93% of AI Shopping Searches End Without a Click. Here's Where the Money Goes Instead.
Key finding: Semrush data from 69 million sessions shows 92-94% of Google AI Mode searches produce zero clicks. Yet Walmart reports 35% higher order values from AI-assisted shoppers. Revenue is shifting from click-based traffic to AI-mediated purchases where product recommendations happen inside the AI interface.
Last updated: February 26, 2026
A Semrush study of nearly 69 million Google Search sessions found that 92 to 94% of Google AI Mode searches end without a single click. No visit. No session. No attribution. The user got what they needed and moved on.
If you've spent years building SEO, content strategy, and conversion funnels around the assumption that people click through to your site before they buy, that number should bother you.
But it also tells an incomplete story. Because the money didn't vanish. It just started flowing somewhere else.
Walmart Just Showed Us Where
During its Q4 earnings call last week, Walmart shared that customers using its AI shopping assistant Sparky have an average order value 35% higher than those who don't. Not 35% more page views or sessions. 35% more dollars per order.
Amazon went further, claiming its AI assistant Rufus generated $12 billion in incremental sales during its first season, though the company hasn't shared the math behind that number.
AI-mediated shopping doesn't produce clicks. It produces purchases.
The Click Was Always a Workaround
Most ecommerce analytics dashboards treat the click as the thing you're optimizing for. But a click was never the goal. It was overhead. Consumers clicked because they had to. Search engines couldn't finish the job, so the only way to actually buy something was to visit the site, find the product again, add it to cart, and check out.
AI agents skip all of that. Semrush's data shows only 6 to 8% of AI Mode sessions lead to an external site visit. That's not a failure of the technology. That's the technology working as intended. The user asked for a recommendation, got it, and either bought in-chat or went straight to checkout.
The metric that matters now isn't how many people visited your site. It's whether the AI recommended your product in the first place.
Retailers Are Already Buying Into This
Williams-Sonoma is testing ads inside ChatGPT, making it one of the first retailers in OpenAI's ad pilot. Target followed with keyword-triggered ads through its Roundel retail media network, also running inside ChatGPT conversations.
Perplexity, meanwhile, went the opposite direction and pulled out of advertising entirely. An executive told the Financial Times that ads make users "start just doubting everything."
Two very different bets, but the underlying assumption is the same: the AI conversation is the storefront now. Whether brands pay for placement or earn it through catalog quality, commerce is increasingly happening inside the AI interface rather than on brand websites.
What Actually Gets You Cited
The research on what makes AI engines recommend your brand is getting more concrete, and it looks very different from traditional SEO.
Content depth beats domain authority. Research aggregated by Position Digital shows that content depth, readability, and freshness matter more than backlinks or traffic volume for AI citations. The old link-building playbook doesn't carry over.
Front-load your best information. A Search Engine Land analysis found that 44.2% of all LLM citations pull from the first 30% of a piece of content. If your key product data is buried below the fold or hidden behind tabs, AI engines won't find it.
Third parties carry more weight than you do. AirOps research found that brands are 6.5 times more likely to be cited through third-party sources than through their own domains. Reviews, press coverage, comparison content, and community discussions all outperform your own product pages in AI recommendations.
AI citations boost what organic traffic remains. When a brand gets cited in a Google AI Overview, its organic click-through rate is 35% higher than when it doesn't. The small percentage of AI searches that do produce clicks go disproportionately to brands that showed up in the AI answer.
The Consistency Problem
AI recommendations are remarkably inconsistent. SE Ranking found that running the same query three times in Google AI Mode produces overlapping results just 9.2% of the time.
Your brand can show up in one query and be completely absent from the next. Unlike holding a #3 ranking on Google for months, AI visibility is volatile by design.
This makes static, one-time optimization useless for AI search. Consistent visibility requires continuously updated, deeply structured product data that AI agents can reliably parse and compare, regardless of how the user phrases their query.
The Funnel Happens Inside the AI Now
Jason Barnard at Search Engine Land put it simply: the entire acquisition funnel, from awareness through consideration to decision, now plays out inside the AI before the user sees a single result. The agent learns about your brand, weighs it against alternatives, and decides whether to recommend you, all in one inference call.
You're no longer trying to pull visitors into a funnel on your site. You're trying to be the answer when the agent runs its own funnel internally.
Early industry estimates put the share of consumers using AI to decide what to buy at around 5% today, with projections pointing to 15% by 2027. Even if those exact numbers don't land, the trajectory is clear.
What Retail Teams Should Do With This
The brands that will win in zero-click commerce are the ones retooling their product data for a world where the buyer is software.
Structured, machine-readable catalogs. Product descriptions built for AI parsers, not just human browsers. A presence in the third-party content that LLMs actually cite. And a recognition that AI visibility is a revenue channel, not a side experiment.
The zero-click number isn't a crisis. It's a signal. The money is still moving. It's just going through a different door.
Related Reading
- The Retailer's Guide to Agentic Commerce
- 41% of Consumers Now Use AI for Product Discovery
- AI Visibility: The New SEO
