Podcast: Download (Duration: 42:56 — 49.4MB)

Hyper-targeting customers for your online store got dramatically easier with the launch of the Klaviyo Data Platform, which turns Klaviyo from an email tool into a full ecommerce CRM. The new platform stores unlimited customer data, uses AI to auto-build segments, predicts which product each customer is most likely to buy next, and folds customer service, SMS, push notifications, and in-app messaging into one unified customer view. My co-host Toni Herrbach and I attended Klaviyo’s launch event and walked away convinced this is what “Salesforce for ecommerce” actually looks like.

If you run a Shopify (or any) store and you have been paying separate SaaS tools for email, SMS, customer service, on-site chat, and reporting, most of that stack is about to collapse into Klaviyo.

Below we break down every major feature Klaviyo announced, the specific use cases where the AI actually pays off, why the new customer-support hub matters more than most owners realize, and where the platform still has gaps.

Key takeaways

• Klaviyo launched the Klaviyo Data Platform, an ecommerce-focused CRM with unlimited customer-data storage and a native Shopify integration.
• AI Forecast automatically builds customer segments by predicting who is most likely to buy a specific product, without you having to write the segment rules by hand.
• The new campaign builder works like a flow. You can send a Mother’s Day promo, then auto-resend only to unopens who visited the site in the last 30 days and have spent $500+, all in one drag-and-drop campaign.
• The Action Center recommends the exact next campaign to send based on your data plus benchmarks from every other store in your vertical.
• Klaviyo now includes a self-service and support hub with AI chatbot, order tracking, and returns management. Customer service can now live inside the same tool as your marketing.
• A unified customer timeline shows every email open, SMS click, purchase, page view, return, and shipment status by real customer name, which Google Analytics 4 cannot do.
• The new open API lets non-Shopify stores integrate Klaviyo with any platform, and opens the door for a future third-party app ecosystem.

What is the new Klaviyo Data Platform?

The Klaviyo Data Platform is a full ecommerce CRM built on top of Klaviyo’s existing email and SMS infrastructure. It offers unlimited storage for every customer’s browsing history, purchases, purchase frequency, email opens, SMS clicks, and returns, all in one profile that pulls directly from Shopify and every other tool Klaviyo already integrates with.

The problem this solves is the one every ecommerce owner knows painfully well. Your customer data is scattered across Shopify, your email tool, your SMS tool, your help desk, your reviews app, and a Google Analytics dashboard that no one enjoys opening. Owners routinely pay contractors thousands of dollars to build custom dashboards that stitch it all together, and those dashboards usually break within six months.

Klaviyo told a customer story at the launch about exactly that scenario. The owner hired a developer, spent real money on a custom aggregation tool, and it did not work well enough to use. The Klaviyo Data Platform is Klaviyo’s answer to that problem for the whole market.

How Klaviyo AI Forecast auto-builds customer segments

Klaviyo AI Forecast automatically builds a customer segment when you describe the outcome you want, instead of forcing you to write the segment rules by hand. You can type something like “create a segment of everyone most likely to purchase a personalized apron in the next month” and Klaviyo assembles the list from every relevant behavioral signal it has stored.

The old workflow was rule-based. You had to know exactly which conditions to apply: bought an apron before, has not bought in six months, opened the last three campaigns.

AI Forecast collapses that into an outcome-based prompt. It also incorporates signals you never would have thought to check, like someone who browsed the apron category two weeks ago but never bought.

Toni already uses Klaviyo’s earlier AI segmentation tool heavily and says the accuracy is very high. The rare misses are almost always because the prompt was written poorly, not because the AI got the data wrong.

The other unlock is downstream. AI does not stop at building the segment; it also sequences the follow-up.

If a segment gets the first email, the AI can decide who inside that segment gets the second email based on whether they took specific actions, and adjust the email copy to reference those actions.

Why the new Klaviyo campaign builder replaces multi-email sales

The new campaign builder works like a Klaviyo flow, so a single “campaign” can now include the send, the resend-to-unopens, the SMS fallback, and the exclusion logic in one drag-and-drop canvas.

Before this, campaigns and flows were two different products. Flows had conditional logic. Campaigns were one-shot broadcasts.

The practical use case is any multi-touch promotion. If you run a three or four day sale, you probably send 4 to 6 separate emails: first day AM, first day PM, midway, last chance AM, last chance PM. Each one needs to exclude buyers, resend to unopens, and coordinate with an SMS blast.

You can now build that whole sequence as one campaign. Resends can be surgical: send only to people who did not open the first email but visited the site in the last 30 days and have spent $500 or more. That level of targeting used to require assembling three or four separate segments manually.

Klaviyo also finally added recent-unopens as a native resend option, which we and other Klaviyo power users have been asking for for years. Small feature, huge time-saver.

How the Klaviyo Action Center tells you what to send next

The Klaviyo Action Center reads your store’s data and gives you a specific campaign recommendation, then builds most of the campaign for you if you click accept. It compares your store’s behavior against Klaviyo’s full dataset of stores in your vertical, so the recommendation reflects what actually works for similar businesses, not just your own history.

An example recommendation might be “send a campaign to people who buy pink leggings, they are 3.4x more likely to buy the new pink dress.” Click it and Klaviyo pre-builds the segment, pre-populates the email template with your brand assets, and sits it in your drafts folder ready to review.

For anyone who has spent hours guessing which segment to target next or building the same campaign template from scratch every week, this is the highest-leverage feature Klaviyo has ever shipped. Testing still matters, but you are testing sharper starting points instead of guessing from a blank page.

Klaviyo’s self-service and support hub turns customer service into a sales channel

Klaviyo’s new self-service and support hub adds an AI chatbot, order tracking, returns management, and live chat handoff to a real agent, all natively inside the same platform that holds your customer profiles. This matters because Klaviyo already knows everything about your customer: addresses, past orders, browsed pages, email engagement, SMS behavior.

Roughly the top question every ecommerce store gets is “where is my order?” That can be answered by AI 95 percent of the time. Once AI handles the routine tickets, your human customer-service reps can focus on the higher-value conversations that actually convert.

At my store, if we can get a customer into a live conversation with our support rep, our close rate on that conversation is around 90 percent. That is the shift most owners miss: customer service is a sales channel, not a rescue boat. Every question about personalization, sizing, delivery timing, or product compatibility is a buying signal.

Having all of that in one platform also means the human agent, when they get the escalation, already sees the customer’s purchase history, browsing history, engagement level, and predicted next purchase. They can suggest the matching apron to the customer who just bought the personalized pillowcase, in the same chat window, without switching tools.

The unified customer view: better than Google Analytics for ecommerce

Klaviyo’s unified customer view shows every interaction a specific named customer has had with your store on one timeline: emails opened, SMS clicked, pages viewed, products purchased, returns filed, shipments in transit. Google Analytics 4 cannot do this because Google prohibits attaching personally identifiable information to analytics profiles.

That single privacy restriction is why so few store owners actually use GA4 for customer-level decisions. It shows you aggregate cohorts, not “what has Sarah done in the last 90 days.” Klaviyo, because it owns the customer profile natively, can show you Sarah’s entire history by name.

For any store with high-consideration or high-lifetime-value customers (custom, personalized, subscription, luxury, B2B), that shift is huge. You stop making decisions off aggregate reports and start making them off individual customer timelines that your CS reps can also see.

Why the Klaviyo open API is a big deal for non-Shopify stores

Klaviyo’s new open API lets you connect Klaviyo to any ecommerce platform via code, so stores on WooCommerce, BigCommerce, custom-built stacks, or (unfortunately) GoDaddy can now get the same integrated experience Shopify stores get natively. Previously, deep Klaviyo integration was effectively a Shopify-first feature.

The API also opens the door for a Klaviyo app ecosystem. Third-party developers can build tools that plug into Klaviyo data, similar to the Shopify app store. That community layer often ends up being more valuable than the core product itself over time.

If you are on a non-Shopify platform, this is the single feature that unlocks the whole rest of the platform for you. The rest of the announcements do not matter if your data cannot get into Klaviyo cleanly.

How Klaviyo AI helps subscription-based stores retain customers

Klaviyo’s new AI features analyze subscription customer behavior to identify when subscribers are about to churn and what messaging keeps them longer. The launch event referenced a 54 percent boost in retention for stores implementing the recommended AI-driven communication strategies.

For subscribe-and-save models, the killer question is always “at what point do customers start to fall off?” Different products, different price points, and different customer types churn at different points on the timeline. Historically you had to build that analysis yourself by hand or hire a data analyst.

Klaviyo’s AI now surfaces those churn patterns automatically and recommends the winback message, discount, or product-swap most likely to save the subscription. If a meaningful portion of your revenue is subscription, this alone is probably worth the platform.

Comparison: Klaviyo Data Platform vs the tools it replaces

Common personalization mistakes Klaviyo AI can prevent

Even with good tools, most stores make the same personalization mistakes because a human forgot to add an exclusion or a suppression rule. The most common ones are all fixable now with AI-driven automation.

The classic bad experience: you buy a pair of shoes and get an email about women’s clothing the next day, even though your account clearly shows you bought men’s shoes. Or you buy a product on Monday and the same product goes on sale in an email on Tuesday for 25 percent off. Or, in Toni’s actual example, you buy a bunch of paint at Lowe’s and the follow-up email pushes more paint, not painter’s tape and brushes.

Every one of those is a segment rule that a human should have set but did not. Klaviyo’s AI segmentation and suppression logic can add those exclusions automatically: exclude anyone who purchased that exact product in the last 48 hours, exclude anyone who just paid full price from the discount send, upsell complementary categories on any post-purchase flow.

The upside is not just fewer angry customers. It is that the inbox becomes genuinely useful to the customer.

Emails that reference the color you actually wear, the size you actually buy, and the products you have actually looked at get opened. Everything else gets deleted or marked as spam.

Frequently asked questions

What is the Klaviyo Data Platform?

The Klaviyo Data Platform is Klaviyo’s new ecommerce-focused CRM. It stores unlimited data on every customer (purchases, browsing, engagement across email and SMS), integrates natively with Shopify and other platforms via a new open API, and uses AI to auto-build segments, recommend campaigns, and predict what each customer will buy next.

How does Klaviyo AI Forecast work?

Klaviyo AI Forecast lets you describe the customer segment you want in plain language (“everyone most likely to buy a personalized apron in the next month”) and builds the segment automatically from your store’s behavioral data. It incorporates signals like browse history and past purchases that you would have to combine manually with the old rule-based segment builder.

Does Klaviyo now include customer service?

Yes. Klaviyo’s new self-service and support hub includes an AI chatbot, order tracking, returns management, and live chat handoff to a human agent. Because it lives inside Klaviyo, the human agent sees the customer’s full purchase and engagement history in the same window when they take over from the AI.

Can I use Klaviyo Data Platform if I am not on Shopify?

Yes. Klaviyo now offers an open API that lets you connect any ecommerce platform (WooCommerce, BigCommerce, custom stacks, and others) to Klaviyo with code. Previously deep integration was effectively Shopify-only, so non-Shopify stores can now get the same unified customer view.

How is Klaviyo different from Google Analytics 4 for ecommerce?

Klaviyo shows a full customer timeline by real name, including every email open, SMS click, page view, purchase, return, and shipment. Google Analytics 4 cannot show you named-customer timelines because of privacy restrictions. For any store making decisions at the individual customer level (custom, personalized, subscription, or high-LTV), Klaviyo’s unified customer view is far more useful.

Is the new campaign builder different from Klaviyo flows?

The new campaign builder brings flow-style conditional logic into one-shot campaigns. You can build a multi-touch promotion (initial send, resend to unopens with specific behavior, SMS fallback, exclusion rules) as one campaign instead of stitching a flow and several campaigns together. Flows still exist for automated ongoing journeys like abandoned cart and welcome sequences.

Will Klaviyo AI features increase my email revenue?

Anecdotal case data from Klaviyo’s launch cited significant lifts, including a 54 percent boost in retention for subscription businesses implementing the recommended AI-driven strategies. The mechanism is straightforward: better segmentation, sharper personalization, and fewer irrelevant sends. Actual lift depends on how sloppy your current segmentation is and how aggressively you adopt the AI recommendations.

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: 42:26 — 48.9MB)

The Sellers Summit 2025 agenda is a curriculum-based, tactical ecommerce lineup running May 6 to May 8 in Fort Lauderdale, capped at roughly 200 attendees, and every session is taught by an operator who is currently running the business they are teaching from. This year’s agenda covers sourcing outside China (with dedicated sessions on Made in USA, Mexico, and Vietnam), Meta ads that survive algorithm turmoil, TikTok Shop from the top seller we know personally, community-building for physical products, YouTube plus email as a growth engine, shoppable YouTube ads, ecommerce SEO in the AI era, and Klaviyo’s newest AI CRM features. My co-host Toni Herrbach and I broke down the full lineup on the show.

If you are trying to decide whether an ecommerce conference is worth the ticket, this post lays out what the agenda actually covers, why the format is different from typical Amazon-only events, and the specific sessions and roundtables that are usually rated highest by past attendees.

Below we cover the two masterminds, every confirmed 2025 speaker session, the roundtable format that most attendees say pays for the ticket by itself, and the practical selection criteria for picking an ecommerce event in general.

Key takeaways

• Sellers Summit 2025 is May 6 to May 8 in Fort Lauderdale, capped at ~200 attendees, ticket prices rise every two weeks until the event.
• Two masterminds run on Day 1: the ecommerce mastermind (revenue floor: $250K+ or $1M+) and the content mastermind (open to any revenue level, focused on YouTube, TikTok, email).
• Confirmed sessions this year include Made in USA sourcing (Annette), Mexico and Vietnam sourcing (Jim Kenimer / Cosmos Sourcing), Meta ads pivots (Andrea), shoppable YouTube ads (Brett Curry), TikTok Shop (Tiffany), community building (Sally Wilson / Caterpillar Cross Stitch), YouTube plus email (Toni Herrbach), building an audience beyond Amazon (Chris Shaffer), AI for Amazon (Ritu), and ecommerce SEO in 2025 (Jeff Oxford).
• Klaviyo, Quiet Light, RPC (freight forwarding with Pam), and attorney Steve Weigler run one-on-one consult sessions at the roundtables. Most attendees say a single 30-minute consult pays for their ticket.
• Every speaker also hosts a roundtable so attendees can go deep on questions the mainstage did not have time for. Speaker roundtables are the most-cited reason people return.
• Focus of the event is diversifying off Amazon-only or off China-only, with real playbooks for building direct-to-consumer brand equity.

What is Sellers Summit and who is it for?

Sellers Summit is an intimate, curriculum-based ecommerce conference for physical-product sellers, capped at ~200 attendees, running its eighth year in Fort Lauderdale, Florida from May 6 to May 8, 2025. Unlike most ecommerce events that focus on inspirational stories or 50,000-foot advice, every speaker is deep in the trenches of an active ecommerce business.

It is not invite-only, despite persistent rumors otherwise. Anyone can buy a ticket. Most attendees already have a brand and are actively selling, but complete beginners are welcome and every year several attendees show up pre-launch with a product idea and use the mastermind to shape it.

Sellers Summit is one of the few ecommerce events that gives roughly equal weight to Amazon and to your own online store. Most competing conferences are Amazon-only. Given how competitive Amazon has become with overseas sellers, most Amazon-only operators we talk to want to diversify and simply do not know where to start.

How to choose an ecommerce conference (what to look for)

The single best filter for an ecommerce conference is whether the speakers currently run the businesses they are teaching from, versus consultants, agency reps, or paid keynote celebrities. Operator-taught curriculum tends to survive contact with reality; theater-taught curriculum does not.

The second filter is size. Events over 500 attendees make it almost impossible to have a real conversation with a speaker or another operator at your level. Small events (100 to 250) usually deliver more actual business value because the hallway conversations become the event.

The third filter is agenda publication timing. Sellers Summit locks the curriculum two to three months out. Toni recently signed up for a newsletter conference where the session lineup was not shared until the week of the event, which is common in the industry and a warning sign about how much thought went into the curriculum.

What is the Sellers Summit mastermind day?

The Sellers Summit mastermind runs on Day 1 (May 6, 2025) as a full-day hot-seat session for qualifying attendees. Groups of ten operators are matched by revenue level and non-competing industry, moderated by a designated mentor, and given the day to workshop each other’s biggest business problems over catered food.

Placement into groups is done by hand. The organizers spend the bulk of their conference-planning time matching personalities (introvert with introvert, extrovert with extrovert), ensuring nobody sells competing products in the same room, and pairing groups with the right mentor. It is the highest-touch part of the event.

Multiple past mastermind groups have kept meeting monthly or bimonthly for years after the event, entirely on their own. Running a business is lonely; getting into a room with nine other operators at your revenue level is often the first time you talk shop with peers who actually understand your problems.

Ecommerce mastermind vs content mastermind: which one to pick

Sellers Summit runs two parallel masterminds on Day 1: the ecommerce mastermind for operators at $250K to $1M+ in revenue, and the content mastermind for anyone (including pre-launch founders and multi-million-dollar sellers who want the content angle). The two masterminds cover different topics and pull different attendees.

The ecommerce mastermind goes deep on ecommerce operations: sourcing, shipping, margin, hiring, unit economics, and any messy operational challenge someone brings to the hot seat. One year an entire mastermind session ended up focused on shipping-cost reduction because every operator in the room had the same problem.

The content mastermind is now in its fourth year and focuses on YouTube, TikTok, email marketing, and community. Last year several million-dollar sellers deliberately picked the content mastermind over the ecommerce one because content was the bottleneck for their next growth stage. Some multi-founder businesses split up: one founder goes to the ecommerce mastermind, the other goes to content.

Sellers Summit 2025 speaker lineup and sessions

The confirmed Sellers Summit 2025 speaker lineup covers ten locked sessions across sourcing, ads, ecommerce SEO, TikTok Shop, YouTube, community, and Klaviyo AI, with several more sessions still being finalized. Every listed speaker also runs a Q&A roundtable so attendees can go deeper after the mainstage talk.

YouTube plus email as a growth engine (Toni Herrbach)

Toni is teaching how to use YouTube and email together to grow an ecommerce brand, using the specific launch playbook she is running for one of her clients right now with strong early results. It is aimed at operators who already have a small email list (5,000+ subscribers is enough) and want a repeatable YouTube launch process that feeds email conversions.

The framework is YouTube for top-of-funnel discovery and email to close the sale. She will walk through the launch mechanics, the cadence, and the specific email flows that convert YouTube-referred subscribers into buyers.

Shoppable YouTube ads (Brett Curry)

Brett runs an ecommerce YouTube ads agency and is teaching the shoppable YouTube ads format, which is YouTube’s answer to TikTok Shop-style in-video commerce. His roundtable is packed every year because he opens attendees’ Google Ads accounts on the spot and audits campaigns live.

Expect a session heavy on real campaign screenshots, the specific ad units to run, and what shoppable formats are converting best right now.

Community-driven ecommerce (Sally Wilson, Caterpillar Cross Stitch)

Sally runs Caterpillar Cross Stitch out of the UK. She sells cross-stitch kits and supplies and has built a community so tight that she runs her own live event for her customers, which is almost unheard of for a physical-product store at her size.

She is teaching how to build that community from scratch and how to convert community engagement into a defensible sales channel. Building community is the single hardest and most underused ecommerce moat, and community-driven brands survive price hikes and algorithm changes that kill ads-dependent brands.

Sourcing outside China: Made in USA (Annette) and Mexico/Vietnam (Jim Kenimer, Cosmos Sourcing)

Two dedicated sourcing sessions cover the two most-requested alternatives to China. Annette teaches Made in USA sourcing (particularly textiles) and how to keep unit costs from destroying margin. Jim Kenimer of Cosmos Sourcing covers Mexico and Vietnam manufacturing, including how to find factories, negotiate MOQs, and structure orders in countries that are less documented than China.

With tariffs and trade policy shifting, sourcing diversification is now a top-three concern for most physical-product sellers. These two sessions are likely to be among the most-attended of the event.

Meta ads that survive algorithm turmoil (Andrea)

Andrea has consistently driven strong ROI on Meta ads for seven years across every platform change. Her differentiator is she knows how to pivot when ads collapse. Her session covers a real case from last year: a client whose Meta ads tanked on Black Friday (worst possible timing), and how she diagnosed the issue and restructured the account into that client’s best December and January of all time.

Her talks are famously no-fluff, timeline-based, and directly implementable during the session itself. Bring a laptop.

AI for ecommerce operations (Ritu)

Ritu had the most-praised session at Sellers Summit 2024, teaching specific ways to use AI in Amazon and DTC operations. She has a technical background (as does her partner Bernie) and uses AI to automate a wide range of ecommerce tasks. Her 2025 topic is not fully locked yet, but based on last year’s response, expect standing-room only.

Building your own audience beyond Amazon (Chris Shaffer)

Chris Shaffer has run brands on Amazon, Etsy, Shopify, and DTC, and consulted as fractional CMO for several others. His 2025 session is on building audience and selling on your own terms so you are not 100 percent dependent on Amazon. He covers SEO, social media, community, and the channel-mix logic that lets a brand shift ad spend between channels as costs move.

His Q&A sessions are legendary. Someone will ask an obscure question and he answers with a specific case study he ran in 2023.

Ecommerce SEO in the AI era (Jeff Oxford)

Jeff specializes exclusively in ecommerce SEO, which is rare (most SEO agencies are horizontal). His 2025 session covers what has changed with Google’s AI-driven updates and how to still get product pages and category pages to rank. He always presents a case study inside the talk and does live SEO audits at the roundtable afterward.

For anyone who has ever been burned by an SEO agency, Jeff is the exception that proves the rule. The ecommerce SEO industry is largely scammy; his approach is data-driven and rooted in what actually moves ecommerce revenue.

TikTok Shop from a top seller (Tiffany)

Tiffany is one of the top-earning TikTok Shop sellers we know personally and is completely transparent about her numbers within TikTok’s terms of service. Past sessions have included live selling walkthroughs that led multiple attendees (including my wife Jen) to launch their own live-selling programs afterward.

Her session is part teaching and part stand-up comedy, and even attendees who have no interest in selling on TikTok fight for seats because of the delivery. Whatever she is working on next, she will show it.

Klaviyo AI CRM features (Klaviyo team)

The Klaviyo team is running a session on the AI features from their newly announced Klaviyo Data Platform, including AI Forecast segmentation, the new campaign builder, the Action Center recommendations, and the customer-service hub. This is the same set of features Toni and I covered in a recent episode after attending Klaviyo’s launch event.

If you are already using Klaviyo for email or SMS, this session shows the features you probably do not know exist yet and how to actually turn them on.

Why the roundtables are the highest-ROI hour of the conference

The Sellers Summit roundtables are one-hour, hosted sit-downs where every speaker sits at a marked table and attendees rotate through to ask any question they want. Most attendees report a single 30-minute conversation at a roundtable pays for the whole ticket.

Every speaker at the event runs a roundtable, so the deep-dive you couldn’t get during the 40-minute mainstage session happens here. Last year one attendee sat down next to Brett Curry, opened their laptop, and had him do a live audit of their Google Ads account on the spot.

Sponsor roundtables carry equal weight. Klaviyo sends product people (not PR reps) who will open your account and fix flows on the spot. Quiet Light advisors run free consults for anyone thinking about selling or buying an ecommerce business.

RPC’s Pam handles freight forwarding questions in real time. She once got an attendee’s stuck container delivered in three days after their existing forwarder ghosted them. Ecommerce attorney Steve Weigler sends a calendar invite ahead of the event where you can book a 30-minute IP and trademark consult for free.

Frequently asked questions

When and where is Sellers Summit 2025?

Sellers Summit 2025 runs May 6 to May 8, 2025 in Fort Lauderdale, Florida. Tickets are capped at approximately 200 attendees and have sold out every year for the past eight years. Ticket prices rise every two weeks until the event, so earlier is cheaper.

Is Sellers Summit invite-only?

No, Sellers Summit is not invite-only and has never been invite-only. Anyone can buy a ticket at sellersummit.com. Beginners are welcome, and pre-launch founders attend every year and use the content mastermind to shape their product idea.

Do I need to have an existing store to attend?

No. You do not need an existing store to attend Sellers Summit. Most attendees have an active brand, but complete beginners and pre-launch founders are welcome. If you are pre-launch, the content mastermind is usually the better fit than the ecommerce mastermind.

What is the difference between the ecommerce mastermind and the content mastermind?

The ecommerce mastermind requires $250K+ or $1M+ in annual revenue and covers ecommerce operations (sourcing, shipping, margin, hiring). The content mastermind has no revenue requirement and covers YouTube, TikTok, email marketing, and community. Both run as ~10-person hot-seat sessions on Day 1 with catered food and a designated mentor.

Will the sessions be too advanced for a beginner?

Sessions at Sellers Summit are pitched slightly above the average attendee’s level on purpose, because beginner content is already free on YouTube and in most ecommerce courses. The value of an event is hearing the cutting-edge strategies working right now, which beginners can then implement several years ahead of the market.

Are there roundtables to ask speakers questions directly?

Yes. Every Sellers Summit speaker runs a roundtable in addition to their mainstage session, and attendees can go to the roundtable and ask any question they want, get a live account audit, or dive deeper into topics the speaker did not have time to cover on stage. Sponsors (Klaviyo, Quiet Light, RPC, attorney Steve Weigler) also run roundtable consult sessions.

What kinds of speakers present at Sellers Summit?

Sellers Summit only invites active operators who currently run the ecommerce business they are teaching from. Speakers include physical-product founders, agency owners who run active client accounts, ecommerce attorneys, freight forwarders, and platform product teams (like Klaviyo). It deliberately avoids consultants or paid celebrity keynote speakers.

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: 52:09 — 60.0MB)

The best million dollar side hustle business ideas for 2025 are the ones you can start this weekend for under a few hundred dollars, in categories where an existing service is either overpriced, opaque, or missing entirely. In this episode, my guest Nick Loper from Side Hustle Nation and I traded a dozen concrete ideas ranging from AI chatbot services for Shopify stores to transparent estate sale eBay agents, TikTok live selling services for ecommerce brands, and a family video biography service for aging parents. Every idea below solves a real problem one of us hit in our own life or business.

None of these are theoretical. Nick runs Side Hustle Nation, one of the largest side hustle communities online, and I run an ecommerce store, a course, this podcast, and a YouTube channel with close to half a million subscribers. The ideas we pitch each other are the ones we would build ourselves if we were not already running other businesses.

Below, you will find the full list, why each idea has money in it, who to sell it to, and how to start with almost no capital.

Key takeaways

• Pillow subscription service: pillows collect up to 16 species of fungi and 10 percent dust mite weight; a quiz-based replacement subscription every 3 to 6 months is an obvious consumables play.
• AI chatbot as a service for Shopify stores: package a trained-on-your-products bot and cold pitch stores via their contact forms. Demand already exists in seller forums.
• Transparent estate sale eBay agent: an itemized, above-board alternative to sketchy estate sale companies that take 40 to 50 percent with no receipts.
• TikTok and Facebook live selling as a service for ecommerce brands: one Chinese seller reportedly made $13.7 million in 7 days livestreaming. You do not need a following to start.
• Software migration specialist: when tools like LeadPages or Zapier hike prices, sell the migration to cheaper alternatives (OptimizePress, Make).
• Family video biography service: film crew interviews aging parents on video; a step beyond the story books that require too much writing.
• Reddit marketing service: paid help getting brands mentioned in Reddit threads now that Google prioritizes user-generated content.
• Digital baby service: lock down domains, email, and social handles for newborns for a yearly fee.
• Smart home garage door “second guess remover”, solar consulting matchmaker, restaurant food photo gigs, rentable Instagram sets, and an AI podcast editor round out the list.

Why “million dollar business ideas” is the right frame for a side hustle

Calling these million dollar side hustle business ideas is not hype. It is a framing choice.

Nick and I both look for ideas where the ceiling is high enough to justify the effort but the entry cost is low enough to test in a weekend. Every idea in this episode meets three tests.

First, there is proven demand. Either people are already paying for a worse version of the service, or a specific complaint keeps showing up in forums, subreddits, or in-person conversations.

Second, the startup cost is under a few thousand dollars, usually well under. A laptop, a Shopify contact-form outreach spreadsheet, or a camera is enough for most of these.

Third, the idea is defensible with a niche. Not by patent or moat, just by being the first person a specific audience thinks of when the problem hits.

How to pick a side hustle idea you will actually finish

Pick the idea that solves a problem you personally hit in the last 30 days. Every idea Nick and I pitched came from something one of us lived through: my mom moving, our garage door second-guessing, a friend’s dating profile, a house break-in, a software price hike.

That is the shortcut. Personal pain gives you free customer research, real conversations to test the pitch, and enough motivation to grind through the first ugly weeks.

If nothing recent qualifies, look at what you pay for or complain about. The friction you already tolerate is somebody’s opportunity.

Pillows as a service (a subscription business hiding in plain sight)

The pillow subscription service replaces your pillow on a 3 to 6 month cadence based on a quick sleep-position quiz. Nick pitched this after reading that a used pillow can contain up to 16 species of fungi and that up to 10 percent of its weight is dust mites and their waste. Experts recommend replacing pillows far more often than most people do, and almost nobody actually does it.

The business model is a classic consumables subscription. Quiz users on back or side sleeper, density, and thickness, ship them a matched pillow, then auto-ship a replacement on a schedule.

The differentiator is education and cleanliness anxiety, not the pillow itself. Amazon sells pillows. What Amazon does not sell is the reminder that you have been sleeping on 16 species of fungi for four years.

AI chatbot service for Shopify stores (train, install, resell)

Sell trained AI chatbots to Shopify stores as a done-for-you customer service tool. This is my idea and it comes straight from my ecommerce store. Most stores get the same two questions over and over: “where is my order?” and “which product should I buy?” A chatbot trained on your product catalog and past support transcripts answers both without human intervention.

Existing services charge enterprise prices for what is now, thanks to modern AI tools, a spreadsheet-plus-prompt job. Feed the bot every product description, an FAQ, and 10 years of past support tickets, then plug it into the store’s help widget.

To find customers, go through the public Shopify store database and cold pitch through their contact forms. Sellers are already asking how to do this in Shopify forums, which means the demand is warm.

The garage door “second guess remover” (smart home consulting)

The garage door second guess remover is a smart home service that installs and configures a phone-checkable garage door, security camera, and door-lock setup so you never have to turn the car around. Nick’s example was the family drive from California to Washington where they realized 20 minutes out that they had left the garage door wide open for two weeks. My example was less charming: our house was broken into a few weeks before we recorded, and we now have cameras everywhere.

This is a productized service, not a product. Charge for the consult, the install, and the app setup. Repeat business comes from adding cameras, sensors, and monitoring over time.

Positioning matters here. Sell peace of mind, not gadgets.

The people who buy this are not shopping specs. They are shopping “I will stop lying awake wondering.”

Live selling services for ecommerce brands (TikTok, Facebook, Instagram)

Live selling as a service means you go live on TikTok, Facebook, or Instagram on behalf of ecommerce brands and sell their products in exchange for a commission. This is already massive in China. One well-known livestreamer reportedly did $13.7 million in sales in 7 days on Douyin (the Chinese TikTok).

In the US, TikTok Shop has made this possible, and the market for freelance live sellers is still wide open.

You do not need an existing following to start. TikTok’s algorithm surfaces consistent creators regardless of follower count. What you need is on-camera presence, willingness to go live multiple times a week, and a demographic match for the brand.

The pitch to brands is pure performance: commission-only, similar to affiliate marketing, no cost until a sale happens. Brands ship product on consignment. You keep a percentage of every order that clears through your stream.

Solar consultant and matchmaker (fix a shady industry)

The solar consulting business is an independent local advisor who tells homeowners whether solar makes sense for their house, brokers the install with vetted local vendors, and cuts through the shady sales pitches. Nick has been trying to figure out his own solar install and ran into exactly the problem this business solves: opaque pricing, confused tax rebates, and salespeople who promise things that are not real.

Charge a flat consulting fee up front, then a referral commission on any install through your vendor network. Because trust is the whole product, you have to be genuinely willing to tell a client “solar is not for you.”

Adjacent side hustle: solar panel cleaning. Solar output can degrade quickly as dirt accumulates. A route-based cleaning service can add recurring revenue in the same neighborhoods.

Home energy backup consultant (a real service after 2024 storms)

The home energy backup consultant designs battery-plus-generator plans for homeowners who lost power in a recent storm and want to be ready for the next one. Nick lost power for 5 days during the November 2024 “bomb cyclone” in the Pacific Northwest, along with roughly half a million other households. He had to throw out a fridge worth of food, and afterward, every neighbor was asking the same question: battery, generator, or both?

The service walks a homeowner through actual load requirements, tradeoffs between whole-home and partial backup, and realistic budgets. Some clients will hear “buy a couple of hotel nights and restock the fridge, that is cheaper than a battery.”

Recurring revenue comes from annual system checks and battery replacements every 5 to 10 years.

Restaurant food photo gig ($3 per organic photo)

Restaurant food photo work pays a per-photo fee to take casual, authentic pictures of restaurant meals for delivery-app listings. I pitched this after ordering Chinese food for Chinese New Year and noticing that the restaurant’s app was offering $3 per photo of their dishes.

The photos are meant to be casual and customer-like, not staged product shots. Restaurants believe organic-looking images boost app conversion.

Start by eating at (and photographing) restaurants you already visit. As you build a portfolio, walk into neighborhood restaurants and offer a batch: dine-in, 20 photos, invoice. If enough places sign on you can turn it into a route.

Scale by hiring college students or part-timers in each city to run the same route locally.

Software migration specialist (ride price-hike anger)

The software migration side hustle helps businesses move from a tool that just hiked prices to a cheaper equivalent, for a flat fee per migration. Nick spent a full day migrating off LeadPages when they nearly doubled their rates after a decade of stability, and he moved everything to OptimizePress at about 25 percent of the price. I did the same thing years ago when Zapier’s pricing changed: I moved everything to what is now called Make (formerly Integromat).

Pick one pair of tools and specialize. Examples that already have migration demand: LeadPages to OptimizePress, Zapier to Make, ConvertKit to Beehiiv, HubSpot to a stack of cheaper alternatives.

Every price hike creates a fresh wave of angry Reddit threads. Reply with a done-for-you offer.

Charge per migration plus optional retainer for optimizing the new tool. Reuse a checklist across every client to keep the actual work fast.

Reddit marketing service (Google now favors Reddit answers)

Reddit marketing services build up organic-looking Reddit accounts and place brand mentions in high-intent subreddits. Google’s recent updates have pushed Reddit and Quora threads to the top of search results for a huge share of commercial queries.

Brands watching their organic traffic collapse are paying real money to appear inside the Reddit answers now stealing their clicks. Our mutual friend Spencer Haws tried one of these services and had good results.

The work looks like ghostwriting for Reddit. Build accounts with genuine karma over months, participate authentically, then place strategic recommendations in the threads where your client would fit the discussion.

This is a gray-area business by Reddit’s rules. If you take it, pick brands whose products genuinely deserve the mention, or your accounts get banned and your reputation with them.

Transparent estate sale eBay agent (kill a shady industry)

A transparent estate sale eBay agent lists an aging parent’s or downsizing homeowner’s belongings on eBay with full itemized reporting and a clear commission, in place of the traditional 40 to 50 percent estate sale companies. I ran into this when my mom moved from a large house to a small one.

Every estate sale company I interviewed in Maryland handed me a check at the end with no itemized list of what sold, for how much, or when. None would come to her house and eBay everything.

The pitch is anti-positioning: transparency in an industry known for the opposite. Sellers see every listing, every winning bid, and every payout.

Get first clients through friends, family, and a partnership with local realtors who already know which of their clients are downsizing. Move to consignment-only pricing so nobody pays until items sell.

The “Robo dating profile” service (AI photos plus copy)

The AI dating profile service uses modern AI photo and copy tools to build an optimized dating profile for busy singles. This idea came from watching my wife Toni help several attendees fix their profiles at Seller Summit. Multiple guys, at an ecommerce conference, asked her to look at Hinge or Bumble because they knew their profile was underperforming.

Now AI tools can retouch existing photos into professional-looking portraits and rewrite bios in minutes. That collapses what used to be a $500 photographer session plus a coach fee into a $99 service.

The market is real: people are getting married later, having kids later, and the loneliness numbers keep growing. If you can help someone match twice as often, that is worth real money.

Family video biography service (a step past the “story book”)

The family video biography service sends a small film crew to interview an aging parent or grandparent on video, then edits the footage into a keepsake documentary. I got this idea from a service that emails your parent a question a week and compiles the answers into a book. My mom found it burdensome (the writing itself was hard), and the finished book, while treasured, was hard to produce.

Video removes the friction. Send prepared questions in advance, arrive with a two-person crew, film for a day, and deliver an edited version plus raw footage.

Upsell: a documentary edit with music, photos, and archival footage. Cross-sell: a companion audio version for podcast-style listening. The Christmas gifting season alone can drive enough demand to fill a small team.

Digital baby service (lock the domain and handles at birth)

The digital baby service secures a newborn’s domain name, primary email, and top social media handles for a small yearly fee. Nick pitched this at around $100 per year.

Almost all of it is margin. The domain renewal is roughly $10 to $12; the rest is pure service.

The value is timing. Parents cannot easily register or think about this in the first exhausted year of a baby’s life, and the names they want will be taken by the time the kid cares.

Bundle a “welcome to life” package including personalized domains, email, Instagram, TikTok, YouTube, and Gmail addresses. Renew automatically. This is a subscription with 15-plus years of runway per customer.

Instagram set rental (a warehouse of fake luxury)

An Instagram set rental business builds and rents time-block access to Instagrammable sets like a fake private jet interior, a luxury hotel suite, or a botanical garden. Nick and I both find the trend a little grim, but the demand is undeniable. There are already jet-set rental businesses charging by the hour.

The economics work because you build once and rent forever. Fit 4 or 5 sets into a warehouse.

Rent 2-hour blocks. Add optional lighting and camera crew for a premium.

Prime customers are influencers, MLM sellers, real estate agents, and any make-money-online creator who needs their backdrop to imply success.

AI podcast editor (the “I’m feeling lucky” cut)

The AI podcast editor takes a raw episode transcript and suggests which sections to cut for a tighter listening experience, on a sliding scale from a 5 percent trim to an aggressive 25 percent trim. Nick has wanted this tool for a decade of podcasting. Existing tools like OpusClip focus on short-form social clips, not the actual long-form edit.

The product feeds the transcript plus a few examples of past episodes the host loved into an AI model and returns time-stamped trim suggestions. A human editor reviews the suggestions before cutting.

This is a paid tool with an obvious audience: every podcast host who currently pays an editor to comb 60 minutes of tape looking for the good 45.

The video scripting service (Nick’s personal bottleneck)

The video scripting service turns a founder’s blog posts, transcripts, and past content into ready-to-record YouTube scripts written in the founder’s voice. Nick admitted this is his bottleneck. My YouTube channel is closing in on half a million subscribers, and even I still spend about 90 minutes scripting a 10-minute video, using AI as a starting point.

The service ingests a body of the founder’s existing writing, trains an AI prompt on their voice, and delivers a script the founder can either read into a teleprompter or narrate as voiceover.

Charge per script with a discount for a monthly package. Add-on: thumbnail concepts, hook lines, and title options. This is a rare service where you can prove ROI in the first video.

The forgotten skills online course library

The forgotten skills course library packages first-principles instruction in skills that have been outsourced to machines: how to read a paper map, Asian calligraphy, basic metalworking, film photography chemical development, Morse code, and simple household repair. I got the idea from spending more time with my mom now that she lives down the street and hearing how much she had to do from scratch.

The audience is people who dislike being helpless when a service goes down or a tool breaks. That audience is larger than it looks, and it overlaps with prepping, homesteading, off-grid, and “digital detox” communities.

Charge a subscription for the full library or single-course pricing for the highest-intent skills (like household electrical repair) that solve immediate real-world problems.

How to launch any one of these million dollar side hustle business ideas this month

Pick one idea today, write one paragraph describing exactly what it is and who buys it, and then go find 10 potential customers by Friday. That is the whole starting move.

Every idea above can be tested with 10 conversations. If 3 of the 10 sound interested, you have a business. If zero, kill it and pick the next idea on the list.

Nick and I both agree the biggest mistake first-time side hustlers make is spending months building a logo, a website, and a Stripe integration before finding out whether anyone wants the thing. Do the ugly cold outreach first. Sell before you build.

Frequently asked questions

What is the easiest side hustle to start with no money in 2025?

The easiest side hustle to start with no money in 2025 is a service-based one where you sell your time and a laptop. Examples from this episode include Reddit marketing, software migrations, dating profile optimization, and food photography. All of them require zero inventory, no software licenses, and a first customer you can find in a weekend of outreach.

How much can a side hustle realistically make in the first year?

A focused side hustle can realistically make between $500 and $10,000 per month within 12 months if you consistently deliver a service for paying clients. The high end depends on pricing, retention, and how quickly you productize. Product businesses (like the pillow subscription) tend to take longer to reach the same revenue because they need upfront capital and marketing.

Are AI-powered side hustles like chatbot services actually viable in 2025?

AI-powered side hustles are viable in 2025 because the tools are now cheap enough that the value you add is packaging and installation, not the AI itself. A trained Shopify chatbot, an AI script writer, or an AI dating profile builder all sell the outcome, not the model. Customers are paying you to save time and avoid the learning curve.

Is TikTok live selling a real business or a fad?

TikTok live selling is a real business that has already produced eight and nine figure sellers overseas and is now scaling in the US via TikTok Shop. In China, one live seller reportedly made $13.7 million in 7 days on Douyin. US brands actively want commission-only live sellers, so the freelance side hustle version has genuine demand.

What side hustle idea has the highest margin?

Digital services and information products have the highest margins because they are near-zero variable cost. Domain registration bundles (like the digital baby service), online course libraries, migration services, and Reddit marketing all clear 80 percent-plus margins. Physical product subscriptions like the pillow business have thinner margins but higher lifetime value.

How do I know if my side hustle idea has a real market?

You know your side hustle idea has a real market when at least 3 out of 10 cold-pitched potential customers say they would pay for it. Do that test before you build anything. Personal pain (an itch you have hit yourself in the last 30 days) is a strong second signal because it usually means at least a few thousand other people hit the same problem.

Which of these side hustle ideas can grow into a full time business?

Any of the productized services and subscription businesses in this list can grow into a full time business, especially the AI chatbot service, the transparent estate sale eBay agent, the family video biography service, and the software migration specialty. Each has a clear ceiling above six figures once you hire a small team and run it as a route or agency.

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: 13:50 — 16.1MB)

Amazon’s dirty tactics for sellers in 2025 fall into five categories: stacked FBA fee hikes with a “safety valve” (AWD) that failed at Q4, a slashed lost-inventory reimbursement policy, aggressive lowest-price enforcement that Temu is exempt from, hostile customer service that treats sellers as guilty until proven innocent, and active suppression of your own brand in favor of Chinese knockoffs and Amazon private label ads placed directly on your listings. After 10 years of selling on Amazon and generating millions on the platform, my take is simple: the platform still moves volume, but the effort-to-profit ratio has collapsed and every seller needs an exit plan.

In this episode I walk through each tactic, name the specific 2024 and 2025 fee and policy changes, and share what real sellers (Amanda from my course, eight-figure seller Molson Hart at Viahart, my own store Bumblebee Linens) are dealing with right now. The takeaway at the bottom: stop treating Amazon as your only channel.

Below we cover the AWD trap, the new lost-inventory reimbursement rules, the lowest-price policy and the Temu loophole, the customer service and copyright-hijack problem, and the brand-suppression tactics that are the single biggest hit to seller profitability.

Key takeaways

• Amazon FBA storage rates, inbound placement fees (27 cents to $1.58 per unit), and low-and-high inventory penalties keep stacking every year.
• AWD (Amazon Warehousing and Distribution) is pitched as a safety net but its Q4 2024 breakdown left sellers with canceled pickups and delayed transfers, and Amazon still raised its price in 2025.
• Effective March 2025, Amazon only reimburses “manufacturing cost” when it loses your inventory, not the retail price or landed cost. Freight, tariffs, and duties are excluded.
• Amazon’s lowest-price enforcement buries your listing if you sell cheaper on Walmart, Target, or eBay. Temu appears to be exempt, which is why the same product can be $40 on Amazon and $1 on Temu.
• Copyright-infringement takedowns are handled guilty-until-proven-innocent. Viahart lost 126 SKUs over Black Friday and Cyber Monday to a false Liahart claim despite owning the trademark.
• Search for your own brand and most top results will be competitors. Amazon places its own private-label ads directly on your product page at a lower price.

Why Amazon’s tactics against sellers matter to shoppers too

Amazon’s dirty tactics against sellers eventually reach shoppers through higher sticker prices, fake sales, and worse product quality. Every fee increase gets baked into the retail price.

Every “deal day” is offset by an inflated base price the other 340 days of the year. And when Chinese knockoffs win the Buy Box, shoppers get an inferior product with the same reviews attached.

Sellers are the canary. When the honest ones leave the platform, what remains is a marketplace of arbitrage, private label, and knockoffs.

Tactic 1: Amazon FBA and AWD fee hikes (the fake safety net)

Amazon’s first dirty tactic is stacking FBA fees every year, then offering AWD as an “escape” while raising AWD prices too. FBA used to be simple: send everything to one warehouse, Amazon does the rest, at a reasonable cost. Now sellers face annual storage rate hikes, holiday storage penalties, and inbound placement fees ranging from 27 cents to $1.58 per unit.

On top of that, Amazon penalizes sellers for both having too much inventory (long-term storage fees) and not enough (low-inventory-level fees). Hitting Amazon’s ideal inventory window is nearly impossible in practice.

The “solution” Amazon markets is AWD (Amazon Warehousing and Distribution). Ship everything to one central warehouse and Amazon sorts, stores, and redistributes as needed.

That is exactly how FBA used to work before the fees stacked up. AWD is a rebranded return to the old model, at a fresh new price.

How AWD failed in Q4 2024

AWD failed in Q4 2024 when demand exceeded capacity, pickups were canceled or delayed, and transfers from AWD to FBA warehouses stalled. Sellers I know missed critical holiday sales because inventory that Amazon had physically taken possession of never made it to the fulfillment centers that needed it. And after that Q4 disaster, Amazon still announced an AWD price increase for 2025.

Tactic 2: The new lost-inventory reimbursement rule (you eat the loss)

Amazon’s second dirty tactic is a March 2025 policy that only reimburses sellers for the “true manufacturing cost” when Amazon loses inventory, not the retail price or the landed cost. Freight from China, customs duties, and tariffs are excluded from the calculation.

Consider a $10 product that costs you $2 to source, plus $1.50 in freight and duties (landed cost around $3.50). Under the old rule, if Amazon lost the unit, you got $10 back.

Under the new rule, you get around $2, minus shipping and import duties Amazon does not credit back. Every warehouse mistake now moves money from your P&L into Amazon’s.

Amazon announced this reimbursement change and the AWD price hike right after publicly claiming there would be no new seller fee increases in 2025. That is the second reason sellers no longer trust Amazon’s fee guidance.

Tactic 3: Lowest-price enforcement and the Temu loophole

Amazon’s third dirty tactic is punishing sellers whose products are cheaper on Walmart, Target, or eBay by suppressing their Amazon listings, while apparently ignoring the same undercutting on Temu. Amazon uses web crawlers to check off-Amazon prices constantly. Get caught with a cheaper listing elsewhere and your Amazon listing effectively disappears from search.

Temu appears to be the one large marketplace Amazon does not price-check against. Temu charges no listing fees and near-zero transaction fees, which lets sellers list the same SKUs at fractions of the Amazon price.

I have seen a floor mat listed at $40 on Amazon and $1 on Temu. As a shopper, once you know that, why stick with Amazon?

This lowest-price enforcement is one of the specific practices cited in the FTC’s ongoing antitrust suit against Amazon.

Tactic 4: Hostile customer service and copyright hijacks

Amazon’s fourth dirty tactic is a support system that treats sellers as guilty until proven innocent on any copyright, trademark, or infringement claim, no matter how obviously false. Getting a human on the phone is nearly impossible, and the canned responses rarely resolve anything.

Amanda, a student in my Create A Profitable Online Store course, designs custom party supplies. She has gone through the work of registering copyrights and trademarks for her top products.

Chinese sellers still file false infringement claims against her listings on a rolling basis. Every time it happens she loses months of sales while she clears her name, and then the cycle starts again.

The Viahart Black Friday takedown

Molson Hart at Viahart, an eight-figure seller, had 126 of his listings taken down over Black Friday and Cyber Monday for infringing a trademark called “Liahart.” His company is Viahart LLC, his seller name is Viahart, he owns the Viahart trademark, and Viahart is in Amazon’s Brand Registry.

Amazon’s canned reply said all products using the term “Viahart” would remain suspended. He got his listings back eventually, and he also lost tens of thousands of dollars on the biggest sales weekend of the year.

Competitors buying out and returning your inventory

At my store Bumblebee Linens we have had a competitor buy out all of one SKU’s inventory almost every year, sit on it through the holidays, and return everything in January. We make zero sales on that SKU during peak season and lose money on FBA and return fees. Amazon’s return policy makes this attack cheap for the attacker and expensive for the seller being attacked.

Tactic 5: Amazon suppresses your brand and puts its own ads on your listing

Amazon’s fifth and most damaging tactic is actively suppressing your own brand in search and then placing ads for its private label, or a competitor’s product, directly on your product page. Search for my store Bumblebee Linens on Amazon and only 3 of the top 10 results are actually my products. The other 7 are competitors, and most of those are Chinese knockoffs paying for Sponsored Brand real estate.

Viahart owns the trademark for Brain Flakes, yet a Chinese knockoff holds the Amazon’s Choice badge on the branded search. Amazon rewards whoever wins the algorithm, not the actual brand owner.

Then Amazon takes it further. On a friend’s emu oil listing, Amazon places an ad for its own private-label emu oil at 30 percent off directly in the middle of her product page. She is paying for advertising on Amazon, and Amazon uses that page to promote its cheaper knockoff.

How much all this has changed the math on Amazon FBA

The math on Amazon FBA has changed enough in the past 3 years that most six-figure private-label sellers I talk to now consider Amazon a distribution channel, not a business. 2024 was the largest single year of seller fee and policy change increases Amazon has ever rolled out. Between higher FBA rates, new AWD fees, holiday inventory penalties, deal-day fees ($150 to several hundred per SKU for a Lightning Deal), and slashed reimbursements, the effort-to-profit ratio has collapsed.

The workload has increased in step. Sellers spend more time defending listings, fighting takedowns, and forecasting inventory to Amazon’s specifications than they do growing revenue.

What to do if you are still selling on Amazon in 2025

If you are still selling on Amazon in 2025, treat it as a paid distribution channel and rebuild everything else around your own audience. Amazon is fine for volume today. It is not fine as your only channel because you do not control any of it: your listing, your price, your reviews, or your customer relationship.

Practical moves worth making now:

• Launch or grow your own Shopify (or comparable) storefront and treat it as your primary brand home.
• Build an email list from every Amazon insert, every offer, and every off-Amazon touchpoint you have. Email is the one asset Amazon cannot take from you.
• Diversify off-Amazon: Walmart, TikTok Shop, Faire (wholesale), your own DTC site, retail placements.
• Trademark your brand and enroll in Brand Registry. It reduces takedown frequency and makes fighting bogus claims faster.
• Stress-test every SKU’s profitability at the new reimbursement rate, new AWD price, and higher inbound fees. Kill anything that only worked at the old numbers.

Frequently asked questions

What are the biggest Amazon fees hurting sellers in 2025?

The biggest Amazon fees hurting sellers in 2025 are the annual FBA storage rate increases, inbound placement fees (27 cents to $1.58 per unit), low-inventory-level fees, high-inventory long-term storage penalties, AWD storage costs, deal-day participation fees ($150-plus per Lightning Deal), and the effective loss from Amazon’s new “manufacturing cost only” reimbursement rule when they lose inventory.

What changed with Amazon’s lost-inventory reimbursement policy?

Amazon’s lost-inventory reimbursement policy changed in March 2025 so that Amazon only reimburses sellers for the estimated manufacturing cost of a lost unit, not the full retail price. The new calculation excludes freight forwarding fees, tariffs, and customs duties. For most private-label products imported from China, sellers now recover only a small fraction of their true landed cost when Amazon loses a unit.

Does Amazon really price-check against Walmart and Target but not Temu?

Amazon actively price-checks seller listings against Walmart, Target, eBay, and other large US marketplaces, and will suppress a listing if it appears cheaper off Amazon. Temu does not appear to be included in that enforcement, which is why many sellers list the same SKUs on Temu at a fraction of the Amazon price without Amazon suppressing them. This is one of the practices cited in the FTC’s antitrust case against Amazon.

What happened with the Viahart Amazon suspension?

Viahart, an eight-figure Amazon seller run by Molson Hart, had 126 listings suspended over Black Friday and Cyber Monday 2024 based on a false infringement claim against the trademark “Liahart.” Viahart owns the Viahart trademark, is enrolled in Amazon Brand Registry, and eventually got the listings restored. The suspension still cost the company tens of thousands of dollars in sales on the biggest shopping weekend of the year.

Why do Chinese knockoffs rank ahead of the real brand on Amazon?

Chinese knockoffs frequently outrank the real brand on Amazon because they aggressively bid on branded search terms, run promotions Amazon’s algorithm rewards, and file infringement claims to suppress the real brand’s listings. Amazon’s search results reward whoever performs best on the platform’s internal metrics, not whoever owns the trademark. Even the Amazon’s Choice badge can end up on a knockoff, as it has on Viahart’s Brain Flakes product.

Should I quit selling on Amazon in 2025?

Keep selling on Amazon in 2025 if the SKU still pencils out, and stop treating it as your only channel. Amazon still moves volume, and with the fee stack, reimbursement changes, and brand suppression, most sellers are better off using Amazon as a paid distribution channel while building a real brand on their own site, email list, and diversified marketplaces like Shopify, Walmart, TikTok Shop, and wholesale.

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: 39:39 — 45.7MB)

Our 2025 content and ecommerce masterplan is built on three moves: shift the primary content platform from written blog posts to short-form video (TikTok, YouTube Shorts, Instagram Reels) built around a pillar overview format, protect the ecommerce business from tariffs and knockoffs by going deeper into personalized and printed goods that are structurally hard to copy, and treat blogs as feed material for AI answer engines rather than as standalone Google plays. In this episode my co-host Toni Herrbach and I walk through what we are changing in both my business (Bumblebee Linens plus the My Wife Quit Her Job media brand) and Toni’s (The Happy Housewife plus her ecommerce coaching work).

The uncertainty going into 2025 is real: a new US administration promising broad tariffs, China preparing counter-moves, Google losing search share to AI, and Amazon copying Temu with the new Amazon Haul marketplace. But the response is not to panic. It is to lean into brand, content, and product moats that no algorithm change can wipe out overnight.

Below we cover the state of blogging in an AI-first search world, the pillar video strategy that is replacing pillar blog posts, our specific ecommerce moves (Bumblebee Linens’ shift into digital printing and handwriting embroidery), the tariff math that is less scary than the headlines suggest, and why Amazon sellers are the most exposed group in 2025.

Key takeaways

• Traditional Google search is losing share, but well-ranked blog content still feeds AI answer engines like ChatGPT, Perplexity, and Google’s AI results. Do not sell your blog.
• Local business search on Google is one of the last strongholds. If you have a physical location or stock in retail stores, blogging plus a basic site still wins.
• Short-form video is the new pillar-and-cluster. Make one overview video (a full Thanksgiving dinner, a room remodel), then split each sub-step into its own TikTok, Reel, or Short.
• Tariffs are less scary than the headlines. On a $1 landed cost / $4 retail item, a 25 percent tariff raises your COGS by $0.25, not $1.
• Amazon private-label sellers are the most exposed to tariffs because Amazon margins already run 10 to 15 percent, and Amazon’s price wars leave no room to pass costs through.
• The Bumblebee Linens 2025 strategy: digital printing plus handwriting embroidery so every occasion (weddings, memorials, holidays) is personalized and structurally hard to knock off.
• Amazon Haul (Amazon’s Temu clone) is undercutting Amazon’s own core. A $7 Amazon jump rope is $3 on Haul and $2.50 on Temu.

Why the 2025 ecommerce and content plan has to change from 2024

The 2025 ecommerce and content plan has to change from 2024 because three forces hit at once: AI search stole a chunk of Google’s traffic, ad platforms (Google Ads and Meta) grew more expensive and less effective, and China’s flood of same-SKU low-price competitors expanded from Amazon to every marketplace including Amazon’s own new Haul storefront. If your plan for 2025 is a repeat of 2024, you will underperform on both traffic and margin.

The response is not more of the same. It is a structural shift toward content formats and product types that survive when algorithms change.

Is blogging dead in 2025?

Blogging is still alive in 2025, though its role has shifted from “primary traffic engine” to “feedstock for AI answer engines and local search.” I would not start a blog today purely as a Google-traffic business. If you already have one, do not sell it because your content is what ChatGPT, Perplexity, and Google’s AI Overviews cite when they generate answers.

Many of my longtime blogger friends are burnt out and selling. Some are shutting down entirely. I understand the exhaustion (Google updates, ad revenue drops, affiliate cuts, all of it), but the asset itself still has real forward value.

Local and travel queries are still Google-first

Local and travel queries are still Google-first because searchers need current, verifiable facts (business hours, phone numbers, whether the reindeer ride is still open). When Toni searched for “where can I go on a reindeer ride” for her upcoming trip, almost every top result was a blog or a destination site, with almost no AI Overview and no Reddit thread in sight. If you run a local business or you cover local topics, this window is still open.

What the AI search shift means for existing blogs

The AI search shift means existing blogs should optimize for citation, not just clicks. That means clean structured data, question-shaped headings, standalone paragraphs that answer questions in the first two sentences, and comparison tables or numbered lists that AI answer engines can lift directly. Traditional traffic will decline for many topics, but the well-optimized posts will keep earning brand mentions inside AI answers.

Pillar video is replacing pillar blog posts for 2025

The 2025 content strategy is to build a pillar overview video, then break each sub-step into its own short-form video. This is the old pillar-and-cluster SEO model applied to TikTok, YouTube Shorts, and Instagram Reels. Toni’s example is Thanksgiving dinner: one full “here is my Thanksgiving menu” overview video, plus one video each for the turkey, the mac and cheese, the scalloped potatoes, and every other dish.

That format is exploding on TikTok because it lets viewers save, share, and text specific recipes to specific people. Toni watched a scalloped potato TikTok because her daughter texted it to her from a family thread that had been going all morning.

How to make pillar videos when you are not a professional editor

Professional editing is optional for pillar videos, though good sound, clear framing, and a willingness to be visible on camera are all required. Toni watched a recipe TikTok where the creator tried an overhead mount, admitted on camera that she did not like the angle, and moved the camera back.

That video hit over a million views. Being real beats being polished.

If you have editing talent (or can hire an overseas editor), snappy edits do help. Toni pointed to “Lady in the Bathroom” as a case where the edits themselves are the growth driver.

The Bumblebee Linens video plan for 2025

The Bumblebee Linens video plan for 2025 is three TikToks a week (ramping to five) filmed at the office, mixing behind-the-scenes clips with employee-shot footage MJ compiles into short videos. My best format is going to be “pack a box with me” where I pack a real customer order, call out the customer’s first name, and drop a small personalized surprise. That combination is cheap (a few dollars per surprise), viral-friendly (people tag friends named Anna), and directly ties the video to a purchase moment.

How to think about tariffs on ecommerce in 2025

Tariffs on ecommerce in 2025 are less scary than the headlines because a 25 percent tariff hits your landed cost, not your retail price. A product that costs you $1 to source and retails for $4 sees a $0.25 tariff. Passing that entirely through takes retail from $4.00 to $4.25, roughly a 6 percent price increase, not a 25 percent one.

We have been here before. Trump’s 2016 term announced 45 percent tariffs across the board and most staples ended up in the 7.5 to 25 percent range. We did not raise Bumblebee Linens prices meaningfully to absorb it.

Who should actually worry about tariffs

The people who should worry about tariffs are Amazon private-label sellers because Amazon margins are already thin (often 10 to 15 percent) and Amazon’s constant price war leaves no room to pass costs through to customers. A quarter added to your COGS on a $10 item is a real dent when your net was already $1.50.

If you sell on your own Shopify or BigCommerce store, you have more pricing flexibility, more brand loyalty, and more room to absorb or pass costs through cleanly.

Chinese factories are also already moving to Vietnam and Mexico (often owned by the same operators) to sidestep tariffs. That will blunt some of the effect for sourcing.

The Amazon Haul problem (Amazon’s own Temu clone)

Amazon Haul is Amazon’s Temu clone and it is undercutting Amazon’s own core marketplace on identical SKUs. The example that stuck with me is a jump rope: $7 on Amazon, $3 on Amazon Haul, $2.50 on Temu.

All three are the exact same product from the exact same factory. Amazon is in a rock-and-a-hard-place because if Haul succeeds it eats their main business, and if it fails, Temu and Shein keep winning the low-end shopper anyway.

For sellers, this reinforces the “brand or be commoditized” split. Anything a shopper can find identical on Haul or Temu will lose the price war.

Why brand is the only durable ecommerce moat in 2025

Brand is the only durable ecommerce moat in 2025 because it is the one thing Chinese sellers cannot instantly clone. Uggs are wildly expensive and the knockoffs (Toni’s daughter has both) feel identical on your foot.

My wife Jen still wanted the real Uggs. That is the brand tax at work, and it is what saves margin when the same SKU shows up for a third of the price on Haul.

Small brands can create the same effect at their own scale. Chili’s grew profits by roughly 40 percent after a TikTok blew up their three-for-$10.99 deal.

GoDaddy’s Super Bowl commercial in the 2000s was a big-money brand moment. Today the equivalent moment costs zero if a single video lands.

The Bumblebee Linens digital printing plus handwriting embroidery move

Bumblebee Linens is going deep into digital printing and handwriting embroidery in 2025 because both create structurally hard-to-copy products with low inventory risk. Digital printing lets us produce any special-occasion design on the same blank substrate: Mother’s Day, Valentine’s Day, weddings, memorials. We do not need to warehouse variations because everything is printed on demand.

Handwriting embroidery takes an actual signed handwritten note (from a wedding vow, a grandmother, a lost loved one) and embroiders it onto a handkerchief or napkin. Nobody offshore can replicate that at the price point, because the customer’s own handwriting is the product.

The barrier for a would-be competitor is real: a DTF printer alone runs into the thousands, breaks down constantly, and requires monthly maintenance and troubleshooting. My printer clogged repeatedly for months before I figured out the cleaning routine. That mess is the moat.

The 2025 formula in one sentence

The 2025 formula for ecommerce and content is a product that is hard to knock off plus a content engine built on short-form video plus a blog that keeps feeding AI answer engines. Everything else (ads, Amazon, marketplace roulette) is secondary. Ad platforms keep getting more expensive and less effective, so free organic distribution on TikTok, YouTube, and Instagram is where the leverage is.

Layered on top: affiliate and influencer marketing on TikTok Shop still works well for interesting products, especially when the price point is impulse-buy friendly and the creator can demo the product on camera.

Frequently asked questions

Is it worth starting a blog in 2025?

Starting a blog in 2025 as a standalone traffic play is a weak play for most people. A blog attached to a real business (ecommerce store, local service, coaching practice) is still worth building because it feeds AI answer engines that increasingly influence buying decisions, helps with local search where AI has not taken over yet, and creates on-site content the business owns forever.

Should I switch my content strategy from blog to video in 2025?

Keep your blog going as feed material for AI search, and shift your primary content investment in 2025 into short-form video on TikTok, YouTube Shorts, and Instagram Reels. Blogs feed AI answer engines and long-tail SEO. Short-form video drives new audience discovery, brand awareness, and cheap distribution that ad platforms no longer provide affordably.

How will Trump’s tariffs affect a small ecommerce business?

Trump-era tariffs on Chinese goods raise the landed cost of imported products by a percentage of that landed cost, not a percentage of retail. A 25 percent tariff on a $1 landed-cost product adds $0.25 to your COGS. For most private-label brands with 3x to 5x markup, the impact on retail price is single-digit percent, not scary. Amazon-only sellers with thin margins feel it more than DTC brands.

Are Amazon sellers in trouble in 2025?

Amazon sellers are more exposed in 2025 because Amazon Haul (Amazon’s Temu clone) is undercutting the same SKUs at 40 to 50 percent lower prices, Chinese sellers are flooding listings with alphabet-soup brand names, and margins were already thin. The path forward for Amazon sellers is either build a real brand outside Amazon or move upmarket into products that cannot be commoditized.

What ecommerce products are hardest to knock off?

The ecommerce products hardest to knock off are personalized or made-to-order goods, products that require specialized equipment (like DTF or embroidery), and branded products with strong recognition. Bumblebee Linens is leaning into digital printing and handwriting embroidery because the customer’s own handwriting cannot be replicated overseas at scale.

Does TikTok Shop actually work for ecommerce brands?

TikTok Shop works for ecommerce brands with interesting, visually demonstrable products under about $50. Free organic reach through affiliates and micro-influencers is where the leverage sits. Brands with commodity products or products that require lots of explanation see weaker results.

How do I do the pillar video strategy for my niche?

To do the pillar video strategy for your niche, make one overview video covering the full topic (a full Thanksgiving dinner, a full room remodel, a full morning routine), then split each sub-step into its own standalone short-form video. Link the shorts back to the pillar in captions and descriptions. Every short doubles as a shareable “here is exactly how to do this step” clip, and the pillar remains the discoverable overview.

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: 41:43 — 48.0MB)

DeepSeek is a free, open-source AI model out of China that benchmarks competitively with paid tools like ChatGPT and Claude while reportedly using a fraction of the compute to train, which is why Nvidia’s stock dropped roughly 16 percent the day the news broke. The catch is data: everything you type into the hosted DeepSeek app is subject to Chinese terms of service, which allow much broader collection than OpenAI’s default settings.

If you download the open-source model and run it on your own PC (roughly a $3,000 hardware setup), no data leaves your machine. If you use their app or web interface, treat every prompt as public.

In this episode, my co-host Toni Herrbach and I unpack DeepSeek, Alibaba’s new Qwen release, and the wider “free Chinese AI vs paid US AI” split. We also cover what each major model is actually best for in real work: ChatGPT as the reliable general Google-replacement, Claude for creative writing and coding, Gemini for free image work, Llama for embedded use, and Grok as the outlier most people can skip.

Below we cover what DeepSeek actually is, the privacy tradeoff nobody wants to hear about, whether DeepSeek will get banned like TikTok, the model comparison for real use cases, and how AI companies will eventually monetize the free tier.

Key takeaways

• DeepSeek is a free, open-source Chinese AI model that reportedly beats ChatGPT on benchmarks and triggered a roughly 16 percent Nvidia stock drop the day it launched.
• Using DeepSeek’s hosted app sends every prompt through Chinese servers under Chinese terms of service. Running the open-source model locally sends nothing.
• Alibaba released a competitive AI model (Qwen) roughly a week later. Both are free. Expect more of the same from Chinese firms.
• DeepSeek reportedly censors sensitive Chinese political topics (like Tiananmen Square), a warning sign for anyone worried about tailored answers.
• ChatGPT is best for general questions, math, and formula help. Claude is best for creative writing, scripts, and coding. Gemini is a solid free image tool. Llama (Meta) is embedded in Facebook and WhatsApp for free.
• The real risk is not individual data; it is aggregate data. When 50 million users’ prompts feed one collector, patterns become intelligence.
• The free AI monetization endgame is almost certainly ads inside answers plus paid brand placement, just like Google did.

What is DeepSeek and why did Nvidia stock crash?

DeepSeek is a Chinese open-source AI model that scored better than or equal to OpenAI’s ChatGPT on several benchmarks while reportedly using roughly 1/10th to 1/20th the compute to train. Nvidia’s stock fell about 16 percent on the news because investors read “less compute needed” as “fewer chip sales ahead.” The model is free to use in DeepSeek’s app or web interface, and the underlying weights are fully open source, so anyone with the right hardware can download the model and run it privately.

The “less compute” claim is contested. Multiple people in the industry think DeepSeek trained on top of OpenAI outputs, or has an undisclosed stockpile of Nvidia GPUs, or tuned the model specifically to pass the benchmarks used in the announcement.

Benchmark-gaming has a long history in hardware and software (Nvidia and AMD have both been caught tuning drivers for known benchmark names). None of this is confirmed, and the “trained on a shoestring” narrative should be treated as unverified.

The DeepSeek privacy question (why using the app is different from downloading the model)

Using DeepSeek’s hosted app or website means Chinese servers see every prompt you send, and their terms of service permit broad collection and use of that data. Downloading the open-source model and running it locally on your own PC means nothing leaves your machine. Those are two completely different products with the same name.

The hosted version asks for name, email, and other account info. Once you are in, everything you type is stored server-side under Chinese terms.

The self-hosted version requires roughly $3,000 in hardware (a solid GPU plus supporting components) and enough technical willingness to run a local model. Realistically, almost nobody in the general public will do this, which is exactly the point.

Why aggregate AI data matters more than individual data

Aggregate AI data matters more than individual data because any single user’s queries are usually unremarkable, but 50 million users’ queries reveal genuine intelligence: what people are worried about, what they are buying, what they are curious about, and where the country’s soft spots are. The national security concern with DeepSeek is not that Chinese authorities care what I ask about my hankerchief inventory. It is what all US prompts, aggregated, tell them.

The same logic applies to the tailored-answer risk. Once a foreign model has hundreds of millions of users, its owners can subtly shape answers on politically sensitive topics.

When a user typed Tiananmen Square into DeepSeek early on, the model reportedly returned nothing. That is a warning signal (well short of proof) and it is worth taking seriously.

Will DeepSeek get banned in the US like TikTok?

DeepSeek will probably face government scrutiny similar to TikTok, and there is a reasonable chance the hosted app gets restricted for federal employees, contractors, and possibly the general public if the current administration continues its stance on Chinese tech. The open-source model is much harder to ban because the weights are already downloaded and mirrored globally. Anyone who wants to run it locally already can.

The awkward truth is that Chinese AI models are arguably more invasive than TikTok because users voluntarily type their most personal questions (medical results, financial situations, private worries) into them. TikTok collected behavioral data. AI models collect the internal monologue.

DeepSeek vs ChatGPT vs Claude vs Gemini: what to use for what

DeepSeek vs ChatGPT vs Claude vs Gemini comes down to what you are trying to do, not which one “won” the benchmark race. Benchmarks measure narrow tasks that are easy to game.

Real work requires picking the tool that handles your job best. Here is where each tool actually earns its subscription.

Comparison table: which AI model to use for which job

Which one I actually pay for

I pay for ChatGPT ($20/month) because it is my daily general-use tool: “how do I write this Google Sheets formula,” “give me the tax math on this scenario,” “help me set up this AWS user permission.” It is reliable and rarely sends me down a rabbit hole. I also use Claude, and I am about to switch my paid subscription over now that I am doing more scripting for video, because Claude produces scripts that stay closer to the source content and its Canvas coding view is genuinely useful.

I recently cancelled Midjourney because free Gemini image generation is now good enough for the YouTube thumbnails I need. As long as I do not need hands, it works.

What Claude and ChatGPT are actually good at (from real use)

Claude beats ChatGPT for creative writing, script generation, and coding tasks that need a live preview. ChatGPT is faster and clearer for math, formula help, and quick factual lookups where “reliable and slightly boring” wins.

Toni uses Claude for a video-editor contract and a launch-doc consolidation that would have taken her hours to draft manually. Her mother uses ChatGPT to draft scientific papers because it makes strong grammar patterns easy to apply.

Both tools respond very differently based on prompt style. Toni finds ChatGPT more sensational on titles; I find Claude more sensational.

That is largely because I prompt with “conversational tone” and she does not. Prompt style matters as much as model choice.

The killer everyday use case: pre-doctor visit

The killer everyday AI use case Toni’s brother uses is pasting medical test results into ChatGPT or Claude to generate context and questions before a doctor visit. The AI can flag which values are actually concerning for your age and history, and it produces a strong list of educated questions for a 15-minute appointment. This is one of the highest-leverage personal use cases for AI right now, and it costs nothing.

The privacy tradeoff on medical data is real. Weigh it before pasting sensitive records into any model.

Should you use DeepSeek or stick with paid US models?

Use DeepSeek only if you are running the model locally on your own hardware and the privacy question actually matters to your workflow. Otherwise, the $20-per-month cost of ChatGPT or Claude is trivial for most professional users compared to the data exposure risk of the hosted DeepSeek app.

For an average shopper, employee, or student, the pull of “free and as good as ChatGPT” is enormous. Expect DeepSeek and Alibaba’s Qwen to pick up massive US user counts in the next few months regardless of the privacy questions.

How free AI tools will eventually make money (and what you give up)

Free AI tools will eventually monetize the same way Google did: ads inside answers, sponsored brand placements, and API sales to third-party apps. There is no such thing as a durable “free forever” AI at scale because the compute cost is enormous.

Someone has to pay. Either the users pay in cash or they pay in data plus ads.

For merchants, that is actually a big opportunity. If tomorrow’s AI answer knows a user is planning a wedding, buying a car, or researching a supplement, brands will bid to appear inside that answer the way they bid on Google keywords today.

Klaviyo already shows what rich behavioral data enables for ecommerce. AI-scale data is that on steroids.

What to do this week if you are new to AI

If you are new to AI, start with the free version of ChatGPT, ask it three questions you would normally Google, then ask it one thing you have been putting off (a formula, a summary, a first draft of an email you dread). Pay attention to what feels faster or higher quality than your old workflow.

Add Claude for creative or script work when the ChatGPT output feels too generic. Consider Gemini for free image generation.

Skip the free Chinese models until the privacy picture is clearer, unless you are willing to self-host and know exactly what that means.

Frequently asked questions

Is DeepSeek safe to use?

DeepSeek is safe to use if you run the open-source model locally on your own hardware, because no prompts leave your machine. Using DeepSeek’s hosted app or website sends every prompt through Chinese servers under Chinese terms of service, which allow broad collection and use of that data. For sensitive queries (medical, financial, business), use a US-based model with a clear privacy policy.

Is DeepSeek really better than ChatGPT?

DeepSeek scores comparably to or better than ChatGPT on several public benchmarks but benchmarks are easy to tune for and do not always match real-world use. Multiple industry insiders believe DeepSeek was trained on top of OpenAI outputs and specifically optimized to pass the benchmark tests it announced. For a typical user’s day-to-day work (writing, coding, questions), ChatGPT and Claude remain highly competitive.

Will DeepSeek get banned in the US?

DeepSeek’s hosted app could face restrictions similar to TikTok, particularly for government employees, contractors, and possibly wider consumer use if regulators view its data practices as a national security issue. The open-source model itself is essentially impossible to ban because the weights have already been downloaded and mirrored globally by researchers and developers.

What is the best free AI tool right now?

The best free AI tools right now are ChatGPT’s free tier for general use, Claude’s free tier for creative writing and coding (with tighter usage limits), Gemini’s free tier for image generation and Google Workspace integration, and Meta’s Llama built into Facebook and WhatsApp. DeepSeek and Alibaba Qwen are also free but carry the Chinese-terms-of-service tradeoff.

What is the difference between hosted DeepSeek and self-hosted DeepSeek?

Hosted DeepSeek runs on DeepSeek’s servers in China and stores every prompt you send under Chinese terms of service. Self-hosted DeepSeek runs on your own PC or server, requires roughly $3,000 in hardware (mainly a strong GPU), and sends no data anywhere. Same model, completely different privacy profile.

Which AI tool should I pay for if I can only pick one?

If you can only pay for one AI tool, pick ChatGPT ($20/month) for general use or Claude ($20/month) for heavy writing, scripting, or coding work. Both are considerably better than the free tiers for high-volume users and both have clearer privacy protections than the free Chinese options.

What is Alibaba Qwen and how does it compare to DeepSeek?

Alibaba Qwen is Alibaba’s competitive AI model, released roughly a week after DeepSeek made global news, that reportedly benchmarks on par with or better than DeepSeek and OpenAI models. It is Chinese-owned and hosted, with the same data-privacy considerations as DeepSeek’s hosted app. It is not as clearly open-source as DeepSeek.

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: 13:52 — 16.1MB)

Amazon’s new FBA reimbursement policy, effective March 10, 2025, changes how the platform pays sellers back for lost or damaged inventory: instead of the retail selling price, Amazon will reimburse only the manufacturing cost, and only Amazon’s own estimate of that cost unless you hand over your supplier invoices. On this solo episode of the My Wife Quit Her Job podcast, I walked through what the policy actually says, what it will cost you per unit, and why it creates a dangerous incentive for Amazon to lose your inventory on purpose.

The math is brutal. If you sell a handkerchief for $10 that cost $2 to manufacture and Amazon loses 100 units, the old policy paid you $1,000. The new policy pays you $2 per unit, minus shipping, customs, and prep fees that Amazon refuses to count, so you actually see about $1.40 per unit.

Below is what changed, the AWD warehouse disaster that made 2024 the worst FBA year in memory, and the two-part game plan for sellers who want to survive the squeeze in 2025.

Key takeaways

• Starting March 10, 2025, Amazon reimburses lost or damaged FBA inventory at the manufacturing cost of your goods instead of the retail selling price. Shipping, handling, customs duties, and prep fees are all excluded.
• Amazon sellers lose approximately 1 to 3 percent of annual revenue to inventory discrepancies each year, according to Getida. Under the old policy this was painful; under the new one it is a direct hit to your cost of goods.
• You have two options for how Amazon determines your cost: accept Amazon’s own “comparable-product” estimate, or submit your actual invoices, which hands Amazon full visibility into your suppliers and margins.
• Amazon Warehousing and Distribution (AWD) had a disastrous 2024 holiday season, running out of capacity and stalling inventory transfers into fulfillment centers. Amazon still raised AWD prices for 2025 right after promising no new FBA fees.
• Amazon has a documented pattern of launching private-label knockoffs of top-selling FBA products at ~30 percent below the seller’s price. Handing over your supplier list accelerates that risk.
• Only high-margin sellers will survive long term. Every mid-margin FBA business needs a serious diversification plan (DTC, Shopify, wholesale, other marketplaces) starting now.

What is Amazon’s new FBA reimbursement policy?

Amazon’s new FBA reimbursement policy, effective March 10, 2025, reimburses sellers for lost or damaged inventory based on the manufacturing cost of the product rather than the retail selling price. Amazon’s stated goal is “greater transparency and more predictability” in how reimbursements are calculated, but the practical result is a large drop in what sellers get paid when Amazon loses their inventory.

Under the old policy, Amazon paid the full retail selling price of the lost unit. Under the new policy, Amazon pays only what it costs you to source the product from a manufacturer, wholesaler, or reseller.

Shipping, handling, customs duties, and prep fees are explicitly excluded, even though they are a real part of your landed cost.

How much money will Amazon sellers actually lose under the new policy?

Under the new policy, the average FBA seller loses roughly 70 to 90 percent of the previous reimbursement amount on a lost unit. A concrete example: if you sell a handkerchief for $10 with a $2 manufacturing cost and 30 percent shipping and customs on top ($0.60), Amazon used to pay $10 per lost unit and will now pay $1.40. Shipping, customs, and prep are all excluded from the “manufacturing cost” definition.

On 100 lost units, that is $1,000 under the old policy versus $140 under the new one. On 500 units (the size of a shipment Amazon has lost on my account before), the delta is $5,000 versus $700.

Amazon sellers lose approximately 1 to 3 percent of annual revenue to inventory discrepancies each year, according to Getida. On a $1 million per year FBA business, that is $10,000 to $30,000 in revenue at risk from lost, damaged, or missing inventory alone. The reimbursement gap now hits your bottom line directly.

How does Amazon determine “manufacturing cost” under the new policy?

Amazon offers two options for determining your manufacturing cost: Amazon can generate its own estimate based on comparable products sold on the platform and through wholesale channels, or you can submit your actual manufacturing invoices. Amazon’s estimate is almost certainly going to be lower than your real cost, because the comparable-product data set includes lower-priced competitors and Amazon’s own private-label pricing.

The second option, submitting your real invoices, means handing Amazon your supplier information, unit costs, and negotiated pricing. That is exactly the data Amazon needs to bypass you and source directly from your manufacturer.

Neither option is good. One underpays you; the other underpays you and gives Amazon the blueprint for a private-label knockoff.

Why is the AWD warehouse debacle relevant to this policy change?

The AWD debacle is relevant because it shows Amazon’s operational reliability is already failing, and the new reimbursement policy shifts the entire financial cost of that failure onto sellers. Amazon Warehousing and Distribution (AWD) had a disastrous 2024 Q4: capacity blew up, Amazon ran out of warehouse space, pickup schedules got delayed or canceled, and inventory sat stalled instead of moving into fulfillment centers in time for the holidays.

Many friends of mine on AWD ended up stocked out during the biggest shopping season of the year through no fault of their own. If the new reimbursement policy had been active during that mess, sellers would have absorbed the full inventory loss at manufacturing cost, not retail.

Amazon then raised AWD prices for 2025, right after publicly promising there would be no new FBA fee increases. The fee promise was technically kept; the reimbursement policy change and the AWD hike more than offset it.

Why does the new policy create a conflict of interest for Amazon?

The new policy creates a serious conflict of interest because Amazon now stands to profit from losing your inventory. When Amazon loses 10,000 units of your product and reimburses you at manufacturing cost with no shipping or duties, Amazon has 10,000 free units of a proven best-seller with no import cost of its own, which it could theoretically list under its own brand at a lower price.

That is not hypothetical. In 10 years of selling on Amazon, I have had products shipped under the wrong listing, customers receiving items my store does not sell (a weighted vest returned instead of our linen towels this past Christmas), and hundreds of units Amazon claimed were lost later reappearing for sale under my own listing from other sellers.

Under the old policy, Amazon lost money on operational mistakes and had a financial reason to fix them. Under the new one, Amazon’s cost of losing inventory drops by 70 to 90 percent, which removes the internal pressure to run a tight ship.

Old vs new FBA reimbursement policy: what actually changed

How does Amazon’s private-label copycat pattern make this worse?

Amazon’s private-label pattern makes this worse because handing over your supplier and cost data (the only way to get an accurate reimbursement) gives Amazon everything it needs to knock you off. Amazon has a long documented history of launching its own private-label version of top-selling third-party products at roughly 30 percent below the seller’s price.

A friend of mine was selling hundreds of thousands of dollars of Emu oil per year on Amazon. Amazon launched its own version, priced it 30 percent lower, and then advertised its branded Emu oil directly on her product listing page. Her sales tanked overnight.

Mike Jackness used to sell gel packs on Amazon at millions of dollars per year. Amazon knocked off the product at 30 percent cheaper and even copied his packaging and photo layouts.

Once Amazon has your supplier, unit cost, and margin data, they can go direct to your manufacturer, negotiate a better price with their volume, and take the market. The reimbursement policy is one more push in that direction.

What should Amazon FBA sellers do to protect their business in 2025?

The action plan is a two-part hedge: cut your Amazon exposure where you can and build alternate channels immediately. Diversification is no longer optional for anyone whose Amazon margin cannot absorb a permanent step-down in reimbursement recovery.

• Model the new reimbursement math on your top 20 SKUs and decide which ones are still viable at Amazon’s new economics.
• Do not casually opt into “let Amazon estimate my manufacturing cost” without checking the estimate against your actual invoices. Amazon’s estimate will almost always be lower.
• Think hard before submitting supplier invoices for accurate reimbursement. The margin recovery on lost units may not be worth handing Amazon your sourcing playbook.
• Aggressively grow a DTC or Shopify store you control, plus a second marketplace (Walmart, TikTok Shop, or wholesale via Faire).
• Track your discrepancy rate obsessively. Consider reimbursement-recovery tools like Getida to file every eligible claim, because the smaller per-unit payout means volume matters more than ever.
• If you are still doing FBA-only, treat this policy as the wake-up call. Amazon’s advertising costs already climbed roughly 20 percent year over year, inbound placement fees appeared this year, and AWD price hikes are baked into 2025.

What else is squeezing Amazon sellers right now?

Beyond the reimbursement change, Amazon sellers are getting hit from multiple directions in 2025. Advertising costs on Amazon rose approximately 20 percent year over year. New inbound placement fees charge sellers per unit just to accept inventory into the fulfillment network.

Amazon Haul, Amazon’s answer to Temu and Shein, is a dropshipping-style marketplace that directly undercuts existing FBA sellers on price. Temu and Shein themselves sell direct from China to US consumers without paying most import duties, which puts a permanent price ceiling above many FBA product categories.

Meanwhile Prime delivery reliability has slipped noticeably. This past holiday I received about 50 percent of my Prime orders inside the promised delivery window. Sellers pay for the Prime badge on the assumption Amazon is holding up its end, and that end is fraying.

Frequently asked questions

When does Amazon’s new FBA reimbursement policy take effect?

The new FBA inventory reimbursement policy takes effect on March 10, 2025. From that date, Amazon will reimburse sellers based on the manufacturing cost of lost or damaged items rather than the retail selling price. The policy was announced in late 2024 and applies to all FBA sellers.

How much less will sellers get paid under the new Amazon reimbursement policy?

Most sellers will receive 70 to 90 percent less per lost unit than they did under the old policy. On a $10 product with a $2 manufacturing cost, the old policy paid $10 per lost unit and the new policy pays about $1.40 after shipping, customs, and prep are excluded from the definition of manufacturing cost.

Should you send Amazon your supplier invoices to get accurate reimbursements?

Submitting your actual supplier invoices does get you an accurate reimbursement, but it gives Amazon full visibility into your suppliers, negotiated unit costs, and gross margins. Given Amazon’s documented history of launching private-label knockoffs of successful third-party products, most sellers should think carefully before handing over that data.

Are Amazon warehouse fees going up in 2025?

Amazon Warehousing and Distribution (AWD) fees are increasing in 2025 despite Amazon’s earlier claim that FBA fees would not rise. Amazon Haul, new inbound placement fees, and advertising cost inflation of roughly 20 percent year over year are also stacking pressure on seller margins.

What is Amazon Warehousing and Distribution (AWD) and why is it failing?

AWD is Amazon’s upstream storage service, positioned as a way to hold inventory before transferring it into FBA fulfillment centers. During the 2024 Q4 holiday season it failed at capacity, with delayed pickups, canceled pickups, stalled inventory transfers, and sellers running out of stock during peak season through no fault of their own.

Should you stop selling on Amazon FBA in 2025?

Stopping Amazon FBA entirely is rarely the right move, but every FBA seller needs a serious diversification plan in 2025. High-margin sellers can absorb the new reimbursement math; mid-margin sellers will struggle. The safest path is to keep your Amazon channel running while aggressively building a DTC store and a second marketplace (Walmart, TikTok Shop, or wholesale) that you control more directly.

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: 50:35 — 58.2MB)

The biggest winners of 2024 were Nvidia, Elon Musk, anything AI-adjacent, Walmart, YouTube and TikTok creators, and breakout personalities like Caitlin Clark and Simone Biles. The biggest losers were Amazon FBA sellers who only sell on Amazon, SEO-only blogs, chain restaurants like Red Lobster and TGI Fridays, department stores like Macy’s (which closed 150 stores), the NBA (down about 58% in ratings), X/Twitter, and anyone who paid $180 to Netflix to watch the Mike Tyson vs. Jake Paul fight stream buffer for three rounds.

This is our first annual awards episode, where my co-host Toni Herrbach and I picked the best investments, worst investments, best products, biggest life hacks, and biggest personal changes of 2024, plus who crushed it and who crashed in the broader economy and ecommerce world. The through-line across almost every category is the same: brands and creators who stayed tuned to what their audience actually wanted won, and the ones that were tone-deaf lost.

Below we cover our best and worst investments of 2024, the best products and life hacks we adopted, the biggest personal changes we are making in 2025, and the biggest ecommerce winners and losers you should actually be paying attention to.

Key takeaways

• Amazon FBA sellers who only sell on Amazon were the biggest ecommerce losers of 2024. Amazon rolled out new inbound placement, low-inventory, and returns fees, plus AWD (Amazon Warehousing and Distribution) had a disastrous holiday season.
• Nvidia was one of the best stock investments of 2024. Steve owned it for a decade before it went vertical, on the theory (from being a gamer) that Nvidia dominated GPUs.
• High-yield savings accounts, including business accounts like Highbeam, were an underused everyday investment. Rates were as high as 4.5% before cuts.
• The best AI life hacks of the year: ChatGPT and Claude for coding and writing (Steve retired his human writers), ElevenLabs voice cloning, and SearchGPT replacing Google for research.
• Biggest ecommerce winners of 2024: Nvidia, AI-adjacent companies, Amazon the corporation (not Amazon sellers), Walmart, fast-casual restaurants like Cava and Shake Shack, and creators building on YouTube and TikTok.
• Biggest ecommerce losers: pure-SEO blogs (killed by Google AI Overviews), chain restaurants, department stores, X/Twitter, and anyone who bought a car or overpriced real estate at 2024 rates.
• The one big shift Steve made in 2024: automating custom-print product creation with a ChatGPT-generated script, so he can target every occasion Hallmark-style at scale.

Best investments of 2024

The best investments of 2024 for us were Nvidia stock (up massively for the year), high-yield savings accounts like Highbeam for business cash, and personal-growth spending like better walking shoes and paid podcast subscriptions. Financial returns and personal-growth returns both count.

Steve has owned Nvidia for about a decade because he was a heavy PC gamer and knew the company dominated GPUs. That thesis paid off well before AI took over, and 2024 was the year everyone else caught up.

Toni’s biggest win was moving money out of a low-interest checking account into a high-yield savings account. On the business side, Highbeam (a bank built for ecommerce) let cash coming in and out of the business earn interest, sometimes as high as 4.5% before the rate cuts.

The overlooked category is personal growth. Buying better walking shoes, subscribing to more podcasts, and putting money into small quality-of-life upgrades produced the best return-on-investment of the year for Toni. Cheap in dollars, big in daily effect.

Worst investments of 2024

The worst investment of 2024 was any luxury European car whose warranty ran out (Steve’s BMW), plus buying a car or overpriced real estate in a market with sky-high interest rates. Toni’s best move was the car she almost bought and didn’t.

Steve’s BMW hit about 50,000 miles, had a catastrophic timing-chain failure right after a $2,500 repair, and BMW refused to help. There was a settled lawsuit on that exact model, but BMW is not legally required to notify owners of settled defect lawsuits. Steve had his son in the car when it started failing.

Toni almost bought a car in 2024 and is glad she didn’t. Interest rates were sky-high, prices were volatile, and used-car values had further to fall. The best investment can be the one you skip.

Real estate in 2024 was a mixed bag. If you had to buy, you probably paid too much at the front of the year. If you sold and downsized (Toni sold her house at the top of the Florida market), it was one of the best financial moves of the year.

Biggest personal wins of 2024

The biggest personal wins of 2024 in our world were selling a house at the top of the Florida market (Toni), automating product-listing creation with an AI script (Steve), and moving family members closer to home (both of us). One is a huge financial win, one is a productivity multiplier, one is a life-satisfaction win.

Toni watched her local Florida market for six years and sold her big, expensive-to-operate house right as it flipped from a seller’s market to a buyer’s market. She then rolled the proceeds into another renovation project. Selling at the top made a large amount of money and cut a stack of ongoing operating expenses.

Steve’s biggest win was a script that automates custom-print product creation for his ecommerce store, Bumblebee Linens. Before the script, every new printed product needed manual Photoshop mockups, a written description, a title, and an upload. With the script, ChatGPT writes the description and title, the image gets Photoshopped onto the three products they print on, and it all uploads automatically. The end goal is to have a printed product for every occasion Hallmark ever invented, because Hallmark literally invented most of those occasions and built a business on it.

The other big win was moving Steve’s mom into his neighborhood, and doing it at a good price after a rainy open-house week and a listing agent who mispriced by $500,000 kept other buyers away.

Coolest moments of 2024

The coolest moments of 2024 for us were bucket-list travel and one bucket-list sporting event. Steve went to the Super Bowl (Niners vs. Chiefs) and the Paris Olympics. Toni was in Budapest for New Year’s Eve, skiing in Austria for her birthday, and in Denmark during Christmas.

Steve had wanted to go to a Super Bowl for years and finally justified the ticket cost because the game was one of the best in recent memory. Paris was his first trip to the city, and beach volleyball under the Eiffel Tower was worth the trip on its own. Olympic basketball was the letdown of the trip because the U.S. only sends its top NBA stars, so every other team plays without theirs and the level drops.

Toni’s happiest coolest moment was less flashy: sitting in an outdoor Christmas-market cafe in Denmark with heaters, heated seats, and blankets, watching people. Those are the moments you cannot put a price tag on.

Steve’s other coolest moment was smaller and closer to home: his daughter made the honor roll for the first time in her life, in high school, after years of struggling. That one you cannot buy tickets to.

Best life hacks of 2024 (AI, voice, and morning routines)

The best life hacks of 2024 for us were using ChatGPT and Claude to write code and long-form content, cloning your own voice in ElevenLabs, a phone-free morning routine with five minutes of guided breathing, and following The Points Guy for travel-hack email tips.

Steve used ChatGPT to code a loyalty program for Bumblebee Linens and a two-step spin-to-win pop-up. The new pop-up increased his email opt-in conversion rate by roughly 3x to 3.5x. He also uses Claude specifically for creative writing, where it beats other tools by a wide margin.

He cloned his voice in ElevenLabs so his overseas assistant can type in any script and get a voiceover that sounds like him, which is only useful if you make content, but if you do it is transformative. The newest ElevenLabs model handles intonation and pacing well enough that the clones are hard to distinguish from the real voice.

Toni’s non-AI life hack is the morning routine. Get out of bed, do not check your phone, walk in the sun, and do five minutes of guided breathing. She was a breathing-work skeptic 18 months ago and it changed her mornings. Her other life hack is following The Points Guy for travel-hacking emails, which is how she has flown almost free to most of her trips.

Best products of 2024

The best products of 2024 for us were noise-canceling insulation (yes, really), Bose open-ear hook-around headphones, and SearchGPT as a full replacement for Google. The two physical products are personal; the third one changed our workflow.

Toni’s construction-related best products came from renovating her house. Sound-dampening insulation (rolled wool-style batting) was the standout: cheap-ish, easy to install, and huge quality-of-life gain if you are building or remodeling.

The Bose Ultra Open Earbuds hook around the back of your outer ear and do not go into the canal at all. That means you can walk, run, cycle, and still hear the world around you. For anyone who dislikes in-canal earbuds like AirPods, they are a genuinely different form factor.

Steve’s best product of the year was SearchGPT, OpenAI’s AI-search product with citations. It has replaced Google entirely for him. Perplexity is also strong, but SearchGPT’s answer format has less clutter and better source citations for his workflow.

Biggest personal changes for 2025

The biggest personal changes we are making for 2025 are exercising five days a week (Steve, using Nintendo’s Ring Fit as one of the workouts) and shifting toward more networking and in-person events (Toni). One is a health system, one is a professional-growth system.

Steve used to play ultimate frisbee weekly, but the group got old, people started getting injured, and the game died. He now plays tennis once a week and does the Ring Fit exercise game on Nintendo Switch on the other days. The goal is five workout days a week, mixed across tennis, weights, and Ring Fit. Gamifying exercise works when nothing else does.

Toni wants to shift, not overhaul. Small tectonic-plate-style movements that create outsized results. Her concrete moves: filling out a wall calendar in one session with a friend for accountability, attending more events, and getting out of the house more (she is a natural introvert who will not leave for weeks if left alone). Steve is going the opposite direction because his kids are still at home and playing volleyball four to five nights a week.

Biggest ecommerce losers of 2024

The biggest ecommerce losers of 2024 were Amazon FBA sellers who only sell on Amazon, SEO-only blogs killed by Google AI Overviews, chain restaurants like Red Lobster and TGI Fridays, department stores like Macy’s, X/Twitter, Aaron Rodgers as a personal brand, and the NBA. The common thread across almost all of them is being tone-deaf to what the audience actually wants right now.

Amazon FBA sellers took the biggest hit of any group we work with. Amazon introduced multiple new fees this year (inbound placement, low-inventory, returns), Amazon Warehousing and Distribution (AWD) failed most sellers during the holidays, and Amazon recently announced it will only reimburse the raw cost of goods (not customs duties or freight) when it damages your inventory. FBA fees now take about a 61% cut of seller revenue on average.

SEO-only blogs are the other quiet loser. Google’s AI Overviews now answer more queries directly, click-through rates on informational content have collapsed, and pure-SEO sites without an email list, community, or social presence lost significant traffic in 2024. This gets a whole post to itself later.

Chain restaurants (Red Lobster bankruptcy, TGI Fridays down), department stores (Macy’s closed 150 stores), and traditional indoor shopping malls all lost to more focused, better-run competitors. Fast-casual chains like Cava and Shake Shack ate their lunch. Outdoor plazas held up better than enclosed malls.

X/Twitter lost users in 2024. Whether that reverses in 2025 is debatable, and Steve is more bullish on X than Toni. And the biggest loser in pure entertainment terms was anyone who stayed up to watch the Mike Tyson vs. Jake Paul fight on Netflix, which buffered so badly that Netflix earned its own spot on the list.

Biggest ecommerce winners of 2024

The biggest ecommerce winners of 2024 were Nvidia, Elon Musk, anything AI-adjacent, Amazon the corporation, Walmart, fast-casual restaurants like Cava and Shake Shack, and creators on YouTube and TikTok. If you were building on YouTube or TikTok in 2024, you had the best distribution tailwind of any channel.

Nvidia and AI were the trades of the year, and Elon Musk went all-in on the election and it paid off in political access. Amazon the corporation is a winner (double the operating income year-over-year in Q2 alone) even though Amazon sellers are a loser. The distinction matters.

Walmart is coming hard at Amazon and is a genuine 2024 winner in ecommerce. Fast-casual restaurants held their price points while chain restaurants slashed and stumbled, which is why places like Cava and Shake Shack kept traffic while Red Lobster went bankrupt.

On the creator side, YouTube and TikTok both had huge years (with the caveat that TikTok’s U.S. future is uncertain). Personal winners include Caitlin Clark (who single-handedly raised WNBA revenue), Simone Biles, Ilona Maher (rugby, then leveraged into massive social followings), and pommel-horse guy Stephen Nedoroscik, who all turned Olympic moments into durable content businesses.

What the 2024 winners and losers actually teach ecommerce sellers

The through-line from every winner and loser of 2024 is the same. Winners stayed tuned to what their actual audience wanted and adjusted. Losers stayed tone-deaf and defended their old model. That is the entire lesson.

Amazon-only FBA sellers who stayed on-platform got squeezed on fees. Sellers who built a brand off-Amazon (a Shopify store, an email list, a YouTube channel) had a second lever to pull. Bumblebee Linens survived because most of the business comes from wedding and event planners via word of mouth, not from Amazon rankings.

Traditional chain restaurants raised prices into a squeezed consumer and hoped nobody noticed. Cava and Shake Shack held pricing and served food people actually wanted. The same test works for ecommerce: are you raising prices while your product experience is getting worse, or are you holding the line and improving?

The Caitlin Clark and Ilona Maher lesson is worth internalizing if you make content or run a brand. Both took one moment of attention and converted it into a durable content business (WNBA franchise value, social followings, Dancing with the Stars, brand deals). Most people get 15 minutes and let it evaporate. Turning attention into a system is the whole game.

Frequently asked questions

Who were the biggest ecommerce winners of 2024?

The biggest ecommerce winners of 2024 were Nvidia and anything AI-adjacent, Amazon the corporation (whose Q2 operating income roughly doubled year-over-year), Walmart (which continued to close the gap with Amazon), and content creators on YouTube and TikTok. Fast-casual restaurant chains also outperformed the traditional sit-down chains.

Who were the biggest ecommerce losers of 2024?

The biggest ecommerce losers of 2024 were Amazon FBA sellers who sold only on Amazon (hit by multiple new fees), SEO-only blogs (crushed by Google AI Overviews and lower click-throughs), chain restaurants like Red Lobster and TGI Fridays, and department stores like Macy’s, which closed 150 stores. The NBA had a rough year in ratings and X/Twitter lost users, though X may rebound in 2025.

Why were 2024 Amazon FBA sellers the biggest ecommerce losers?

Amazon introduced multiple new seller fees in 2024, including inbound placement fees, low-inventory fees, and updated returns fees. Amazon Warehousing and Distribution (AWD), pitched as the fee-saving alternative, failed most sellers during the holiday season. On top of that, Amazon announced it will only reimburse the raw cost of goods (not customs duties or freight) when it damages seller inventory. FBA fees now consume roughly 61% of seller revenue on average.

Was Amazon a winner or loser in 2024?

Amazon the corporation was a big winner in 2024, with Q2 operating income roughly doubling year-over-year to $14.7 billion. Amazon third-party sellers, on the other hand, were the biggest ecommerce loser group of the year, since almost all of Amazon’s profit growth came from higher seller fees and advertising revenue, not from actual ecommerce sales growth (which was up only about 4.6%).

What was the best AI tool of 2024 for ecommerce?

The best AI tools of 2024 for ecommerce work were ChatGPT (or Claude) for coding and content, SearchGPT to replace Google for research, and ElevenLabs for voice cloning if you make video or podcast content. Using ChatGPT to code a Shopify or Klaviyo automation (like a new opt-in pop-up) can lift conversion rates by 3x or more without hiring a developer.

Are chain restaurants really dying in 2024?

Traditional sit-down chain restaurants like Red Lobster and TGI Fridays had a very rough 2024. Red Lobster filed for bankruptcy and installed a new CEO to restructure, and TGI Fridays closed multiple locations. The chains that thrived were fast-casual concepts like Cava and Shake Shack, which held their price points and served food people wanted, while traditional chains raised prices as service quality declined.

What is the biggest lesson from the 2024 winners and losers?

The biggest lesson from the 2024 winners and losers is that being tuned in to what your customer actually wants beats every other strategy. Winners (Walmart, Cava, Caitlin Clark, YouTube creators) adjusted quickly to changing demand. Losers (Amazon-only FBA sellers, chain restaurants, X/Twitter, Aaron Rodgers) either stayed defensive or ignored the shift. For ecommerce sellers, the practical takeaway is to build off-Amazon channels (email, social, brand) so you have leverage when a platform decides to raise fees.

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: 13:14 — 15.4MB)

Trump’s proposed tariffs (a 10% global tariff and a 60% tariff on Chinese goods) will hit Amazon FBA sellers, but nowhere near as hard as the headline numbers suggest. Even if the full 60% China tariff passes exactly as proposed and gets fully absorbed into your cost of goods, the actual retail price increase on Amazon is closer to 15% than 60% because of the standard 4x retail markup. In practice, the tariffs will most likely roll out in staged brackets (like Trump’s first term, where his 45% campaign promise became a mix of 7.5% and 25% by product category), consumer prices will rise about 1% to 15% depending on category, and Amazon FBA sellers who source from China will absorb some of the cost rather than pass all of it through, because Amazon rankings collapse with even a 60-cent price bump.

This post is my take (as an ecommerce seller who has sourced from China for 17 years and sold on Amazon since 2014) on why the tariff numbers everyone is panicking about are almost certainly overstated, what actually happened last time under Trump 1.0, and what Amazon FBA sellers should do right now before late January.

Below I cover Trump’s proposed tariff policies, what a 60% China tariff would really do to Amazon prices, why sourcing outside China is harder than most sellers realize, and the exact playbook for Amazon FBA sellers to prepare over the next four to eight weeks.

Key takeaways

• Trump has proposed a 10% global tariff and a 60% tariff on Chinese goods. Over 70% of the products sold by wholesalers and retailers on Amazon are made in China, so the exposure is real.
• JP Morgan chief U.S. economist Michael Feroli estimates a 60% China tariff would raise the U.S. price level by only about 1% if fully passed through, and less if sellers absorb the cost or shift sourcing.
• In Trump’s first term, his 45% campaign promise on Chinese goods was actually implemented in staged brackets from 7.5% to 25% depending on product category, with many exemptions. Expect the same pattern this time.
• Standard retail markup is roughly 4x cost. A 60% tariff on a $1 product raises the landed cost to $1.60, which at a 4x markup would sell for $6.40 versus $4. Almost no seller passes that full increase through on Amazon.
• Amazon rankings punish price hikes hard. A 60-cent bump can drop your listing from position 1 to position 10, so sellers usually absorb 60% to 100% of a tariff increase to hold the buy box.
• Trump’s proposed corporate tax cut from 21% to 20% or even 15% would offset most of the tariff hit for U.S.-based ecommerce businesses.
• Sourcing outside China is much harder than it sounds because of vertical integration. Many Mexican, Indian, and Southeast Asian suppliers still buy components from China. Chinese factories have also opened plants in Mexico to bypass tariffs directly.
• What to do right now: place your China orders before late January if you are worried, expect higher freight rates in the near term, and plan for a small margin hit over the next two to three quarters, not a business-ending catastrophe.

What are Trump’s proposed tariffs for his second term?

Trump has publicly proposed a 10% global tariff on every imported product regardless of origin, plus a 60% tariff on all products coming from China. That is the campaign promise. History strongly suggests the actual implemented tariffs will be smaller, staggered, and full of category exemptions.

For context on how much exposure Amazon FBA has here, over 70% of the products sold by wholesalers and retailers on Amazon.com today are manufactured in China. A pure 60% tariff on all of them would be a serious event.

The likelihood of that pure 60% tariff actually landing on all Chinese goods is low. Michael Feroli, JP Morgan’s chief U.S. economist, put the probability of a blanket 10% global tariff at relatively low and a blanket 60% China tariff even lower. Expect targeted brackets, not one flat rate.

What would a 60% China tariff actually do to Amazon prices?

A full 60% China tariff would raise the overall U.S. price level by about 1% if fully passed through to consumers, according to Michael Feroli. The effect could be even smaller if sellers absorb part of the cost into margins or substitute suppliers away from China.

That 1% headline number sounds too small. It is not, once you understand retail markups.

In a typical retail environment, products carry roughly a 4x markup. Source a product for $1, sell it for $4. Add a 60% tariff and the sourced cost rises to $1.60. At the same 4x markup, the new retail price would be $6.40, a 60% price hike. Almost no seller does this in practice.

The real-world math looks different. On Amazon, price elasticity is brutal. A 60-cent price increase on that $4 product can drop your listing from search position 1 to position 10, which crushes sales far more than the tariff crushes margin. Instead of raising the price to $6.40, most sellers hold the price at $4 and eat the extra $0.60 of cost. Profit drops from $3 to $2.40, which is a 20% profit hit that reads like a 15% effective discount on the product.

That is what actually happened during Trump’s first term. Most of our products at Bumblebee Linens fell into the 7.5% tariff bracket, not the 25% one, and we barely raised prices in our store at all.

Why the 60% China tariff will probably roll out in brackets

Trump’s proposed 60% tariff will almost certainly be implemented in staged brackets by product category rather than as a single flat rate. That is exactly what happened in his first term.

Trump 1.0 campaigned on a sweeping 45% global tariff on Chinese goods. What actually got implemented was tariffs from 7.5% to 25% depending on the product category, with numerous exemptions. Many everyday consumer goods landed in the low 7.5% bracket. The rhetoric was 45%; the reality was much softer.

There is a logical reason for the bracket approach. It makes no policy sense to tariff products the U.S. does not manufacture at all, because you cannot substitute domestic production for something no one here makes. The highest tariffs will probably target industries where the U.S. is actively rebuilding domestic capacity, like automotive, semiconductors, and consumer electronics.

Everyday consumer goods, homewares, and low-tech commodity products will likely see much smaller tariff increases. Bold campaign numbers make for stronger political messaging than a spreadsheet of 47 category-specific brackets, so expect the announcement to sound bigger than the implementation.

How Trump’s corporate tax cuts offset the tariff hit

Trump’s proposed corporate tax cut (from 21% to 20% or even 15%) offsets most or all of the tariff impact for U.S.-based ecommerce businesses. That was the story of Trump 1.0: tariffs went up, but the corporate rate dropped from 35% to 21%, and small businesses like mine ended up making much more profit than before, not less.

For a typical Amazon FBA business with a 20% net margin, a further corporate rate cut from 21% to 15% is a large annual tax savings that goes straight to the bottom line. That savings alone can more than offset a small tariff-driven margin hit.

The right framing for FBA sellers is not “tariffs will kill me.” It is “my landed costs go up by X, my competitors go up by the same X, my corporate tax rate drops, and I can adjust prices modestly across the board.” When everyone’s costs move together, the competitive impact is much smaller.

Can Amazon FBA sellers source outside China to dodge tariffs?

Sourcing outside China (Mexico, India, or Southeast Asia) is a real strategy but is much harder than most tariff-panic articles imply, for four reasons: cost, vertical integration, quality lead time, and Chinese factories already bypassing tariffs through Mexico.

On raw cost, China is still cheaper than almost every alternative. Chinese labor costs are about 4x lower than U.S. labor. Even a 60% tariff often leaves China cheaper on landed cost than Mexico or the U.S. for most product categories.

China’s vertical integration is the underrated moat. Most products are assemblies of many components, and China has suppliers for virtually every component in close geographic proximity. Take the iPhone: even a U.S.-assembled version still needs chips from TSMC in Taiwan, RAM from Korea, and PCB components from China. Replicating that supply-chain density elsewhere takes decades, not one presidential term.

We ran into this exact problem when we sourced intricate lace handkerchiefs from an Italian supplier. The materials still came from China. We were paying an Italian markup for the same Chinese inputs. This is common in “Made in Mexico” and “Made in India” too.

And Chinese-owned factories have already set up operations in Mexico specifically to sell into the U.S. tariff-free, since Mexico has no U.S. tariffs today. Chinese manufacturers are always three steps ahead of tariff enforcement.

What Amazon FBA sellers should do right now (before late January)

Amazon FBA sellers who source from China should place their orders now (before Trump takes office in late January) and get inventory into the U.S. as fast as possible. Expect freight rates to spike in the short term because every other seller is doing the same thing, then normalize once the initial rush ends.

Concrete steps for the next four to eight weeks:

• Place reorders now on your top-velocity SKUs to lock in pre-tariff pricing on 60 to 120 days of inventory.
• Get product into the U.S. before late January if possible, expecting freight rates to rise 10% to 30% in the short term.
• Model your worst case at a full 25% tariff on your specific product category, not the 60% headline. Most consumer categories historically fall in the 7.5% to 25% range.
• If your margin is under 15%, plan a modest price increase (roughly 5% to 10%) once competitors move first. Do not lead the price hike on Amazon.
• Look at Chinese factories with Mexico operations as a medium-term hedge. Do not switch primary sourcing on a rumor.

Longer term, tariffs actually give the U.S. some real negotiation leverage that has been missing for years. China has enjoyed structural advantages for a long time, and even imperfect tariffs will help rebalance parts of that.

What Trump’s presidency means for Amazon FBA overall

Trump’s second term is likely net neutral for Amazon FBA sellers, and possibly slightly positive. Tariffs add cost. Lower corporate taxes add profit. And ongoing FTC scrutiny of Amazon (which will almost certainly continue under Trump because of his well-known Bezos and Washington Post feud) probably forces Amazon to behave better toward sellers.

Amazon was already sued by the FTC in 2023 for illegally maintaining its monopoly. That case is proceeding. Trump has repeatedly called Amazon a monopoly, so the antitrust pressure is not going away. That is why Amazon suddenly announced no fee changes for 2025, which is the first good news for FBA sellers in years.

Net-net for FBA: small margin headwind from tariffs, small margin tailwind from taxes, and slightly better Amazon behavior because the FTC is watching. Not the doomsday scenario the media is selling.

Frequently asked questions

How much will Trump’s tariffs raise prices for U.S. consumers?

According to JP Morgan chief U.S. economist Michael Feroli, a full 60% tariff on Chinese goods would raise the overall U.S. price level by only about 1% if fully passed through to consumers, and possibly less if sellers absorb the cost or shift sourcing to other countries. Individual product categories with heavy China exposure will see larger increases (typically 5% to 15%), while categories with domestic production may barely move.

Will a 60% China tariff really increase Amazon prices by 60%?

No. A 60% tariff on the cost of goods does not translate to a 60% retail price increase because retail products carry roughly a 4x markup. A $1 sourced product with a 60% tariff has a new landed cost of $1.60, which at a 4x markup would price at $6.40, but almost no Amazon seller passes the full increase through because a 60-cent price bump can drop their listing from position 1 to position 10. Expect actual Amazon price increases in the 5% to 15% range on China-sourced products.

Should Amazon FBA sellers order more inventory before Trump takes office?

Yes, if you source from China and expect a tariff impact on your category. Placing reorders now on your top-velocity SKUs locks in pre-tariff pricing for 60 to 120 days of inventory and buys you time to see how the tariff brackets actually roll out. Expect freight rates to rise in the short term because most other sellers are doing the same thing.

Can Amazon FBA sellers just source from Mexico or India instead?

Sourcing outside China is possible but much harder than it sounds. China is still cheaper on raw cost (labor is about 4x lower than the U.S.), and China’s vertical integration means that many Mexican and Indian suppliers still buy their components from China. Chinese factories have also opened plants in Mexico to sell into the U.S. tariff-free, so “Made in Mexico” can still be Chinese-made economically.

Did Trump’s first-term tariffs actually hurt Amazon FBA sellers?

Trump’s first-term tariffs on China had a much smaller impact on Amazon FBA sellers than the media predicted. His 45% campaign promise landed as staged brackets from 7.5% to 25% by product category, with many exemptions. Combined with the corporate tax cut from 35% to 21%, most U.S.-based FBA sellers made more profit during his first term, not less.

Will Amazon Prime prices go up because of Trump’s tariffs?

Amazon Prime membership pricing is separate from tariffs and is driven by Amazon’s internal cost structure and pricing strategy. Tariffs would raise the cost of goods on Amazon’s third-party marketplace, which Amazon usually passes to sellers rather than absorbing. Amazon has announced no third-party seller fee changes for 2025, which is a first, likely driven by ongoing FTC pressure.

Is now a good time to start an Amazon FBA business with tariffs coming?

Yes, tariffs do not fundamentally change the case for starting an ecommerce business. Trump’s proposed corporate tax cuts (from 21% to potentially 15%) are one of the biggest reasons to be a business owner rather than a W-2 employee, because business owners get deductions and preferential rates that employees never see. The tariff hit is real but manageable, especially for categories with modest China exposure.

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: 45:12 — 52.0MB)

Word-of-mouth marketing is the single highest-leverage traffic channel for an ecommerce store because it costs almost nothing per new customer, converts far better than paid ads, and doubles as the signal AI search engines like Perplexity and SearchGPT now use to decide which brands to cite. Building it comes down to three things: designing a customer experience worth talking about, personally showing up for your best customers (real replies to emails, real gifts they actually want, real comments on their social posts), and putting yourself out on social media and podcasts so your brand becomes an entity that AI search and Google recognize.

My co-host Toni Herrbach and I walk through why word-of-mouth is now more important than SEO for ecommerce, using specific examples from Bumblebee Linens (my linen store), Toni’s local Italian restaurant that upsold her with a free half-pour of wine and a jar of salt, and Gary Vaynerchuk’s $400 autographed Jay Cutler jersey stunt that turned into a viral wine order.

Below we cover why people actually shop where they shop, how to build customer relationships that generate repeat sales without ads, the corporate gifting trick that most brands get wrong, the social media and personal-brand play that now drives Google and AI visibility, and the checkout tactic (asking for a customer’s Instagram handle) that turns one purchase into ongoing organic reach.

Key takeaways

• People shop for two reasons: consistency (Amazon) or customer experience (local boutique, direct-to-consumer brands). If you cannot beat Amazon on consistency, you win on customer experience.
• The bar for great customer service is embarrassingly low. Replying personally to customer emails is enough to make people say “I cannot believe a real person responded.”
• Corporate gifting works, but only if the gift is something the recipient actually wants. Do not send them a mug with your logo. That is an ad, not a gift.
• Gary Vaynerchuk spent $400 on an autographed Jay Cutler jersey for a small customer with a large Twitter following. One month later, the customer sent him a massive wine referral order.
• The most reliable modern word-of-mouth engine is social media. TikTok and YouTube are how people now find new restaurants, products, and destinations. Ignoring both is a strategic mistake for an ecommerce brand.
• AI search (SearchGPT, Perplexity) is starting to matter for ecommerce. Perplexity now has one-click checkout via ShopPay. Getting cited requires a strong brand entity, which requires you personally showing up on social, podcasts, and press.
• Steve’s biggest recent SEO win: linking every press mention on his personal site. Traffic to his blog doubled in one day and kept climbing, because Google now confidently ties him to his content.
• The checkout tactic worth testing: ask for the buyer’s Instagram handle at checkout, then have someone comment on their posts or reshare their content. Low cost, high emotional payoff.

Why word of mouth is still the best marketing channel for ecommerce

Word-of-mouth marketing is the best channel for ecommerce because it delivers your two highest-intent buyers (referrals from happy customers) at effectively zero customer acquisition cost, and it compounds with every good experience you deliver. Bumblebee Linens gets most of its wedding and event-planner business through word of mouth, and event-planner customers convert at rates paid ads cannot touch.

Every other channel has gotten worse in the last three years. Google organic click-through rates have collapsed under AI Overviews, Meta and Google ad costs have kept climbing, and Amazon FBA fees now eat about 61% of seller revenue. The relative value of a channel that costs nothing per new customer has never been higher.

Word of mouth is also the input to modern AI-search visibility. AI engines cite brand entities they recognize, and recognition comes from being mentioned across the web (podcasts, press, social media, reviews). A brand that a lot of people talk about is a brand that AI engines can confidently cite.

Why people actually shop where they shop

People shop for one of two reasons: consistency or customer experience. Amazon wins on consistency. If you cannot match Amazon on speed, price, and predictability, your only durable moat is customer experience.

Amazon’s customer experience is actually pretty good for buyers. Returns are usually easy, refunds are usually fast, and shipping is usually reliable. The reason people default to Amazon is not the experience, it is the predictability. Place an order at 11 p.m., have it on your porch by 5 p.m. tomorrow. That certainty is why Amazon owns over 50% of U.S. ecommerce.

Every other successful retailer wins on experience. The local Italian restaurant Toni goes to (owner from Italy via Brazil, always at the bar shaking your hand) has built such a strong experience moat that Toni drives past closer restaurants to get there. He gave her an extra half-pour of wine unprompted and a small jar of a salt they used in an appetizer she asked about. She has recommended that restaurant to everyone in town since.

Almost none of what that owner did cost him real money. It cost him attention. That is the entire template for ecommerce word of mouth: give customers small, specific, generous moments they will remember and repeat.

How to build customer relationships that turn into repeat sales

The two highest-ROI relationship tactics for an ecommerce brand are personally answering customer emails (yes, from the founder) and sending thoughtful year-end gifts to top customers. Both are extremely cheap and both routinely produce shocked, grateful reactions from the customer.

I try to answer every email I get personally, and at least once a week someone replies with “I cannot believe a real person responded.” That is how low the current bar is. Most ecommerce brands hide behind a support ticket queue with a canned auto-response. A founder reply on a real question, especially in the first 12 hours, converts a first-time buyer into a lifetime customer.

Jen (my wife) sends small holiday gifts to Bumblebee Linens’ best customers every year. I initially thought it was pointless (they were already regulars), but many of those customers have kept ordering for over a decade. The gift is not what closes the next order. It is what makes the customer feel seen enough to keep choosing you over Amazon when the next purchase comes up.

Relationships cannot be automated. You can systematize them (SOPs for who gets a gift, who gets a reply), but the actual moment of “someone at this brand knows me” has to be human.

The corporate gifting mistake almost every ecommerce brand makes

The biggest mistake ecommerce brands make with gifting is sending customers branded swag (a mug, a t-shirt, a pen with your logo) instead of something the customer actually wants. Branded swag is an ad you are asking the customer to display for you. That is not a gift.

Good gifts are useful, unbranded, and slightly personal. A Yeti-style cup in a color the customer likes. A holiday ornament tied to the year or the customer’s interests. Something they would keep on their desk or in their kitchen instead of leaving in a drawer.

For your top 20 to 50 customers, spend 5 to 10 minutes each on their social profiles first. Find out what they are actually into. Match the gift to that. It sounds like a lot of work, and it is worth it because a matched gift is remembered for years.

Gary Vaynerchuk ran the extreme version of this on his wine business. He audited his customer list, found people who spent modest amounts but had large Twitter followings, and identified one customer who was a huge Jay Cutler fan. He sent that customer a $400 autographed framed Jay Cutler jersey. Nothing happened for a few weeks, then exactly one month later Gary got a massive wine referral order sent by that customer. That is what a $400 gift buys when it is thoughtful and targeted.

How to get your ecommerce brand cited by AI search (SearchGPT, Perplexity)

AI search engines like SearchGPT and Perplexity cite brand entities they recognize, and recognition is built by having your brand and your founder mentioned across the web (podcasts, YouTube, press, Reddit, social media, and other blogs). Getting cited by AI requires the same brand-authority signals that used to work for SEO backlinks, just distributed across more surfaces.

Perplexity now has one-click checkout. Ask Perplexity for a product, click the shopping button, and the order gets placed directly through a Shopify store using ShopPay credentials the user has already saved. That is a full purchase, in the AI answer, before any Google click happens. If your brand is not in that AI answer, the transaction never touches your store.

Getting into those AI answers requires strong brand authority signals. That means the founder or the brand being talked about across the web, not just on the brand’s own site. Podcasts, YouTube guest spots, real customer reviews, Reddit threads, and press mentions all feed the entity graph AI engines use.

I made one small change to my personal site (adding a section that links every podcast, press mention, and speaking gig I have done) and my traffic doubled in one day and kept climbing. Google finally had enough signal to confidently associate me with my own content. Same principle applies for a product brand: if your brand name appears in five podcast transcripts, three roundup articles, and 20 authentic Reddit threads, AI search will start citing you.

How to use social media as an ecommerce word-of-mouth engine

Social media is now the primary word-of-mouth engine for ecommerce because a single TikTok or Reel about your product can drive more organic discovery than a year of Instagram posts, and clips of you (or the person behind the brand) build the entity signal AI search uses. YouTube and TikTok are where people find new restaurants, products, and destinations in 2025.

You do not need to make content directly about your product. There is a woman in Vancouver who runs a swimwear brand and posts under the handle Strawberry Milk Mob about dating advice. She has 2.2 million followers, never mentions the swimwear in her videos, and drives her whole business through the link in her bio. Her content matches her target demographic (women, roughly her age); the product simply lives adjacent.

For most ecommerce brands, the easiest starting content is behind-the-scenes. Set up a camera in front of the embroidery machine when someone orders a funny custom message. Film how the packaging gets designed. Talk through what actually breaks in the business day-to-day. People love watching how things get made, and a 5,000-subscriber niche channel with an engaged audience can support a service business or a specialty ecommerce brand entirely on its own.

Toni’s biggest push into this in 2025 is content built around Bumblebee Linens as a real operating business: what breaks, what runs, what a print job looks like, and the surprising day-to-day of running a linens company. The behind-the-scenes format works because it does not require the founder to be a comedian or a personality. It just requires showing up consistently.

The one checkout tactic that turns customers into content creators

The highest-leverage single tactic to test right now is asking for the customer’s Instagram handle at checkout, then having someone from your brand comment on their posts or reshare their content when they post something you made. Low cost, high emotional payoff, and it converts customers into ongoing organic reach.

The instinct that makes this work is the same one behind the “you replied to my email” reaction. People are starved for genuine, positive attention on social media. When a brand they love comments on their post (not asking for anything, just being nice), it feels good and gets talked about.

I rented outfits from Rent the Runway for years. Every time I posted an outfit I loved and tagged them, they would reshare it to their stories. I knew exactly what they were doing (it was brand-building for them, obviously), and it still felt good. I told other people about them. Even a cynical marketer’s brain responds to being seen. Your customer, who is not a marketer, will respond even more strongly.

This can be outsourced once you build the SOP. You are not writing a novel. A single-sentence “these turned out beautifully, thanks for sharing” comment from a real brand account, sent consistently, converts customers into free ongoing distribution.

Frequently asked questions

What is word-of-mouth marketing for an ecommerce store?

Word-of-mouth marketing for an ecommerce store is any strategy that gets existing customers to actively recommend, tag, review, or share your brand with new customers. It includes referral programs, thoughtful gifting, personal customer service (like a founder-reply email), a memorable unboxing or customer experience, and content that customers organically post about your product on TikTok or Instagram.

Why does word of mouth work better than paid ads for ecommerce?

Word of mouth works better than paid ads because a personal recommendation from a trusted person has near-100% intent match, whereas an ad interrupts someone who was doing something else. Word-of-mouth customers convert at higher rates, spend more per order, and are far more likely to become repeat buyers. It also does not stop working when your ad budget runs out.

How do I get more word-of-mouth referrals for my ecommerce store?

The three highest-leverage moves are: reply personally to customer emails (especially from the founder), send unbranded thoughtful gifts to your top 20 to 50 customers each year, and design a customer experience worth talking about (unboxing, handwritten note, follow-up thank you, small unexpected extra). Then make it easy for customers to share by asking for their Instagram handle at checkout and engaging with what they post.

Do I need to be on social media to build word of mouth for my brand?

Yes, in 2025 you need at least one social channel (TikTok or YouTube) as a distribution surface, because word-of-mouth recommendations increasingly happen inside those platforms rather than in text messages or emails. You also need presence outside your own site (podcasts, press mentions, Reddit, other blogs) so AI search engines like SearchGPT and Perplexity will start citing your brand as an entity they recognize.

What kind of gift should I send my best ecommerce customers?

Send unbranded, useful, slightly personal gifts. A quality drinkware item in a color they would pick, a holiday ornament tied to something they mentioned, or a small artisanal food item are all good. Never send branded swag with your logo (a mug, t-shirt, pen), because that turns the gift into an ad. If you have time for your top customers, check their social profiles first to match the gift to something they actually like.

How does word of mouth affect my ecommerce brand’s AI search visibility?

AI search engines like SearchGPT and Perplexity increasingly cite brand entities they recognize, and recognition is built by your brand being mentioned across the web (podcasts, YouTube, press, Reddit, reviews, other blogs). Traditional word-of-mouth activity generates most of those mentions naturally, which is why word-of-mouth marketing and AI-search visibility now feed each other directly.

Can I run an ecommerce store without doing customer service?

You can, but you will grow much slower and get far fewer repeat customers than a store that treats customer service as a growth channel. Real customer service (fast personal replies, easy returns, follow-through on problems) is the single cheapest way to earn recommendations. If you do not want to talk to customers at all, Amazon FBA is the platform that lets you offload that, but you give up brand ownership and margin in exchange.

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!