AI search is influencing ecommerce sales. Your analytics just aren’t very good at admitting it yet.
A shopper can ask ChatGPT to compare products, see your brand recommended, then search for you on Google or come back to your website later. The AI platform helped shape the decision, but your analytics may give all the credit to organic search, paid ads or direct traffic.
See the problem?
If you only measure clicks directly from AI platforms, you are probably underestimating their influence. But if you treat every increase in branded search as an AI win, you are giving the channel far too much credit.
The truth sits somewhere in the middle. Finding it takes more than one report.
Direct traffic and revenue from AI platforms
Let’s start with the clearest signal: people clicking from an AI platform to your website.
Traffic is useful, but it is only the beginning. Ecommerce brands also need to understand which pages these shoppers visit, how they behave and whether they buy.
Keetrax client data shows why this is worth watching.
One home and garden retailer generated 681 AI referral sessions over 12 months, up 449% year on year. Those visits produced $845 in tracked revenue, up 147%. AI still represented only 0.15% of total sessions, but ChatGPT delivered 655 of those visits and all of the directly attributed AI revenue.
Small channel? Yes.
Growing quickly and already producing sales? Also yes.
Another Keetrax client, an outdoor cooking retailer, recorded $3,706 in ChatGPT-attributed revenue over 12 months, up from $912 the year before. That is still modest beside its larger channels, but it has moved well beyond the “maybe one day” stage.
These figures should be treated as a confirmed minimum. Referral information can disappear when someone copies a link, changes device or returns later. Google Analytics records visits without a clear source as direct traffic, so some AI-influenced demand remains hidden.
Visibility when customers are making decisions
Clicks only show what happened after someone visited your website. They do not show how often your brand appeared before that point.
This is a big gap.
A shopper might ask an AI assistant for the best product in a category, compare two materials or find something suited to a specific problem. If your brand is recommended, you may make the shortlist without receiving an immediate click.
If your competitor is recommended instead, you might lose the sale without seeing any obvious change in your traditional search rankings.
That is why AI search reporting needs to look at brand mentions, citations and product recommendations across the questions that matter commercially.
It should also check whether those answers are accurate. Being mentioned is not much of a win if the AI recommends an unavailable product, quotes an old price or completely misunderstands what you sell.
How your visibility compares with competitors
A mention count means very little on its own.
Ten citations might sound pretty good until you discover your closest competitor has 100.
Share of voice adds the missing context. It shows which brands are dominating AI-generated comparisons, which sources are repeatedly cited and where your products are missing from high-intent conversations.
This is also why waiting carries a risk.
AI search systems use retrieval-augmented generation to find current sources and support their answers. Brands with clear product information, genuinely useful content and strong authority across the web have more opportunities to become recurring sources.
The earlier your brand builds that presence, the harder it becomes for competitors to own the conversation. Let them get there first and you may spend the next year playing catch-up.
Which products and pages are doing the heavy lifting
A sitewide traffic number can hide the most useful part of the story.
Which pages are AI platforms actually sending people to? And do those pages convert?
For the home and garden retailer mentioned earlier, the homepage received the most AI visits but generated no tracked conversions. Several product pages converted AI visitors at rates between 1.96% and 3.03%.
The opportunity was not simply “get more AI traffic.” It was to build stronger visibility around the products and pages most likely to turn interest into revenue.
That distinction matters. More visibility sounds lovely, but the right visibility pays the bills.
Assisted conversions and hidden demand
AI search often starts a buying journey that another channel finishes.
Someone discovers a product through ChatGPT, searches for the brand two days later and buys after seeing a remarketing ad. Last-click reporting celebrates the ad while AI search quietly slips out the back door without any credit.
No analytics platform captures this perfectly.
Brands need to understand the relationship between AI visibility, branded search, direct visits, returning customers and assisted conversions. One signal alone will not prove that AI caused a sale. When several move together, however, a much clearer picture starts to form.
This is why judging AI search only on immediate revenue misses part of its value. It can influence product discovery, shortlist inclusion and the return visits that eventually lead to a purchase.
Google Search Console only tells part of the story
Google began rolling out a dedicated Generative AI performance report in Search Console in June 2026. It separates impressions from AI Overviews and AI Mode, with information about the pages, countries and devices involved.
There is a catch. Of course there is.
The report is still limited to a subset of websites. It focuses on impressions rather than dedicated AI clicks, queries, click-through rate or revenue. AI clicks remain included in the standard Web performance report.
Other AI platforms provide even less first-party visibility.
That leaves ecommerce brands trying to connect search data, website analytics, citation monitoring and actual sales. No single platform provides the full picture, and third-party tools only offer directional data because they cannot see inside the AI companies’ ranking and retrieval systems.
Technical readiness and brand authority
Measuring performance is one side of the job. The other is finding out what is holding the brand back.
AI visibility still depends on strong search foundations. Product pages need to be accessible. Product information needs to be accurate. Your website needs original content that answers real customer questions, along with enough authority across the web to be trusted as a source.
Google confirms that its generative search features remain connected to its core Search systems and crawlable web content. Accurate ecommerce information through platforms such as Merchant Center also matters.
There is no magic schema type, AI file or clever shortcut that guarantees a citation. Sorry to anyone selling that dream.
The goal is to remove the barriers preventing search systems from finding, understanding and trusting your brand.
So, what should your reporting tell you?
A useful AI search report should answer four commercial questions:
- Is your brand appearing when customers research the category?
- Are the right products and pages being mentioned?
- Are you gaining ground on competitors?
- Is that visibility contributing to qualified traffic and revenue?
AI search may be a small direct channel today, but it is already shaping decisions that analytics often credit elsewhere.
Waiting for the reporting technology to catch up gives competitors more time to claim the citations your customers see.
A Keetrax AI SEO audit uncovers your real AI search footprint, identifies the technical and content barriers holding you back, benchmarks your brand across major AI platforms and gives you a prioritised roadmap for attracting more high-intent traffic.
No guesswork. Just a clearer view of where your brand stands and where the hidden revenue opportunities are.
