AI Search Optimization (AEO) · focused work
AI Overviews for ecommerce
AI Overviews sit above the ordinary results for many of the questions shoppers ask before they buy: which type to choose, what to look for, whether a product suits a use. The overview is assembled from pages Google trusts and, for product queries, from the Shopping Graph that Merchant Center feeds. A store can be one of those sources or it can be summarized without being named. This work is about being the source.
This is the right work if
- Informational queries in the store's categories now show an overview and clicks to the store's guides have fallen
- Competitors' pages appear as cited sources for product-type questions and the store's do not
- Product pages state key attributes only in images, in tabs loaded by script or in PDFs
- Merchant Center prices or availability disagree with the website
- The store has deep category knowledge that is not written down anywhere on the site
What it is
What the work covers
AI Overviews for ecommerce is the work of making a store's pages the kind of source an AI Overview draws on and cites, for the queries that lead to a purchase in its categories. It begins with a query map: the questions that trigger an overview in the store's product types, the pages currently cited, and where the store's own pages sit. From there the work is editorial and structural: guide and category content that answers a question in its first sentences, product pages with attributes stated as text, and comparison content that a summary can quote accurately.
The second half is data. Product mentions in an overview show prices, availability and ratings drawn from Merchant Center and from structured data on the page, so a feed that disagrees with the site or a page missing Product markup is invisible to that part of the overview. The work aligns feed, markup and page, keeps crawler access open, and builds a way to measure a channel reporting tools do not break out. Nothing here promises a citation; it removes the reasons a store would not be chosen.
How it is done
The work, in order
What changes
- The store's guides and categories appear as cited sources for the pre-purchase questions in its categories
- Product mentions inside overviews carry the store's current price and stock because feed and page agree
- Question-led content on the site opens with the answer, which helps human readers as much as the summary
- A tracked query set shows whether overview exposure is producing visits and revenue
Map which queries trigger overviews
Run the store's commercial and pre-purchase queries through a rank tracker that records AI Overview presence, and note the cited sources for each. Group the queries by category and by intent, then mark where the store already ranks, where it is cited and where it is absent.
Audit the pages that should be sources
Check each candidate page for a direct answer in the opening lines, attributes and specifications in HTML text, headings that match the question asked, an organization the page can be attributed to, and a last-updated date. Pages that bury the answer are rewritten so the answer leads.
Align product data across feed, markup and page
Compare Merchant Center attributes with on-page Product markup and with the visible page for price, availability, GTIN, brand and shipping. Fix the source of any mismatch, usually a stale feed or a theme that prints markup from a cached value. Confirm robots.txt allows Googlebot, which is the crawler overviews rely on.
Write the guide layer
For each category, produce the buying guide, the comparison between the main types and the FAQ a shopper needs before choosing. Each page opens with the answer, uses the vocabulary of the query, and links to the category and the products it describes. It is written from product knowledge, not assembled from other sites.
Measure by query set
Because reporting does not isolate AI Overview traffic, measure the mapped query set instead: impressions, clicks and position in Search Console for those queries over time, alongside citation checks from the tracker. Report changes per category and re-run the map quarterly, since which queries trigger an overview keeps shifting.
Platform notes
Shopify
Product data reaches the Shopping Graph through the Google and YouTube channel, so its sync schedule and attribute mapping decide how current the product facts in an overview are; theme JSON-LD should print live price and inventory, not a cached value.
WooCommerce
Feeds usually come from the Google for WooCommerce plugin or a feed plugin, and core Product schema can be overridden by SEO plugins; make sure only one set of Product markup is emitted and that it matches the feed.
Is there a way to make sure a store is cited in an AI Overview?
No. Which sources an overview cites is decided by Google's systems, and it changes between queries and over time. What a store controls is whether its pages are eligible and useful: crawlable, direct, accurate and current. The work removes the reasons a page would be passed over.
Do AI Overviews reduce traffic to product pages?
They mostly affect informational queries, where a summary can answer without a visit. Queries that name a product or a product type still need a store to buy from, and overviews for those queries show products with prices and links.
Does the store need separate content for AI Overviews and for ordinary search?
No. A page that answers a question directly, states facts in plain HTML and stays current serves both. The difference is emphasis: overviews favor a direct answer at the top and clearly stated attributes, which are also the traits that make a page rank and convert.
How is this different from optimizing for ChatGPT or Perplexity?
The principles overlap, but the sources differ. AI Overviews draw on Google's index and the Shopping Graph, so Merchant Center and Googlebot access matter most. Other assistants use their own crawlers and product data routes, each with its own access rules.
Related work and answers
Find the leak.
A free Growth Analysis ranks what your store should fix first, by revenue at stake.