The New Shopping Experience
When someone asks Gemini for "the best running shoes for flat feet," there's no sponsored placement, no ad auction. So how do AI engines decide what to recommend?
This is the $40 billion question that's reshaping e-commerce as we know it.
No Ads, No Auctions
In traditional search, the rules were clear. You could pay for placement. You could optimize for keywords. You could build backlinks. The system was gameable—and brands got very good at gaming it.
AI recommendations work differently. There's no ad slot to buy. No keyword to target. The AI synthesizes information from across its training data and decides what's "best" based on patterns it learned during training.
Where Recommendations Come From
Our research team spent three months tracking AI recommendation sources. Here's what we found:
Review aggregators account for roughly 35% of cited sources. Sites like Wirecutter, Consumer Reports, and niche review blogs appear frequently in AI reasoning.
Reddit and forums contribute about 25% of the signal. Real user discussions—particularly those with detailed comparisons—heavily influence recommendations.
Brand websites only account for 15% of direct citations. And even then, it's usually product specs rather than marketing copy.
The remaining 25%? A mix of news articles, academic papers, and miscellaneous web content.
The Influence Gap
Here's what keeps CMOs up at night: you can't buy your way into AI recommendations.
Traditional digital marketing budgets—the ones focused on paid search, display ads, and sponsored content—have almost zero impact on how AI engines recommend products.
The brands winning in AI recommendations are those with:
- Strong presence in review publications
- Active, positive community discussions
- Clear, factual product information
- Consistent brand messaging across sources
What This Means for E-commerce
The $40B question isn't just about where recommendations come from. It's about where marketing budgets should go.
Brands spending heavily on traditional SEM are seeing diminishing returns as more consumers shift to AI-assisted shopping. The smart money is moving toward influence strategies that shape how AI perceives and recommends products.
The Path Forward
Understanding AI recommendation mechanics is the first step. The second is auditing your current AI visibility—how do AI engines actually describe and recommend your products today?
That gap between reality and AI perception is where opportunity lives.