AI Agents Read Your Size Guides and Compatibility Data, So Write Them Clearly
Shoppers now ask agents whether a part fits, whether a size runs true, whether two products work together. Agents answer from your guide pages, if those pages are readable at all.
A growing share of product questions never reaches your support team. Shoppers ask an AI agent instead: will this case fit that phone model, does this brand run small, will this filter work with a three year old machine. The agent answers from whatever it can find and read, which usually means your size guide, your compatibility table, and your product specifications.
If those pages are images, PDFs with scrambled text, or charts with no labels, the agent guesses or sends the shopper elsewhere. Structured, readable fit and compatibility data has quietly become sales content.
Compatibility questions are decided by tables
Will it fit and will it work are yes or no questions, and agents answer them from tables that pair models with products. A compatibility table in proper markup, with model names spelled the way manufacturers print them, lets an agent answer in seconds with your store as the source. A chart uploaded as a picture might as well not exist, and an outdated table is worse than none, because the agent will quote it confidently.
Size guidance needs words, not just numbers
Agents look for fit language as well as measurements: runs small, true to size, size up if between sizes, cut slim through the thigh. Those phrases, written in text near the size selector and in the guide, are what let an agent advise a shopper whose size sits between two of yours. A bare grid of numbers forces the agent to stay silent or improvise, and improvised sizing advice ends in returns with your name on them.
Version and date your guides like it matters
Compatibility changes as models launch and products revise. Date your guide pages, note which product versions they cover, and keep one canonical guide rather than a new orphan page per season. Agents weigh recency when sources conflict, and a store with one current, dated guide beats a store with four contradictory ones. When a new model launches, updating the table in the first week captures the questions everyone is asking.
Feed data and page data must agree
Agents cross check your structured product data against your guide pages. If the feed lists a different weight limit, size range, or model list than the guide, trust in both drops. Assign one owner to fit and compatibility content, review it when products change, and treat a mismatch report from a customer as a bug, because the next reader who finds it may be an agent answering at scale.
Related reading
If this topic is useful, these earlier guides go deeper on connected questions: How AI Agents Read Product Reviews Before Recommending a Store, How AI Shopping Agents Compare Shipping and Returns Policies Before Buying, and How AI Shopping Agents Handle Product Variants and Options.
Frequently asked questions
Where do AI agents get sizing and compatibility answers?
From the store's own guide pages, specification tables, product feeds, and reviews, in that rough order of trust. If those sources are images, missing, or contradictory, the agent says so or sends the shopper to a competitor with clearer data.
Why is a PDF size guide a problem?
Many PDF guides extract as jumbled text with measurements detached from their labels, and some are scanned images with no text at all. An HTML table keeps each number attached to its size and measurement, which is what an agent needs to answer accurately.
What fit language helps agents most?
Short factual phrases in plain text: runs small, true to size, size up between sizes, plus a dated note of who the fit was tested on. Combined with a real measurements table, that is enough for an agent to give grounded advice.
How often should compatibility tables be updated?
Whenever a relevant product or model changes, and reviewed at least quarterly. Date the page visibly so agents and shoppers can judge freshness, and keep a single canonical table instead of seasonal copies.