Google tests health ads in AI Mode: what’s the impact on your e-commerce strategy?
Summarize this article with AI
37% less traffic in 3 months: the call that alerted me
A client calls me one Thursday morning. He sells sleep devices and melatonin supplements. His catalog has 1,200 items. For three months, organic traffic has been tanking.
I check his Search Console. 37% fewer sessions over the period. The drop is brutal. No manual penalty. No major core algorithm update recorded. So what changed?
I isolate the pages losing the most clicks. All are tied to health queries. All appeared in AI Overviews. And suddenly, I get it.
22% of this site’s traffic came from generative answers displayed by Google. In recent weeks, those same answers now include ads. Ads that push organic links lower. Or remove them from the screen entirely.
The mechanism is relentless. And it’s not limited to health.
What exactly is Google testing with health ads in AI Mode?
On June 2, 2026, Search Engine Land revealed the test. Google confirms inserting ads in AI Mode for health queries. A spokesperson clarified it’s a limited experiment with no official rollout timeline.
Concretely, a user asking AI Mode a medical question (for example, « what natural sleep aid for sleep? ») gets a detailed generative answer. Below, a carousel of links. Today, an ad block can appear between them. It’s marked « Sponsored. »
The format mirrors Shopping Ads. Image, title, price, stars. A direct call to buy. Commercial intent is captured by Google’s ad network before the user even scrolls down to traditional organic results.
« This test opens the door to massive monetization of AI responses, far beyond health. » – Barry Schwartz, Search Engine Land.
For now, the test is geolocked to the US. But the history of Google tests shows one thing: what works in the US reaches us in 12 to 18 months.
Why your health e-commerce traffic risks collapsing
Health organic traffic has one peculiarity: it’s captive. The user is searching for a solution, not a website. When Google serves them a ready-made answer with a buy button built in, the odds they click an organic link drop sharply.
My observations across 14 health e-commerce sites confirm it. On average, 8% of organic traffic already comes from AI Overviews. And that number climbs 3 points each quarter. If Google turns this space into an ad network, the share of non-sponsored clicks inevitably falls.
Take a nutraceutical site I audit regularly. Its melatonin category pages captured 1,800 clicks per month via AI Overviews. After ads appeared, that dropped to 1,470 clicks. A drop of 18%. Without any other changes to the site.
The trap? This lost traffic doesn’t shift to classic organic positions. It’s absorbed by Google’s ad ecosystem.
For e-commerce businesses, the challenge is no longer just ranking first in blue links. It’s existing in the answer itself, as a cited source, as a trusted citation.
To counter the impact of ads in AI Mode, I applied the DOSE framework developed by Guillaume Attias. This flow diagram illustrates the four sequential steps that helped a nutraceutical site reverse its traffic decline.
The DOSE framework: four steps to reclaim visibility in AI Mode
How to adapt your semantic architecture and content for generative answers
How a supplements site reversed the trend
One of my clients in nutraceuticals suffered the same erosion. 12% less traffic in 45 days. I applied the DOSE framework (Describe, Orient, Signal, Establish) taught by Guillaume Attias at BMO Academy.
First step: Describe. Every product sheet was enriched with a « Proof » section containing 3 sourced clinical studies, each summarized in one sentence. I gave the AI the data it cites in its answers.
Second step: Orient. I structured pages in a semantic cocoon, linking each supplement to a condition, user profile, age range. Internal linking showed our topical expertise.
Third step: Signal. I implemented FAQPage, Product, and MedicalScholarlyArticle markup everywhere it mattered. Google displays excerpts for AI Overviews thanks to structured data.
Fourth step: Establish. We published 3 detailed case studies, written by a partnering doctor, with measured results from patients. Medical authority counts for AI answers.
Result? In 60 days, the site regained 9% of its traffic on health queries. Better yet, it’s cited in 47% of AI Overviews in its market. Even when an ad appears, its name stays visible.
Beyond health: prepare your catalog for AI ad arrival
The health test is just a lab. Ads in AI Mode will expand to other sectors: food, sports, tech, fashion, automotive. Any sector where purchase intent is strong.
Look at what happened with Google Shopping Ads. At first, they covered products. Today, they occupy 60% of mobile screens for a transactional query. The same pattern will repeat with generative AI. Organic link space will shrink. SEO doesn’t disappear, but it evolves.
Here’s what I recommend to my e-commerce clients:
- Identify your high-intent pages. Find queries triggering an AI Overview and check if your page is cited. I do this systematically in audits: in 35 recent audits, 14 sites had no idea their citation rate in AI.
- Adopt semantic markup. Structured data is your ticket into the AI ecosystem. Without it, Google can’t extract your information for a generative answer.
- Build authority outside your site. Citations in AI answers often come from trusted external sites (medical societies, journals, independent reviews). Getting a mention on one of them boosts your chances of appearing as a source.
Don’t bet only on ads. The best approach combines organic authority with paid backup. One reinforces the other.
3 concrete actions starting today to not vanish from AI Overviews
You don’t sell health products? You’re still affected. Three initiatives to launch now.
1. Enable AI traffic tracking. Google Search Console doesn’t yet segment clicks from AI Overviews. Use a UTM parameter on your links or analyze mobile traffic with the performance tool. I helped 8 clients set this up in 2026: all found between 4 and 11% of their traffic already came from AI answers.
2. Turn product sheets into answers. A standard sheet describes. A rethought sheet for AI answers a specific question. Example: replace « Running Shoe X, cushioned sole » with « What running shoe for a marathon in hot weather? The X with its thermoregulating foam sole avoids overheating below 30°C. » Short format, assertion, data point.
3. Build a semantic cocoon. Google no longer looks at one page alone. It evaluates your topical coverage. Connect your pages by intent, not product. A domotics client saw AI traffic jump 37% after restructuring content around 5 core usage questions, each becoming a branch of the cocoon.
These actions don’t depend on ads. They create lasting authority, which holds despite the ads.
When ad boxes land in your vertical, you’ll already be cited as a reference.
Your SEO audit might not be tracking the right metrics anymore.
I spend 45 minutes on your site, live. I show you where AI Overviews are stealing your traffic, and how to structure your content to appear in them despite the ads. No commitment, just facts.
Book a strategic call — 45 minFrequently Asked Questions
Will ads in AI Mode completely replace organic results?
No. Google needs organic content to feed its models. But the share of clicks on high-intent commercial queries will drop. Being cited as a source in the AI answer is a lever.
My site isn’t in health: should I really worry?
Yes. This test is a prototype. It will generalize ads in AI answers. In 18 months, all e-commerce will be hit. Better to prepare your content for AI now.
How do I know if my pages are already used in AI Overviews?
Go to Google, type your main queries, switch to mobile, and see if an AI answer appears. Check if your site shows up in it. No native tool does this automatically. Do regular manual crawls—it’s essential.
Is local SEO affected too?
For now, the test covers only général health queries. Google is testing local ads elsewhere. The merge will come to AI Mode as well.
Should I invest in Google Ads on top of SEO to counter this shift?
Ads and SEO are complementary. Ads give immediate visibility in AI answers. SEO ensures lasting presence as an organic source. The idea is to combine both.
