AI Overviews are most of the reason your informational blog content is losing traffic while your product and category pages are mostly untouched. Your ecommerce SEO strategy needs to account for this now. AI is answering the questions at the top of the funnel and we are living in a zero-click world where most searches end without a click. Shoppers used to land on your guides, your comparisons, your breakdowns and then purchase down funnel.
If you haven’t focused on where shoppers pull out a credit card and actually purchase at the bottom of the funnel, it’s time to re-think your ecommerce strategy.
The real question is not whether SEO still works for ecommerce, it is where your effort earns a return. The fix is knowing where AI is taking clicks away from ToFu traffic and informational ecommerce content because those are pages you can’t win at getting shoppers to your site. Those are now just impressions driving awareness in a Zero-Click world and that is totally okay!
Where AI Overviews Are Taking Ecommerce Traffic
AI is taking traffic from your informational content and almost none from your buying pages. Get that straight before you change anything else.
The numbers are not subtle. Ahrefs studied 146 million SERPs and found AI Overviews on about 21% of searches, and of the keywords that trigger one, 99.9% are informational. Commercial queries sit at 5.5%, transactional at 1.2%, navigational at 0.1%. Break it down by category and Shopping is one of the lowest at a 3.2% AI Overview share.
Then there is what an overview does to the click. When one appears, the top ranking page now earns about 58% fewer clicks than it did before, based on Ahrefs’ December 2025 data. The queries AI answers do not lose a sliver of traffic, they lose most of it.
What AI replaces is the guide a shopper used to read on the way to buying, not the page where the buying happens.
Your “what is,” “how to,” and “ideas” content is where AI is taking the clicks. Your category pages, product pages, and buying queries are mostly untouched. In the ecommerce accounts we run, that shows up as a guide that keeps its impressions, or even gains them, while its clicks fall off, because the answer now sits in the Overview above it. The collection and product pages that guide feeds hold their numbers.
Run an Ecommerce SEO Audit for AI Exposure
Sort every page into one of two buckets, exposed or protected. It takes an afternoon and it decides where your next quarter of effort goes.
We kicked off a customer recently and asked them to pull this data and find all the queries AI Overviews are showing up for, and another where they aren’t, so we could tailor the approach for the two sides of how we approach seo and content for their ecommerce funnel.
Exposed pages are the informational ones:
- Top of funnel guides and explainers
- “What is” and definition style posts
- How to and ideas content that fully answers a question on the page
Protected pages are the commercial and transactional ones:
- Category and collection pages
- Product pages
- “Best [product]” and “[product] for [use case]” pages
- “[Brand] vs [brand]” and alternatives comparisons
Then rank inside each bucket, because not every page deserves the same attention. On the protected side, sort by revenue and margin so you work the pages that pay you first. On the exposed side, sort by whether the page still earns links, citations, or assisted conversions, which separates the keepers from the retire pile. For a skincare brand, the protected set is the category pages for each skin concern, the product pages, and the comparison content, while the exposed set is the routine walkthroughs and ingredient explainers an AI Overview now summarizes in three sentences.
To confirm what is actually under pressure, open Search Console and look for queries where impressions hold flat or climb while clicks fall. That divergence is the fingerprint of an AI Overview answering the question for you. You can also filter for the AI Overview SERP feature in Ahrefs to see exactly which of your keywords trigger one, which turns the audit from a guess into a list. It also doesn’t hurt to do some quick manual searches to look at SERPs and AIOs to find out what you’re customer is actually looking at.
Which ecommerce pages still earn clicks
Your category, product, and comparison pages are where shoppers still click, so that is where your ecommerce SEO strategy spends first. These are the money pages, and the data says they are largely safe. The work is making each one the most useful version of itself.
Category Page SEO
Give each one copy that answers the questions a shopper has at that stage, not a thin line wedged under the product grid. A running shoes collection should speak to support, terrain, and use, because that is what the buyer is weighing. Handle faceted navigation so filters do not spawn thousands of near duplicate URLs, point canonicals where they belong, and link related collections to each other so the set reads as one coherent hub instead of scattered pages.
Product Page SEO
These win on the specifics a shopper compares before buying. Write descriptions that name the real detail. “Holds a 15 inch laptop and a change of clothes” beats “spacious and versatile.” Wire up price, availability, and ratings with product schema so the engines read your catalog correctly. Surface genuine reviews and buyer questions on the page, and handle variants so your color and size URLs do not split or duplicate the ranking signals you have built.
Comparison and Best-Of Pages
Pages like “best running shoes for flat feet” or “[your brand] vs [competitor]” carry heavy buying intent, rarely trigger an AI Overview, and double as the source AI leans on when it does assemble a shortlist. They earn the click today and keep you in the consideration set tomorrow.
The Technical SEO Floor
None of this ranks if a crawler cannot reach it or render it. Keep the foundation tight with a technical SEO checklist so your money pages load fast, stay indexable, and do not burn crawl budget on filtered junk URLs.
Internal Links to Your Money Pages
The informational content you keep still earns authority, so route it deliberately. Link from each guide down to the collections and products it supports, and the authority those guides build flows to the pages that actually convert.
We’ve done this exact work and rethought how we approach these “buyer pages” as I like to call them. Calling out to call out specific features, unique differentiators, fabrics, leathers, and so on. Also, we have spent a lot of time thinking through “informational” content that is a bit more downfunnel where an AI Overview might not be.
Which is a great segway, diving into that.
Rethink Your Ecommerce Content Strategy
Your informational content still has a job. It just stopped being “drive clicks at scale.” Treat it as a brand and authority asset instead of a traffic engine and the spreadsheet starts making sense again. The positive of this was, we always thought about bottom line and not just vanity metrics, so we didn’t really spend time driving clicks, just to drive them. But, this has us tightening our focus even more.
Keep the pieces that earn their place:
- Content that builds topical authority feeding your category and product pages
- Content that gets your brand cited in AI answers during the research stage
- Content deep or specific enough that an AI summary cannot stand in for it, like comparisons grounded in real testing, original data, or a genuine point of view
- Content that supports internal linking down to your money pages
Cut or consolidate the rest. The thin posts written only to catch top of funnel traffic are the exact pages AI now answers in place, and keeping fifty of them only thins out your authority. Folding five overlapping guides into one strong page usually does more for the products underneath it than the five ever did apart.
For the keepers, the target moves from ranking for the click to being the source the answer quotes. When the engine responds, you want your brand inside the response. The structure that earns it, a direct answer near the top of each section, specific entities and numbers, and verifiable authority, is the same content design that separates AEO from traditional SEO.
How to get your products cited in AI answers
A growing share of product discovery now runs through AI, so getting named in the answer is its own job. If ChatGPT, Perplexity, or an AI Overview returns three brands for “best [product]” and you are not one of them, you never make the shopper’s short list, and no amount of on-page work fixes that after the fact.
Getting named pays off more than it used to, because the answers link out more now. In May 2026 Google started placing citations next to the exact line they support, added a block of related article links under each answer, and began crediting Reddit and forum voices with a link. ChatGPT and Perplexity already cite as they go. So a citation is no longer only brand presence, it can earn a click. Keep it in proportion though. Pew finds only about 1% of people click a link inside an AI Overview, so this wins back a slice of traffic, not the volume you used to get.
AI decides who to name by cross referencing the web, not by reading your own claims. Telling shoppers your blender is the best does nothing if the engine finds no corroboration anywhere else. So the work is building the signals it actually reads:
Third party mentions. Editorial coverage, niche blog roundups, and “best of” lists are how an engine confirms you are a real option. Pitch products into the roundups that already rank for your “best [product]” queries.
Reviews with real detail. AI reads review text to form answers, not just the star average. A review that says “this tent held up through a three day windstorm in Patagonia” hands the engine a verifiable use case to quote when someone asks for “windproof tents for severe weather.” Prompt buyers for the specific, not the generic.
Structured product data. A clean feed and product schema let the systems read your catalog, pricing, and availability accurately, which is table stakes for being included at all.
Brand presence. Consistent mentions across credible sites and a steady stream of branded search tell the engine you are an established entity, not a long shot.
Then measure it like you measure rankings. Track how often your brand surfaces in AI answers for your core “best [product]” and comparison queries so you can watch the consideration set form, or catch where you are missing from it.
One caveat worth holding onto. Semrush’s tracking shows AI Overviews slowly creeping into commercial and transactional queries, and Google has started folding shopping and ads into AI surfaces. The protected zone is real today, but the line can move, so treat it as a front to watch rather than a wall you can stop thinking about.
How to capture zero-click ecommerce demand
Even on your protected pages, some shoppers now finish researching inside an AI answer and show up ready to buy without ever reading your guide. Being the cited source wins some of those clicks back, but plenty of that demand never clicks at all. It is still real revenue, and you want a way to catch it that does not depend on a click you are no longer getting.
Build the paths that survive a zero click search:
- Defend your branded search so when a shopper looks you up by name, you own that result instead of a marketplace or a competitor
- Capture email and SMS so a single visit becomes a relationship rather than a one time hit
- Retarget the shoppers who bounced back into an AI answer before they bought
- Invest in repeat purchase and loyalty, because a customer who returns direct never has to be reacquired through search at all
- Build the brand so more people search for you by name, which is the one query type AI tends to reinforce rather than answer away
Because so much of this never registers as a click, your reporting has to change with it. Remember, all of this is okay, we are in a world that is moving more Zero-Click daily. So, you need to think about how your ecommerce seo strategy marries that Zero-Click reality and whether you are doing your job to drive awareness that will convert down funnel. show up in AI Overviews and ChatGPT can do the talking for your ecommerce SEO by generating awareness while the buyer intent driven pages can do the walking by generating actual revenue.
Key Takeaway
A strong ecommerce SEO strategy in the AI era is mostly an act of reallocation. AI answers the informational questions now, so stop pouring budget into the thin content it keeps and move that effort to the category, product, and comparison pages where shoppers still click and convert. Make your informational keepers citable so your brand stays in the AI answers that shape buying decisions, and build email, loyalty, and brand demand to catch the buyers who never click at all. If you want an outside read on where your money pages stand and where AI is quietly draining your traffic, our SEO services team can map it with you.
FAQ
Is SEO still worth it for ecommerce with AI Overviews?
Yes, SEO is still worth it for ecommerce, because the queries that drive sales are the ones AI leaves alone. AI Overviews concentrate on informational searches, while shopping and transactional queries rarely trigger them, so your category and product pages keep earning clicks. The work shifts from chasing top of funnel traffic to protecting and growing your buying pages.
Which ecommerce pages does AI hurt most?
AI hurts your informational pages most, meaning blog guides, how to posts, and definition content. Those sit in the informational bucket where 99.9% of AI Overview keywords live, so when an overview appears the top page can lose about 58% of its clicks. Category, product, and comparison pages mostly sit outside that exposure.
Do AI Overviews show on shopping queries?
AI Overviews show on shopping queries only rarely. Ahrefs found the Shopping category had one of the lowest AI Overview shares at around 3.2%, which means most product and buying searches still return normal results with clickable links. That is exactly why protected ecommerce pages remain a strong traffic source.
How do AI Overviews affect ecommerce conversions?
AI Overviews affect ecommerce conversions less than the traffic headlines suggest, because they mostly answer research questions rather than buying ones. The real risk is at the top of the funnel, where a shopper who once found you through a guide now gets the answer in the overview and never reaches your site. Your category and product pages, which sit closest to the purchase, keep earning clicks.
How do I get my products cited in AI answers?
You get cited by making your content extractable and your brand authoritative. Use structured product data, direct answers near the top of each section, real reviews, and consistent brand mentions across credible third party sites. AI engines lean heavily on off site authority when deciding which brands to name, so earning those mentions matters as much as anything on the page.
What content should ecommerce brands create for AI search?
Ecommerce brands should create two kinds of content for AI search: commercial pages that earn the click and citable pages that earn the mention. The first group is category, product, and comparison content for buying queries that rarely trigger an AI Overview. The second is deep, well sourced informational content that AI engines pull from when they build a shortlist, which keeps your brand in the answer even when the click does not come.
Should I delete my informational blog content?
No, do not delete it wholesale, but do reframe its job. Keep the informational pieces that build authority for your money pages, earn AI citations, and support internal linking, and consolidate the thin posts that only ever chased traffic AI now answers in place. The goal moves from ranking for the click to being the source the answer quotes.
About the Author: Tony Adam
