Google AI Mode Sends 57% of Product Links to Google Shopping

Have you ever wondered what happens to Google traffic as users increasingly search for products through AI Mode instead of typing keywords into the traditional search box? The answer is not just a dry number. It is a wake-up call for anyone building an e-commerce strategy based on the belief that “good SEO simply guarantees traffic.”

A new study from AI display measurement platform Findrix analyzed 18,400 buyer queries across 35 markets in the US, collecting 55,198 responses from ChatGPT, Gemini, and Google AI Mode between October 14 and 17, 2026. The results shocked many SEO specialists and marketers: 57% of links in AI Mode product responses lead straight to Google Shopping. Furthermore, in 46.5% of product responses in the US, AI Mode only linked to Google-owned pages.

Join me, SilverZ, as I dissect every figure in this study. Because behind these percentages lies a structural shift in how Google distributes attention—and if you don't understand it, you might be letting competitors get ahead on your own home turf.

What is Really Happening Inside AI Mode?

To understand why the 57% figure matters, you need to look at how AI Mode operates. When a user asks a shopping query—such as “what swim trunks are good for beginners” or “which sunscreen is suitable for oily skin”—AI Mode does not just return a list of blue links like traditional search. It synthesizes information, provides direct answers, and inserts relevant links.

And according to Findrix data, most of those links do not lead to brand or retailer websites. They lead to Google Shopping—where Google displays product images, prices, reviews, and stock availability, all within an interface controlled by Google.

Yosif Merman, co-founder of Findrix, explains the reason: Google’s product database contains prices, ratings, and inventory information. AI Mode can use that data to build product answers with fewer links to external websites. In other words, Google does not need to send users away from its ecosystem to provide shopping information. They already have the data to do it right on the spot.

This creates a painful paradox for sellers: your product may appear in AI Mode responses, but users don't need to click your website to buy. They can view prices, compare, and even complete transactions without ever leaving the Google interface. And referral traffic—which SEOs have long viewed as a measure of success—no longer fully reflects the true visibility of a product.

Which Industries are Hardest Hit?

Not every sector is impacted equally. Findrix’s study shows a clear divergence across industries.

Fashion and beauty are the hardest hit groups. Specifically:

• Swimwear: 60.4% of responses only linked to Google pages.
• Women's everyday fashion: 57%.
• Sun protection skincare: 54.1%.
• Active sportswear: 53.8%.

Conversely, sports nutrition is a notable exception: only 7.6% of responses exclusively linked to Google. This shows that not all products are “trapped” in the Google ecosystem to the same extent. Industries with high specialization, requiring complex consulting or comparisons, may still need external links to provide complete information.

For local services—plumbers, lawyers, dentists—only 5.1% of links pointed to Google. And for software, that figure was a mere 1.1%. This difference reflects a reality: Google does not have a comparable data advantage in these fields as it does with physical products. Therefore, AI Mode is forced to rely more heavily on external sources.

Comparison with ChatGPT and Gemini: Three Different Linking Philosophies

Interestingly, the study also compares how the top three AI platforms handle links in product responses. And the differences are vast.

Google AI Mode has the fewest links: an average of only about 3 links per response. And as mentioned, most of those lead to Google Shopping.

ChatGPT has significantly more links: about 25 links per response. In product responses, ChatGPT links to Walmart in 45.1% of cases and Amazon in 6.1%. This shows that ChatGPT acts more like a “referrer” to external retailers rather than trying to keep users within its own ecosystem.

Gemini had no links pointing to google.com within the measurement scope. This may reflect how Gemini is designed to function as a more independent assistant, or simply that its linking structure differs from AI Mode.

The difference also shows in how platforms cite Reddit. ChatGPT cites Reddit in 35.6% of all US responses, compared to just 5.1% for AI Mode and 1.5% for Gemini. This indicates that each platform has its own “source ecosystem,” and optimization strategies must vary for each.

Why Does This Matter to Marketers?

You might be thinking: “This study only surveyed the US market, what does it have to do with global markets?”

The answer is: a lot.

First, Google AI Mode is gradually rolling out globally. What is happening in the US is a preview of the near future everywhere else. As AI Mode becomes more popular, the trend of “trapping” links within Google Shopping is likely to repeat, especially for fast-moving consumer goods, fashion, and beauty—sectors with massive shopping search volumes.

Second, the “referral traffic” model is being challenged. If users can view prices and product details right inside AI Mode without clicking a website, metrics like organic traffic and referral traffic will no longer accurately reflect brand visibility. Marketers need to start tracking metrics around AI visibility, not just rankings on traditional search engine results pages.

Third, content strategy needs to evolve. If AI Mode only points to Google Shopping in many product answers, optimizing a website to “get clicks” may no longer be as effective as optimizing product data to be displayed in Google Shopping. This means elements like product feeds, structured data, competitive pricing, and product reviews become more important than ever.

Fourth, opportunities still exist for “difficult” industries. The study shows that fields like local services and software still have much higher external link rates. This means if your business operates in these sectors, AI Mode remains a potential traffic referral channel. But you need to understand how the platform selects sources and optimize your content accordingly.

What Should You Do Starting Today?

Facing these changes, here are concrete action steps I recommend taking.

1. Measure AI Presence, Not Just Rankings

Start tracking whether your brand and products appear in AI Mode, ChatGPT, and Gemini answers. AI visibility measurement tools are becoming increasingly popular. If you can't measure it, you can't improve it.

2. Optimize Product Data for Google Shopping

Ensure your product feed is complete, accurate, and regularly updated. Pricing, stock availability, reviews, and high-quality images are factors AI Mode uses to construct answers. If your data is missing or inconsistent, your products may not be displayed.

3. Build In-depth, Hard-to-Replace Content

Simple content that merely lists product specifications can easily be synthesized and replaced by AI. Focus on content that AI cannot generate on its own: hands-on reviews, detailed usage guides, experience-based comparisons, and unique product insights. These are reasons for users to still click through to your website.

4. Diversify Distribution Channels

Don't put all your eggs in the Google basket. Build a presence across multiple AI platforms—ChatGPT, Gemini, and other search tools. Each platform has different source selection and link display methods, and understanding each will help you reach more customers.

5. Monitor and Adapt

AI Mode is still evolving and changing. What works today might differ tomorrow. Establish a routine monitoring and periodic evaluation process so you can detect shifts in how Google distributes links early and adjust your strategy accordingly.

When “Being Seen” No Longer Means “Being Clicked”

I have worked in SEO for two decades, and I have witnessed this industry change enough times to force professionals to relearn everything from scratch. From Panda and Penguin algorithms, to continuous core updates, to the rise of generative AI. Every time, the survivors are those who understand the nature of change and adapt the fastest.

The research on Google AI Mode and Google Shopping is a reminder that visibility no longer equates to traffic. Your product may appear in AI answers, but users may never step foot on your website. This does not mean SEO is dead. It means SEO is changing shape—from optimizing for rankings to optimizing for visibility, from attracting clicks to building brand trust and authority.

And in a world where AI increasingly acts as the information “gatekeeper,” the question is no longer “How to rank #1 on Google?” but “How to make AI trust and choose my brand as a source?” That is a different game, with different rules, and those who understand that earliest will lead the pack.

Knowing that Google AI Mode is driving most product links back to Google Shopping instead of retailer websites, how will you change your SEO measurement approach—continue relying on traffic, or start tracking brand visibility in AI answers? And if you operate in a hard-hit industry like fashion or beauty, what will you do to ensure customers still find and trust your website, even when they don't need to click through to shop? Share your perspective below, because in this AI search era, whoever understands the rules best wins!