store In 5 Easy Steps.

In this 6 day mini course, I reveal the steps that my wife and I took to earn 100 thousand dollars in the span of just a year. Best of all, it's absolutely free!

Podcast: Download (Duration: 38:55 — 44.8MB)

To optimize product listings for AI shopping in 2026, you need to rewrite your product copy around the emotional intent and specific use cases in your buyer’s mind, mine your reviews for the exact phrases they use, and spell out every attribute an agent would need to make a decision. On this episode of the My Wife Quit Her Job podcast, my co-host Toni Herrbach and I dug into what Amazon’s decision to block Perplexity from indexing its product catalog means for every seller with a store online.

The short version: agentic shopping is coming faster than most sellers think, and the sparse, keyword-stuffed listings that ranked in Google circa 2015 are now invisible to the AI agents doing the buying. If your description reads “100% cotton, 6 oz, machine washable” and stops there, an AI shopping agent has nothing to recommend you for except a spec query.

Below is the exact playbook Toni and I walked through: why Amazon just picked a fight with Perplexity, why agentic commerce breaks old copy, and the six-part product-listing rewrite that gets your products into AI shopping answers.

Key takeaways

• Amazon has blocked LLMs from indexing its product catalog and just won a court case forcing Perplexity to delete every scraped Amazon listing from its database. Amazon products are effectively invisible in AI shopping today.
• Agentic commerce (Anthropic’s Claude, OpenAI’s ChatGPT agents, and specialized shopping agents like OpenClaw) is already assembling shopping lists, comparing components, and in some cases checking out on behalf of users. This is a two-year, not a ten-year, shift.
• Old product copy fails because AI agents cannot infer a benefit from a bullet list of specs. Winning listings spell out the buyer’s problem, the use case, the fit for specific customer types, and the emotional payoff.
• The biggest information source for AI-friendly copy is your own reviews. Real ecommerce brands (True Classic Tees is the cited example) already pull emotional phrases like “I feel better looking in the mirror” straight out of customer reviews into their listings.
• Brand storytelling shifts, though it does not disappear. The “founder journey” section matters less to an agent; storytelling embedded in the product copy (who this is for, what problem it solves, when it fails) matters much more.

Why did Amazon block Perplexity and every other AI shopping agent?

Amazon blocked Perplexity and other AI shopping agents to protect its advertising business, which is now its highest-margin revenue stream and its main growth engine. Amazon’s core retail business has been growing slowly for several quarters, and its ad revenue is the piece Wall Street rewards. If shoppers move to AI agents that skip Amazon’s ad-heavy search results, that revenue stream collapses.

The Perplexity court case is likely the first of many. Perplexity now has to delete all scraped Amazon product listings from its database, and every other LLM shopping agent (ChatGPT, Claude, Gemini, and specialized tools like OpenClaw) is probably next. Amazon is drawing a hard line: shop inside Amazon.com, use Rufus (Amazon’s own AI assistant), or do not see Amazon products at all.

This decision is a bet that Rufus will grow fast enough to replace lost AI-agent discovery. Rufus is buried behind a small icon inside the Amazon app and site, and most shoppers do not know it exists. In the meantime, Walmart and other retailers who stay open to LLMs are getting a discovery advantage that Amazon has voluntarily walked away from.

What is agentic commerce and how fast is it coming?

Agentic commerce is any shopping flow where an AI agent researches, compares, and in some cases purchases products on behalf of a human. It is arriving now in phases: research and shopping-list assembly today, direct product recommendations with clickable links today on platforms like ChatGPT and Perplexity, and full agent-driven checkout emerging on OpenAI’s platform as of late 2025.

Real examples I have already seen:

• A friend using OpenClaw to build the parts list for a custom PC. The agent researched components, checked prices across retailers, and returned a shopping list. Amazon products did not appear.
• My own Claude session asking for the best router for the money. It returned a specific recommendation from a non-Amazon retailer.
• My wife using an AI styling service that ingests photos of her wardrobe, then assembles 10-piece capsule outfits she can approve or reject. Products that are not in the agent’s data set never make the cut.

The pattern is the same across every example. If the agent cannot see enough information to compare your product against alternatives, it cannot recommend you, and the shopper never learns you exist. This is why product-listing optimization for AI is now urgent for anyone selling anything more considered than a commodity.

Why do old-school product listings fail with AI agents?

Old-school product listings fail with AI agents because agents need explicit information to make a decision, and most ecommerce copy assumes a human eyeball will fill in the blanks. A human sees a nice hero image, reads “100% cotton t-shirt,” and buys on gut.

An AI agent has no gut. It has only the text on the page, the reviews (if it can read them), and the structured attributes in the schema.

A typical Shopify listing today looks like this: bullet list of specs (fabric, weight, dimensions), one paragraph of marketing fluff, care instructions. There is no clue about who the shirt is for, when it fits best, what problem it solves, or which reviewer profiles love it.

The result is that when an agent gets a query like “find me a t-shirt that fits well on tall lanky guys and looks less baggy than a Hanes,” the sparse listing has nothing to match on. A competing listing that explicitly says “designed for tall guys, stretches across the shoulders, tapers at the waist to eliminate the sloppy look” wins by default.

How do you rewrite product copy for AI shopping in 2026?

You rewrite product copy for AI shopping in 2026 by treating every listing as an answer to the specific search prompts your buyer would type into ChatGPT or Claude. That means front-loading the buyer’s problem, the customer profile the product is designed for, the use cases it solves, the reasons someone rejects it, and the emotional payoff people report in reviews.

The rewrite has six parts:

• Problem statement. One or two sentences naming the specific problem the product solves (“Regular t-shirts hang like sacks on tall guys with narrow shoulders”).
• Ideal customer profile. Who this is designed for, including body type, use case, skill level, family situation, or any other filter an agent would need.
• Use-case scenarios. Three to six concrete situations where the product shines. “Great for weekend errands,” “wears well under a blazer for hybrid office days.”
• Emotional payoff, pulled from reviews. The literal phrases customers use in reviews. “I feel better looking in the mirror,” “my grandmother’s apple pie every Thanksgiving.”
• Explicit specs and dimensions. Every attribute an agent would need to compare: measurements, weight, materials, compatibility, warranty. Do not assume the image conveys size.
• Who this is NOT for. One or two sentences filtering the wrong buyer out. Agents use exclusion signals to decide when NOT to recommend a product.

The True Classic Tees listing is a textbook example. Its Amazon description explicitly names the tall, athletic-build buyer, promises the “less sloppy look,” and reflects reviewer language about feeling better looking in the mirror.

That is not accidental. Someone at True Classic is pulling that copy directly out of the review corpus.

Why are customer reviews the biggest input for AI-friendly copy?

Customer reviews are the biggest input for AI-friendly product copy because they are the only place your buyer describes the product in their own words, using the exact vocabulary an AI agent will match on. Your marketing team writes “premium construction.” Your buyer writes “I do not have to iron this shirt before a meeting.” Those are the phrases that hit the agent’s semantic search.

There is a whole category of new tools built to scrape a store’s own reviews (or a competitor’s reviews on Amazon, where scraping is much harder), cluster the emotional and functional themes, and feed them into product copy and paid ad creative. I mentioned an ads company I am working with on my direct-to-consumer store; their entire workflow is review-driven copy generation.

Practical version you can do this week:

• Export your last 500 reviews (Shopify, Amazon Seller Central, or a review app like Judge.me).
• Paste them into ChatGPT with a prompt: “Cluster these reviews by the top five themes buyers mention. For each theme, quote the three most representative sentences verbatim.”
• Rewrite each product listing to work in the top three themes and at least one verbatim quote (paraphrased into your voice) per section.

That single exercise moves most stores from AI-invisible to AI-considerable.

How does agentic commerce change branding and brand loyalty?

Agentic commerce weakens brand storytelling on a homepage while strengthening brand loyalty at the product level, because agents ignore founder stories but shoppers still form attachments to products they discover through an agent. The “why we started this company” hero section becomes background music. The specific product experience, the packaging, the customer service, and the post-purchase moment become the loyalty engine.

Toni made the point cleanly with Yeti, Stanley, and Owala. Nobody researches a cooler; they buy the Yeti, and nobody researches a tumbler; they buy the Stanley or the Owala.

That brand loyalty short-circuits the AI research step entirely, because the buyer tells the agent what to buy rather than asking for a recommendation.

The play for a new brand is to earn one of those short-circuits in a specific category. Do it by shipping a product experience so good that reviews naturally include the brand name and the emotional payoff. Once that language shows up in enough places (your reviews, TikTok, Reddit, YouTube), agents start recommending you by name.

What kinds of purchases will AI agents dominate, and which will they not?

AI agents will dominate commodity and considered-utility purchases (groceries, household staples, electronics, tools, appliances) and will make almost no dent in high-touch luxury, in-store retail, and identity-driven purchases. The line runs along how much emotional experience is bundled into the purchase.

The AI-dominant column:

• Groceries and household consumables (Walmart’s aggressive Subscribe & Save push is already pointed here).
• Standard electronics and networking gear.
• Home appliances, tools, replacement parts.
• Basic apparel and accessories.
• Books, media, office supplies.

The AI-resistant column:

• Luxury goods where the appointment, the fitting, and the store experience are the product (Hermes, high-end watches).
• Vehicles, especially through personality-driven sellers. The “Baddie in a Benz” TikTok example is exactly this: buyers fly to Georgia to buy a Mercedes from a specific salesman.
• Luxury travel where a human agent handles the crisis (the “call Betty when you are stranded” service model).
• Anything highly emotional (weddings, kids, gifts) where storytelling and shopper agency matter.
• In-store retail, which is actually growing year over year as shoppers crave real-life experience.

Interesting data point: recent retail analyses suggest the top 10% of US consumers now drive roughly half of all retail spending. Your customer mix determines how much of the AI-agent shift actually affects your business.

Should you optimize for Amazon Rufus or ignore it?

You should still optimize your Amazon listings for Rufus because Amazon is betting the entire agentic-shopping outcome on it, and if it works, sellers who did the copy work will win the first wave of Rufus-driven traffic. The same six-part rewrite that works for external LLMs works for Rufus: problem, customer profile, use cases, review-derived language, explicit specs, and negative filters.

Whether Rufus actually beats third-party agents is the open question. Amazon has structural reasons to bias Rufus recommendations toward high-margin products, sponsored listings, or its own private labels. Skeptical shoppers (especially anyone who has watched Amazon over the years) will either supplement with an external agent or bypass Amazon entirely.

The pragmatic move is to rewrite both channels. Your own website has to be AI-agent friendly for external LLMs, and your Amazon listings have to be Rufus-friendly for on-platform search.

Doing one and skipping the other leaves half your discovery surface unlit.

What should you do this week if you sell products online?

The most important thing you can do this week is audit your top 10 SKUs against a simple test: paste your product page into ChatGPT or Claude, ask “who is this product for, what problem does it solve, and when would you NOT recommend it?” and see what comes back. If the answers are vague or wrong, your listing is not ready for AI shopping.

Fixes ranked by leverage:

• Rewrite the description on your top 10 SKUs using the six-part structure above (problem, ICP, use cases, review language, specs, negative filters).
• Add structured data (JSON-LD for Product with brand, category, size, material, review count, rating). Agents extract these fields with high confidence.
• Make sure your product content is in the raw HTML, not client-side rendered. Most AI crawlers do not execute JavaScript.
• Publish comparison content (“X vs Y” pages) so agents can cite you as a source rather than only as a product.
• Encourage detailed reviews. Longer, more specific reviews improve both human conversion and agent-visible signal.

The stores that get this right in the next 12 months will be the ones agents recommend by default for years. The stores that wait will be invisible.

Frequently asked questions

Will AI agents actually start buying products on behalf of shoppers?

Yes, though the first wave is agent-assembled shopping lists and product recommendations rather than agent-triggered checkout. OpenAI has already announced direct checkout inside ChatGPT for select partners, and specialized agents like OpenClaw are experimenting with full checkout flows. Full agent-driven buying is a matter of quarters, not years.

Is Amazon blocking AI agents a permanent policy?

Amazon’s block on AI agents is likely to hold as long as ad revenue is Amazon’s dominant profit line, which means for the foreseeable future. Amazon may eventually license API access to select agents for a fee, similar to how it monetizes seller access, but free scraping is done. Sellers should plan around Amazon being an isolated channel.

Do product listings need to be different on Amazon vs my own website?

The core copy structure (problem, ICP, use cases, review language, specs, negative filters) is the same on Amazon and your website, but you have to write for two different agents. Your website copy needs to work for external LLMs like ChatGPT and Claude, while your Amazon copy needs to work for Rufus. Duplicate the structure, tune the details for each channel.

How do I get review language into my product listings legally and safely?

You can paraphrase and adapt themes from your own reviews freely because customers granted a license when they submitted the review, and quoting a customer with attribution is standard marketing practice. You cannot copy competitor reviews verbatim from Amazon or other platforms. Stick to your own review corpus and any reviews on your site.

Will branding still matter in an AI-shopping world?

Branding will still matter, though it moves from homepage storytelling to product-level trust and post-purchase experience. Buyers who love a Yeti do not research coolers; they tell the agent “buy a Yeti.” Earning that brand loyalty in a specific category is now the moat, because it lets you skip the agent’s comparison step entirely.

What is the fastest low-cost first step to make my listings AI-ready?

The fastest low-cost first step is to export your last 500 reviews, cluster them in ChatGPT for the top five themes buyers mention, and rewrite your top 10 SKU descriptions to work in those themes with the buyer’s own language. That single exercise moves most stores from AI-invisible to AI-considerable in a week.

Ready To Get Serious About Starting An Online Business?

If you are really considering starting your own online business, then you have to check out my free mini course on How To Create A Niche Online Store In 5 Easy Steps.

In this 6 day mini course, I reveal the steps that my wife and I took to earn 100 thousand dollars in the span of just a year. Best of all, it's absolutely free!

Podcast: Download (Duration: 47:45 — 54.9MB)

The Meta ads change that is quietly 10x-ing returns for Shopify stores right now is to swap short punchy ad copy for a long-form story ad built on a five-part master story: a category-of-one headline, an origin story, a differentiation stack, a testimonial stack, and a risk-reversing offer. On this episode of the My Wife Quit Her Job podcast, I sat down with Scott Cunningham of Merchant Mastery, who runs both a Shopify agency (Socialite) and a coaching community, and who has been quietly running this playbook since he adapted Eugene Schwartz’s 1966 book Breakthrough Advertising for the Meta ads era.

The specific tactic that made me record this episode is Scott’s long-form Meta ad. Ten of his community members in the past week alone reported their best-ever ad performance running the exact same five-part written story as a Facebook ad, paired with a still image or a founder video. The playbook is a wholesale rejection of the short “scroll-stopping” ad copy Meta agencies have preached for the last five years.

Below is the exact framework Scott walked me through: the four desire buckets that produce a category-of-one headline, the five parts of a master story, the ad structure that works after Meta’s September Andromeda update killed interest targeting, and the way he validates the whole system on $60 a day.

Key takeaways

• Meta’s September 2025 Andromeda update killed interest audiences. All targeting now happens through the creative and copy, which is exactly why long-form story ads suddenly out-perform short punchy ones: they give Meta more signal about who to serve the ad to.
• Scott’s five-part “master story” ad framework is: category-of-one headline, five-part origin story, differentiation stack, testimonial stack, and offer. All five parts run inside a single long Facebook ad.
• Every strong headline starts by identifying the buyer’s most prominent desire, drawn from four buckets: mass instinct (permanent), innovation problem (permanent), trend (changing), and mass education (changing). The framework comes from Eugene Schwartz’s 1966 book Breakthrough Advertising.
• Every Shopify seller has an origin story, even the ones who insist they don’t. Scott’s client at American Vinegar Works swore he had no story, then revealed he found an 1800s fermentation method in a French library and rebuilt it with the University of Massachusetts.
• The launch structure is a $60/day break-even test with two ad copies and three creatives each (owner video, UGC, static), all in a single ad set post-Andromeda, with organic social layered on only after ads validate the message.

What is Scott Cunningham’s Story Selling Formula for Meta ads?

The Story Selling Formula is a five-part narrative structure that runs as a single long-form Meta ad and gives Facebook enough signal to auto-target the right buyers post-Andromeda. The five parts are: a category-of-one headline that names the buyer’s top desire, a founder origin story, a head-to-head differentiation stack against alternatives, a testimonial stack that answers common objections, and an offer that removes purchase risk.

Scott built the formula by combining Eugene Schwartz’s 1966 direct-response classic Breakthrough Advertising with what he sees working across the Shopify brands his Socialite agency runs ads for. It is deliberately old-school. Long, direct, benefit-first, and structured to hold attention rather than shock it.

The reason it works in 2026 has less to do with copywriting theory and more to do with Meta’s algorithm change. Interest audiences are gone, and the ad copy itself is now the targeting layer.

A 400-word story ad tells Meta exactly who this product is for, what problem it solves, and which readers stop and read; a five-word “Buy Now!” ad tells Meta almost nothing.

Why did Meta’s Andromeda update kill short-form ads and reward long-form?

Meta’s Andromeda update rolled out in September 2025 and removed the ability to target by interest audiences, which forced the algorithm to infer audience fit from the ad creative and copy instead. Short punchy ads no longer give the algorithm enough information to find your buyer, so campaigns that used to run on tight interest stacks now underperform.

Long-form ads give Meta a rich text signal about the buyer profile, the problem, and the product category. The algorithm reads that signal, matches it to on-platform behavior, and delivers the ad to lookalikes of the users who actually read to the end. The ad copy is now doing double duty as both persuasion and targeting.

The new post-Andromeda structure Scott’s agency uses is a single ad set that consolidates all creatives (no more separate interest / open / lookalike ad sets), driven by a rich long-form primary text with three creative variants per copy variant. This is a significant departure from the “one ad set per audience” playbook that dominated 2020 to 2024.

How do you find the “most prominent desire” that anchors your ad?

The most prominent desire is the single buyer motivation with the highest intensity, staying power, and audience scope, chosen from four brainstormed buckets: mass instincts (permanent human drives like family, wealth, health), innovation problems (unsolved specific problems like eczema or migraines), trends (short-lived opportunities like COVID face masks or K-pop merchandise), and mass education (the current level of consumer awareness in your category).

The four-bucket exercise for Scott’s Palmerus Puzzles client (double-sided wooden puzzles handmade in Calgary) played out like this:

• Mass instinct: family connection, hobby, shared memories, screen-free time.
• Innovation problem: traditional puzzles lose pieces, get soggy, collect dust after one build, and lose novelty.
• Trend: renewed interest in tactile hobbies as an antidote to screen time.
• Mass education: growing awareness that puzzles improve memory formation and reduce screen fatigue.

Once the list is built, you grade every desire on intensity (how urgent is it?), staying power (will people keep buying to satisfy it?), and scope (what share of your customers share it?). The highest-scoring desire becomes the anchor for your category-of-one headline.

What is a category-of-one headline, and how do you test it?

A category-of-one headline is an opening line for your ad that positions your product as the best possible answer to the buyer’s most prominent desire, in language a competitor could never plausibly copy. Scott’s test is simple: if any competitor could say the same thing, the headline is not category-of-one and needs to be rewritten.

Weak headline: “Premium wooden puzzles for adults.” Any puzzle brand can claim it.

Category-of-one headline: “The double-sided poplar-wood puzzle designed to hang on your wall as artwork when you’re done.” Only Palmerus can plausibly say that.

The headline sits at the top of the primary text of the Facebook ad, above the origin story. It is the piece of copy Meta serves first, and it decides whether a scrolling user stops long enough to read the story. Everything else in the master story is downstream of getting this line right.

How do you write your origin story if you think you don’t have one?

Every Shopify seller with product validation has an origin story; they usually just do not recognize it as remarkable, because they are too close to their own decision to sell. Scott’s rule is that something led you to this specific product, and documenting that something is the origin story. His five-part origin-story framework is: struggle, journey, spider bite (the moment of revelation), revelation, impact.

The American Vinegar Works example illustrates how invisible a great origin story can feel to its owner. When Scott’s team asked Rodrigo, “What led you to selling vinegar?”, he shrugged and said grocery vinegar felt bland. Then out came the full story: while traveling in France he found an 1800s fermentation manual in a library, consulted the University of Massachusetts to reverse-engineer the process, and built a facility to ferment vinegar in oak barrels for nine days a batch.

Rodrigo genuinely did not think that was remarkable. His Meta ads were running at a 0.5 ROAS before Scott’s team added the origin story. Once the story appeared in the primary text, the ads became profitable, because the story is exactly the signal a buyer needs to prefer a $20 bottle of vinegar over a $4 grocery bottle.

What are the five parts of Scott’s master story ad?

The five parts of Scott’s master story ad are the sequence a single long-form Facebook ad walks a buyer through, top to bottom, inside the primary text field. Each part has a specific job and a specific length target.

• 1. Category-of-one headline. One or two sentences. Names the buyer’s top desire and positions your product as the uniquely best answer.
• 2. Origin story. Two to four short paragraphs. Uses the five-part struggle / journey / spider bite / revelation / impact structure so the reader trusts you did the work they would not do themselves.
• 3. Differentiation stack. A short list or two of “they do X, we do Y” comparisons across features, materials, sourcing, or customer support. This is the section that fends off direct competitors.
• 4. Testimonial stack. Two or three real customer quotes that each answer a common objection (price, trust, quality, longevity). Objection handling by proof, not by argument.
• 5. Offer. One to two sentences. Reduces purchase risk (guarantee, free shipping, no-questions returns) or adds an incentive (bundle, first-order discount, gift with purchase). Ends with a specific call to action.

Total length is typically 300 to 500 words inside the primary text. Yes, that is longer than what Facebook has told agencies to write for years. That length is the point.

What ad campaign structure works post-Andromeda?

The post-Andromeda ad campaign structure Scott’s agency runs is a single CBO campaign with a single ad set, containing two long-form copy variants tested against three creative variants each, for a total of six ads on the “ice” at once. The old three-ad-set split (interest / open / lookalike) is gone; everything consolidates into one ad set because interest targeting is gone.

Scott’s hockey-analogy launch plan:

• Build a bench of 5 ads. Draft five different long-form story ads based on the master story exercise, each with a different angle on the primary desire.
• Put 2 on the ice. Pick the two you would bet the business on and launch those first.
• Pair each copy with 3 creatives: a founder video, a UGC video, and a static image or infographic. That is 6 ads live simultaneously.
• Start at $60/day. Goal is to break even in the first week. Break-even validates the message and lets you scale.
• Scale after 10K/month. Once you’re consistently hitting five-figure monthly revenue from ads, layer on organic social and email retention.

The scale-up rule matters. Organic social is a great retention and vetting channel once ads have proven the message, and it is a very slow validation channel if you use it first. Scott’s rule of thumb is that pure-organic Shopify growth typically takes four years of daily posting to hit meaningful revenue, versus roughly 30 to 90 days on paid.

Can beginners use AI to write the master story?

Beginners can use AI to draft the master story if they use a custom GPT trained specifically on the framework rather than a blank ChatGPT prompt. Scott built a proprietary tool called Story Selling AI (available at merchantmastery.io) that interviews the merchant through the four desire buckets, the origin story, the differentiation stack, the testimonials, and the offer, then produces drafts of each section using 40+ pages of internal templates and examples.

The critical rule is that AI cannot invent your category-of-one headline. Generic ChatGPT pulls language from the entire public web, which by definition is what your competitors already say; a true “breakthrough” headline in Schwartz’s sense has to say something no one else has said. AI is a brainstorming partner for the raw material and a first-draft writer for the connective tissue, and you supply the specifics of your product, story, and customer.

Scott’s practical workflow is to answer the tool’s interview questions with your own words, review the draft output, and then edit it into a voice that reads like you. Most merchants who try to short-cut the interview step end up with copy that sounds like every other ecommerce brand.

What does the ad look like paired with creative?

The winning creative pairing for a long-form story ad is a still image or short founder video that stops the scroll while the long-form text holds attention. The creative’s only job is to buy the primary text a chance to be read. The text does the actual selling.

Scott’s three creative variants per copy:

• Founder video (from-the-owner). The founder on camera talking about the origin moment or the product’s key differentiator. Works because it visibly attaches a real person to the story in the copy.
• UGC video. A customer using or reacting to the product. Works as ambient testimonial support to the testimonial-stack section of the copy.
• Static image or infographic. Product on a clean background, or a simple comparison graphic. Works because it does not compete with the copy for attention; the reader’s eye lands on the text quickly.

Rotate which creative wins by testing all three against both copy variants for the first two weeks, then let Meta’s optimizer concentrate spend on the best-performing pairs.

When should a Shopify seller add SMS to the mix?

Shopify sellers should add SMS to their Klaviyo (or equivalent) flows as soon as they have a working ads engine and a growing email list, because SMS has higher open and click-through rates than email and doubles the recovery rate on abandoned checkouts. Scott’s agency Socialite is a Klaviyo Platinum Diamond partner, and their standard build sends an email first, then follows up with an SMS to anyone who did not open the email within a set window.

The list-building step matters more than the tool choice. Get SMS opt-in at checkout and in your welcome flow from day one; the compounding value of a bigger opted-in SMS list is much larger than the incremental performance difference between platforms.

Skip SMS entirely until you are past the validation stage. On $60/day of ads with no clear message-market fit, SMS adds cost and complexity without meaningful upside. Once you are hitting $10K to $50K a month and have a solid email flow, SMS is the highest-ROI add-on channel.

Frequently asked questions

Do long-form Meta ads still work if my product is a low-priced impulse buy?

Long-form story ads still work for lower-priced products, though the emphasis shifts. On sub-$30 items you can shorten the origin story and lean harder on the differentiation stack and testimonial stack, because the buyer decides fast. The core reason long-form beats short-form (Meta needs targeting signal from the copy post-Andromeda) applies regardless of price point.

How much should I spend testing a new Meta ad campaign?

Scott’s baseline test budget is $60 a day with a goal of breaking even within the first week, run against a single ad set with two copy variants and three creatives each. That is enough spend to give Meta learning signal without burning cash before you know whether the message works. Scale up only after you hit break-even at that budget.

Is the Story Selling Formula only for Shopify stores, or does it work for Amazon sellers too?

The framework works for any DTC ad channel, though the ad itself has to live outside Amazon, because Amazon does not run long-form primary-text ads. Amazon sellers use the same master story to rewrite their product listing bullets, A+ content, brand story page, and external Meta ads that funnel to Amazon. The underlying desire-and-story mechanics are the same.

Where does the “Breakthrough Advertising” book fit in, and do I need to read it?

Eugene Schwartz’s Breakthrough Advertising (1966) is the source material for the four-desire framework and much of Scott’s story architecture. The book is out of print and typically costs $200 to $1,000 secondhand, so most operators use derivative frameworks like Scott’s rather than the original. Reading it helps if you write a lot of copy, though the ROI on Scott’s Story Selling course or AI tool is faster for most merchants.

How long before I know whether my new story ad is working?

You should see a clear signal within the first 5 to 7 days at $60 a day of spend, which is roughly $300 to $420 total. If cost-per-purchase is trending toward your break-even number by day 7, the message is working and you can scale. If you are well above break-even by day 7 with no downward trend, kill the ad and rotate in a copy variant from your bench of 5.

Do I still need organic social if my ads are profitable?

You still need organic social as a trust-and-retention layer once your ads are working, because a decent share of ad-clicks vet your brand by clicking through to your Instagram or TikTok profile. An empty or stale social presence kills conversions from otherwise qualified traffic. It does not have to lead sales; it has to prove you are a real, active brand.

Ready To Get Serious About Starting An Online Business?

If you are really considering starting your own online business, then you have to check out my free mini course on How To Create A Niche Online Store In 5 Easy Steps.

In this 6 day mini course, I reveal the steps that my wife and I took to earn 100 thousand dollars in the span of just a year. Best of all, it's absolutely free!