Google AI Overviews have been live long enough now that we can move past speculation and talk about what's actually happening for ecommerce sites. The early panic ("AI Overviews will kill organic traffic") and the early optimism ("just optimize for AI and you'll get featured") have both given way to a more nuanced picture.
Here's what we're observing across the ecommerce retailers we work with.
Which ecommerce queries trigger AI Overviews?
Not all product searches generate an AI Overview. The pattern we're seeing is roughly:
High AI Overview frequency:
- Comparison queries ("best X vs Y for Z")
- Research-stage queries ("what to look for in hiking boots")
- Problem-solution queries ("how to fix squeaky hardwood floors")
- Category exploration ("types of espresso machines")
Low AI Overview frequency:
- Direct product searches ("Nike Air Max 90 black size 10")
- Navigation queries ("REI hiking boots")
- Price-specific queries ("cheap running shoes under $50")
- Very specific long-tail ("size 12 wide waterproof steel toe work boots")
The pattern makes sense: Google generates AI Overviews when the query implies a need for synthesis or explanation. When the intent is clearly transactional or navigational, the traditional SERP layout persists.
For ecommerce, this means AI Overviews primarily affect the research and comparison phase of the buying journey - the queries where category pages and buying guides compete.
What types of ecommerce pages appear in AI Overviews?
This is where it gets interesting. The pages cited in AI Overviews for product-related queries aren't always what you'd expect:
Category pages with strong descriptive content get cited more than thin category pages that are just product grids. If your "men's hiking boots" page has a substantive introduction explaining boot types, key features, and selection criteria, it's more likely to be cited than a page that jumps straight to product tiles.
Buying guides and comparison content are overrepresented in AI Overview citations. Retailers who publish genuine buying advice - not just SEO content, but useful guides that help someone make a decision - are seeing significant visibility.
Product pages with detailed specifications get cited for specific attribute queries. When someone asks "what's the waterproof rating of Gore-Tex hiking boots," Google pulls from product pages that have explicit, structured specification data.
Review and rating content surfaces frequently when the query implies evaluation. Retailers with authentic review content (not just star ratings, but actual review text) see their pages cited more often.
What this means for your category page strategy
The biggest tactical implication: thin category pages are now a liability, not just an underperformance.
In the pre-AI Overview world, a category page with minimal content could still rank if it had good links, authority, and product relevance. In the AI Overview world, those pages get bypassed. Google doesn't cite pages that don't say anything useful.
This is a fundamental shift for ecommerce sites that treated category pages as navigational waypoints rather than content destinations. The pages that perform now are the ones that:
- Open with substantive content that addresses the searcher's likely questions
- Include structured information about product types, features, and selection criteria
- Link contextually to related categories and relevant buying guides
- Have proper schema markup so Google can extract structured facts
The New Pages Agent and Content Agent at Similar AI are designed specifically for this problem - making sure every category page has the content depth that AI Overviews require, matched to actual search demand.
The click-through rate question
The concern most retailers raise: "If Google answers the question in an AI Overview, why would anyone click through?"
The data tells a more nuanced story. For informational queries, yes - click-through rates have declined. If someone asks "what is Gore-Tex" and gets a complete answer in the AI Overview, many searchers don't click further.
But for purchase-intent queries, the impact is different. AI Overviews for product queries tend to increase click-through to cited pages because they function as endorsements. Being cited in "best hiking boots for wet trails" with a link to your category page is effectively a Google recommendation.
The net effect for ecommerce: traffic from pure informational queries is declining, but traffic from research-stage queries (which are closer to purchase) is shifting toward cited pages. If your pages are cited, the traffic you get is higher quality. If they're not, you lose share.
Practical steps
Based on what we're seeing, here's what ecommerce retailers should prioritize:
Audit your category pages for content depth. Any category page with less than 200 words of useful content above the product grid is at risk. That doesn't mean adding filler - it means adding genuinely useful information about the product category.
Invest in structured data. Product schema, FAQ schema, and review schema are the structured signals that AI Overviews lean on. If your structured data is incomplete, fixing it is probably your highest-ROI activity right now.
Create genuine comparison and buying guide content. Not "top 10 best" listicles, but real guides that help someone understand the differences between product types and make an informed decision.
Strengthen internal links between related content. When your category page links to your buying guide which links to your product pages, you're building the content cluster that AI Overviews reward.
For a comprehensive overview of the AI Overview landscape and optimization strategies, see our detailed guide on Google AI Overviews.
The long view
AI Overviews are not going away, and they're going to expand to cover more query types. The retailers who adapt their content strategy now - creating genuinely useful, well-structured, deeply linked content - will have a compounding advantage over those who wait.
The good news for mid-market retailers: this is an area where quality of content matters more than volume. A site with 5,000 well-crafted pages will outperform a site with 50,000 thin ones in the AI Overview landscape. That's a game where smart retailers can compete regardless of size.