Your About Page Says Who You Are. No Machine Can Use It.
Your About page tells people who you are. The product data AI shopping assistants work from rarely carries any of it. Here is how to connect them.

In this article: the part of Interpretability Engine Optimization (IEO) that deals with the brand, not the product. Who makes it, where, how, and with what. The job of this part is simple to state: whatever your brand says about itself to people has to reach the machine as the same facts, attached to the products it sells.
Here is what happens without it.
A shopper asks an AI assistant for "a leather belt made in Portugal by a small family workshop." Your About page answers every word of that. It tells the story of a grandfather who opened a workshop in Guimarães, Portugal, the two generations that followed, the vegetable-tanned leather from Italian tanneries, the small runs cut and stitched by hand. It's a good page. People read it and buy.
The assistant may never see it. What usually reaches an assistant is product data: title, category, price, description, a few attributes. On the belt, none of those says "Portugal", "family workshop" or "vegetable-tanned". So the assistant recommends a belt from a brand that put those three facts in its product fields, and your shopper buys from them. You lost a sale to a story you told better.
Why the About page doesn't travel
An About page is written for a person who is already on your site. A shopping assistant usually isn't on your site. It's working from what reached it through a feed or a catalog connection, and that's product data. Some systems do read pages; what you can't count on is any of them connecting paragraph three of your About page to belt number four.
Platforms say as much. Google's new conversational attributes let you attach documents to each product, and the rule for them is to focus on the product, not on documents that represent your company. Shopify tells merchants to describe products in literal language and to map custom fields so its Catalog can pass them to AI channels. The door into the catalog an assistant uses is the product record. Brand facts walk through it only if they're attached to the product.
Brands rarely attach them. The provenance lives on one page, the products on another, and the link between them is a menu a machine has no reason to click.
The technical name: the mirror principle
In IEO this has a name: the mirror principle. What a person reads and what a machine reads have to state the same facts, because they come from the same source. The equivalence is factual, not literal: the machine's version can carry things the person never sees, such as category codes or normalized units, and the words don't have to match. But every fact in it has to trace back to something a person could read on your site, and every fact a person reads should have a place where a machine can find it.
Your About page breaks the mirror in one direction. The person gets the whole story. The machine gets a belt called "The Porto".
Fixing it means three moves:
- Pull the facts out of the story. Made in: Guimarães, Portugal. Leather: vegetable-tanned, from Italian tanneries. Production: family workshop, small runs, hand-stitched. Not every line of the story is a fact; keep the ones a shopper might ask about.
- Attach them to the products they're true for. If every belt you sell is made in that workshop, that fact belongs on every belt, as an attribute, not as a link to a page.
- Keep the fact the same everywhere; the words can differ. The story can say "the north of Portugal" and the field can say "Guimarães, Portugal". One is more precise than the other, and they agree. What the mirror can't survive is two facts: Guimarães on one page, somewhere else in the data.
Where it lands: the product, not the page
Here is the belt, before and after the brand facts reach it:
| The shopper asks | Where the store says it | What the machine had | What it has now |
|---|---|---|---|
| Made in Portugal? | About page: "a workshop in Guimarães, Portugal" | Nothing | Made in: Portugal (Guimarães) |
| What leather? | About page: "vegetable-tanned leather from Italian tanneries" | Nothing | Leather: vegetable-tanned · Tanned in: Italy |
| Small workshop? | About page: "small runs cut and stitched by hand", two generations after the grandfather | Nothing | Production: family workshop, small runs, hand-stitched |
| Who makes it? | About page: the family makes them | Brand: (the store's name) | Brand: (the store's name) · Maker: the brand itself |
Nothing in the last column is new. It's the About page, cut into facts and attached to the product. The page stays as it is, because people still read it.
Two places carry these facts. The first is the product itself: the attributes and custom fields your platform sends on. The second is your site's own markup, where the organization can be described in a structured way, so that a machine reading your pages can tie the brand to its products. The first is what shopping assistants work from. The second is what keeps the mirror whole.
What to do this week
- Read your About page as a list of facts, not as a story. Write down every claim a shopper might ask about: where, how, by whom, with what.
- For each one, ask: which products is this true for? All of them, or some? A fact that's true for every product goes on every product.
- Find the field, then check that it travels. Google's and Meta's product data both have a material attribute and a brand field. For the rest, use the custom fields your platform lets you send on. Then open the channel's catalog and look for the value. Shopify, for example, documents how to map sources into title, description and category for its Catalog; a custom field that isn't mapped stays home. Creating the field is half the job. Seeing it arrive is the other half.
- Keep the fact the same everywhere. The wording can change from the story to the field; the fact can't.
- Leave out what the page doesn't say. If the About page says "cut and stitched in Guimarães" and never says "Portugal", you can add the country, because the page names where the making happens. If it says "leather from the best tanneries" and never says where, the country of the leather stays empty.
Practical example
A hot-sauce brand's About page says: "Founded by a chef in 2019. Every batch is cooked in our own kitchen in Austin, Texas, with no artificial preservatives."
Its best-selling sauce has these fields: title, price, size, "medium heat".
A shopper asks an assistant for "a hot sauce made in the U.S. with no artificial preservatives."
Before you read on, decide which facts move to the product, and in what words.
Moves to the product: Made in: USA (Austin, Texas). Preservatives: no artificial preservatives. Both are on the page, in plain words, and both are true for every sauce the brand makes. Keep the brand's own scope: the page says no artificial preservatives. Turning that into "preservative-free" would be adding a claim the page never made.
Travels as brand information: "Founded by a chef" is a fact about the company, and it can go with the product as a brand fact: Brand: founded by a chef. What it must not become is "chef's recipe" or "restaurant-grade" on the sauce. The page never said that. The founder's background describes the company; the product makes no claim of its own.
That's the mirror principle from section 4.3 of the IEO Methodology, applied to the brand object described in section 10.1: identity, provenance, manufacturing facts, and the evidence behind them.
Check what's explicit, what's missing and what contradicts itself on one of your products → Run the free audit
To go deeper → The IEO Methodology, sections 4.3 and 10.1 (doi.org/10.5281/zenodo.22416784) · nsolvia.com/ieo
Sources: Google Merchant Center Help, "Document link [document_link]" · Shopify, "Agentic-Ready Product Data" and Shopify Help Center, "Shopify Catalog and product discovery for agentic storefronts" · Google Merchant Center product data specification and Meta "Catalog Fields" (material and brand attributes).
← A Shopper Asked Google About Your Product. The Answer Was Wrong. · ⌂ IEO · Next: Your Product Page Is Perfect. The Sale Fails at Shipping and Returns. →
Series: Interpretability Engine Optimization · Knowledge Domain: Interpretability (IEO)
Interpretability Engine Optimization (IEO)
The UNDERSTAND Layer
nsolvia.com/ieo