546: Google Pisses Off Sellers And The Latest Updates From TikTok

Podcast: Download (Duration: 41:20 — 47.6MB)

Google just stopped accepting credit cards for ad payments, TikTok launched AI avatars and gave creators real search-metric data for the first time, Reddit rolled out in-content ads, and Adobe quietly tried to grant itself rights to your creative work. This week’s episode of the My Wife Quit Her Job podcast covers the biggest platform shifts hitting online sellers right now, and my co-host Tony Rowland and I break down what each one actually means for how you run your business.

The credit-card payments change is going to cost heavy Google Ads spenders millions in lost credit-card rewards. TikTok’s new avatar tools and search analytics start closing the gap on Google. And every platform announcement this week points in the same direction: content-creation costs are approaching zero, and the moat is now trust, demonstration, and real expertise.

Below is the full rundown of every update: what changed, why the platform did it, and what to change in your own business today to stay ahead of the shift.

Key takeaways

• Google Ads no longer accepts credit cards. Payment options are ACH, wire transfer, or paper check. This wipes out a common travel-hacking play where advertisers earned huge credit-card rewards on ad spend.
• TikTok launched licensable AI avatars that let creators publish videos without appearing on camera, and creators can also license their own likeness.
• Adobe quietly updated its terms of service to imply broad rights to user content for AI training, and reversed course after backlash.
• Reddit started running native in-content ads inside threads. Combined with Reddit’s SEO surge, this means ad fatigue is landing on a platform that used to be ad-light.
• TikTok now shows creators real search metrics inside the app, mirroring what Google Search Console used to provide before Google removed keyword data.
• User-generated content ads are underperforming plain image ads for several sellers, likely because audiences can now detect scripted, paid UGC.
• Wix released an AI website generator, but Wix still lacks the third-party app ecosystem to support a serious ecommerce business.

Why Google stopped accepting credit cards for Google Ads payments

Google stopped accepting credit cards for Google Ads payments to save on interchange fees, which typically run 1.5 to 3 percent of every transaction. Advertisers must now pay via ACH bank debit, wire transfer, or (surprisingly) paper check. For high-spending accounts, the interchange savings for Google are enormous, but the collateral damage to advertisers is real.

For years, savvy advertisers used credit-card rewards on Google Ads to earn massive travel points, cash back, and sign-up bonuses. Some of my friends earned millions of points a year on ad spend alone, funding first-class travel entirely from advertising budgets that were going to be spent anyway. That perk is gone.

Why ACH is a worse deal than a credit card

ACH is worse than a credit card in three ways: no rewards, no float, and much weaker dispute rights.

• No rewards. You forfeit the 1 to 5 percent cash-back or points you were earning on the same spend.
• No float. ACH pulls money directly out of your bank account within days. A credit card gives you roughly 30 days before the balance is due.
• Weaker dispute rights. Credit-card chargebacks are well-defined consumer protections. ACH disputes exist but are slower and less favorable to the account holder.

The other risk is standing authorization. Once Google (or any vendor) has your ACH details, they can pull funds essentially any time. If Google over-charges or double-charges, the money is already gone from your account while you dispute it.

Why Facebook and TikTok Ads are likely to follow

If Google can push advertisers off credit cards, Meta and TikTok will notice. Facebook and TikTok have exactly the same interchange-fee bill, and if they can eliminate it without losing advertisers, they will. Expect similar payment-method restrictions from those platforms in the next 12 to 24 months.

What TikTok’s new AI avatars mean for content creators

TikTok’s new AI avatars let creators generate video content without ever appearing on camera, by combining a stock or custom avatar with a voice (yours or a licensed one) and a script. TikTok showed working demos where the avatars are close to indistinguishable from real humans on screen.

The two use cases are stock avatars (a library TikTok provides) and custom avatars (a digital twin trained on your likeness). The custom-avatar version is what changes content strategy, because you can now film once, then generate hundreds of videos of “you” saying anything, in any language.

The avatar production pipeline creators are already using

Some creators are already stringing this together with automation tools. The flow looks like this.

1. ChatGPT writes 20 TikTok scripts in one prompt.
2. Make.com or Zapier passes each script into ElevenLabs, which converts the script into your cloned voice.
3. The voice file feeds into TikTok’s avatar tool (or a comparable video generator), producing a publishable video.
4. The video posts automatically to TikTok.

The economics are dramatic. A creator who used to film and edit ten videos a week can now publish 100-plus, at close to zero marginal cost per video, without touching a camera.

The identity theft risk from AI avatars

The obvious risk is impersonation. If your face and voice can be cloned by anyone with a few seconds of your content, someone can create videos of “you” endorsing products, courses, or scams that you never approved. TikTok requires creators to label AI-generated video, but enforcement will be uneven.

Our friend Andrew Youderian recently cloned my voice as a demo and made me say things I would never say in real life. The technology to protect against this does not exist yet, which is why some sellers (like Emmaloos Boutique’s Tiffany Ivanovsky) are already dealing with cloned TikTok accounts diverting sales to knockoff stores.

Why user-generated content ads are underperforming plain image ads in 2024

User-generated content ads are underperforming plain image ads because audiences have learned to spot paid, scripted UGC and are no longer swayed by it. Several Facebook ad buyers have reported the same trend independently: creator-style talking-head ads used to convert better than static images, and now the pattern is flipping.

The reason is exposure. Once shoppers see enough perfectly-lit, perfectly-spoken, perfectly-attractive “regular people” reviewing products, they figure out that these are paid UGC creators reading a brief, not real customers. The illusion breaks and the persuasion breaks with it.

What still works for ecommerce video ads

Product demonstration content still works, because demonstration is harder to fake than testimonial. A video of the product being used, installed, cleaned, or in action shows the buyer what they will actually experience, and the format resists AI replication because it requires real physical setup.

The kitchen sink ad Tony has been served 30 times in a week is a textbook example: a real person drops the sink in without removing the countertop, washes glasses on the built-in bar, uses the built-in cutting board, and demonstrates each feature. That kind of demo is far more persuasive than a talking-head UGC creator saying “I love this sink so much.”

Expertise-signal content also still works. A licensed pediatrician recommending a baby product converts far better than a Kim Kardashian mention, because the expertise is directly relevant to the purchase.

A friend of ours had Kim Kardashian organically post about his baby blanket and it drove roughly four sales. A “Dr. Sharon Williams, board-certified pediatrician” placement on the same product could plausibly do 10x that.

Adobe’s terms-of-service AI training scandal, explained

Adobe quietly updated its terms of service in early June 2024 with language that implied Adobe could access user-created content and use it to train AI models. After massive public backlash from Adobe’s professional creative user base, Adobe reversed course and issued a clarified terms-of-service update within days.

The scandal matters because Adobe touches roughly every professional creative workflow on the planet: Photoshop, Illustrator, Premiere, InDesign, Lightroom. If Adobe can train on user content by default, the entire creative industry’s output becomes AI training data by default.

The precedent is worse than the specific reversal. Every major creative-software vendor is now on notice that professional users will read the TOS and revolt if training-data language appears. Expect very careful, very lawyered-up TOS updates going forward.

Reddit’s new in-content ads: what they look like and what they mean

Reddit rolled out native in-content ads that sit inside thread pages, rather than only in the sidebar or above the header. This is a meaningful expansion of Reddit’s ad inventory, and it lands at the exact moment Reddit is receiving the largest search-traffic bump in its history from Google.

Reddit went public in March 2024, which is the classic trigger for a platform to raise ad load. Public shareholders demand quarterly earnings growth, and expanding ad inventory is the fastest path to it. Expect Reddit’s ad density to keep climbing.

How Reddit search traffic and in-content ads collide

The unfortunate combination is heavy search-traffic entrants (new users who arrive from Google without knowing Reddit’s UX) plus native ads that look like real posts (users can accidentally click ads thinking they are content). For search-driven sellers, this means the Reddit citation strategy for AI answer engines is still valuable, but user experience on Reddit is degrading.

TikTok search metrics: the analytics Google took away

TikTok now shows creators real in-app search metrics, including which queries brought users to their content, search-driven view counts, and search-based discovery data. This is data Google Search Console used to provide before Google removed keyword-level data in 2011 with the “not provided” change.

Why TikTok search is the next Google for many queries

TikTok search has been quietly replacing Google search for younger users on a growing set of query types. My kids search TikTok for restaurants, product reviews, and how-to content. TikTok returns a video showing the food, the menu, the atmosphere, and real reactions, which is a better answer than 10 blue links plus an AI Overview.

For ecommerce sellers, TikTok search analytics unlock a Google Search Console-style workflow: find the queries driving your content, optimize titles and captions for those queries, expand into related queries. This is a real SEO opportunity, and it exists inside a platform where search intent is already visual.

Wix’s new AI website builder: why I still would not recommend Wix for ecommerce

Wix released a new AI website generator that builds ecommerce sites from a text prompt describing what you want to sell and how you want it to look. The generated sites look good visually, but Wix remains a poor choice for a serious ecommerce business for one reason: the third-party app ecosystem is not there.

Why designing the website is the easiest part

Designing the website is the easiest part of an ecommerce business, and it only has to be done once. The hard parts are the ongoing work: email marketing, SMS, upsell flows, subscriptions, reviews, loyalty, wholesale, POS integration, accounting, ad-platform integrations. Every one of those needs a third-party app, and most of the best ones do not support Wix.

Klaviyo integration with Wix requires jumping through additional hoops. OneClick Upsell and similar Shopify-native tools do not exist in the Wix ecosystem. Every integration friction point compounds into recurring maintenance costs.

What to use instead

For a serious ecommerce business, use Shopify, WooCommerce, or BigCommerce. Each has a mature third-party app ecosystem, deep integrations with every ad and email platform, and a large enough developer community that whatever you want to build will already exist as an app.

What every online seller should do in the next 90 days based on this week’s changes

The specific actions that matter for online sellers based on this week’s platform news are: adjust ad-payment workflows, experiment with TikTok avatars carefully, shift ad creative toward demonstration over UGC, and audit your Reddit and TikTok search presence.

1. Set up an ACH funding account for Google Ads that is separate from your main operating account, so a mistaken charge cannot drain everything. Keep only 1 to 2 weeks of ad spend in it.
2. Test TikTok avatars for supplementary content (evergreen product education, restock announcements) but keep a real human on camera for your brand-defining content.
3. Rebalance ad creative away from generic UGC talking-head ads toward product-demonstration videos and expert-voice-of-authority content.
4. Audit your TikTok search analytics the moment they are available in your account, and optimize captions and hashtags for the queries driving your discoveries.
5. Read the terms of service on every SaaS you use. Adobe will not be the last vendor to try to expand rights to user content for AI training.
6. Do not migrate to Wix even if the AI site generator looks compelling. The third-party app tax will hurt you within a year.

Frequently asked questions

Can I still pay for Google Ads with a credit card?

You can no longer pay for Google Ads with a credit card. Google has switched to ACH (bank debit), wire transfer, and paper check as the only accepted payment methods. This change is happening across accounts globally and removes the credit-card rewards play that heavy advertisers used to fund travel and cash back.

What is a TikTok AI avatar?

A TikTok AI avatar is a digital character (either from TikTok’s stock library or based on your own likeness) that TikTok can animate to speak any script in a chosen voice. The result is a publishable video that looks like a person talking to the camera, without any actual filming. Creators use them to scale video production without appearing on camera themselves.

Is Adobe using my content to train AI?

Adobe rolled back the June 2024 terms-of-service update that implied broad rights to user content for AI training, after significant backlash from the creative community. As of the revised terms, Adobe states it is not using customer content to train its Firefly AI models. Users should still read updated terms carefully whenever any creative-software vendor issues a TOS change.

Are user-generated content ads still effective?

User-generated content ads are still effective for some audiences and products, but effectiveness has declined as consumers have learned to recognize paid UGC. Several ecommerce advertisers report that plain image ads and product-demonstration videos now outperform talking-head UGC ads in their accounts. The safest current strategy is to test formats side by side in your own account rather than assume UGC always wins.

How do I check TikTok search metrics for my account?

TikTok search metrics are available inside the TikTok Creator Center under the analytics tab, and inside the Business Center for business accounts. The rollout has been gradual, so if you do not see search data yet, check back in a few weeks. The data mirrors what Google Search Console offers, showing which queries drove users to your content.

What is the safest way to fund a Google Ads account after the credit-card change?

The safest way to fund a Google Ads account is a dedicated bank account holding only 1 to 2 weeks of expected ad spend, connected to Google via ACH. This limits exposure if a billing error or fraudulent charge occurs, since ACH pulls funds directly and dispute processes are slower than credit-card chargebacks. Keep your main operating account separate.

Is Wix a good platform for a new ecommerce store?

Wix is not a good platform for a new ecommerce store planning to scale, because Wix lacks the mature third-party app ecosystem that ecommerce operations require. Klaviyo integration is more difficult, popular upsell tools are not available, and every integration friction becomes ongoing maintenance cost. Shopify, WooCommerce, and BigCommerce are all better foundational choices.