AI shopping readiness, defined as six things you can measure
"AI shopping readiness" is the kind of phrase that gets used before it gets defined. Here is a definition you can hold a store to.
A store is ready for AI shopping when an assistant asked for what the store sells can do four things: find the products, read them correctly, trust the store, and reach it. That breaks down into six measurable things. This is the framework the Blunt Findings audit scores, and it is set out here so that you can apply it yourself, with or without the audit.
1Catalogue readability
What it is.
Whether the product data itself, before any page is rendered, says what each product is.
Ready looks like.
Titles that name the brand, the product type and the attribute a shopper would ask for. Descriptions that carry a measurable specification: material, dimensions, weight, capacity, what is in the box, with units. Vendor and product type filled in on every product. More than one image per product, with alt text.
How to measure it.
Open /products.json on your store and read it as if you had never seen the products. Count how many titles say what the thing is. Count how many descriptions state a number with a unit.
Why it is first.
Every other signal is a pointer to this one. A perfect structured-data block around a description that says "built for everyday wear" gives an assistant a well-formed nothing.
2Structured data
What it is.
The Product JSON-LD block in each product page's source.
Ready looks like.
Present on every product page, with price, currency and availability in the offer, a brand, a SKU, and a GTIN on every variant. Ratings where you have them.
How to measure it.
View the source of a product page and search for application/ld+json. Check for "@type": "Product", then for gtin. Repeat for a handful of products from different collections, because a theme change or an app can strip the block from some templates and not others.
3Agent access
What it is.
Whether the crawlers behind the assistants are allowed in.
Ready looks like.
robots.txt allows OAI-SearchBot and PerplexityBot on /products/ and /collections/. The sitemap lists the product pages. An llms.txt at the root names the store and points at the catalogue. Whether to allow training crawlers such as GPTBot and Google-Extended is a separate choice; it does not affect readiness.
How to measure it.
Read /robots.txt and /sitemap.xml. Both are one click away.
4Trust and policy
What it is.
Whether the pages an assistant checks before recommending a store exist, are linked, and can be read by a machine.
Ready looks like.
Shipping, returns, contact, about and privacy pages all exist, are linked from the footer, are not hidden from crawlers, and state facts rather than burying them: dispatch time, delivery time, cost, free-shipping threshold, returns window, how refunds are made, an email address and a phone number.
How to measure it.
Check where your policies live. If they are under /policies/, check whether robots.txt disallows that path. Shopify's default does.
5Speed
What it is.
How quickly the store loads on a phone.
Ready looks like.
A mobile performance score an assistant would not penalise, and pages that do not shift about while loading.
How to measure it.
Google's PageSpeed Insights, on a product page rather than the homepage, because the product page is the one an assistant sends a shopper to.
A note on weight.
This is the section where a store owner has the least direct control and the assistants have said the least. It is worth measuring and not worth panicking over.
6Visibility
What it is.
Whether the store actually appears when an assistant is asked for what it sells.
Ready looks like.
Appearing in answers to buyer-style questions that do not name the brand.
How to measure it.
Ask an assistant eight questions a real shopper would ask about what you sell, half naming a specification you publish, none naming your brand. Note whether you appear and who appears instead. Then treat the result as what it is: a spot check on one day, not a measurement of your share. Assistants change their answers, and eight questions is a sample.
What ready adds up to
A ready store is one where nothing the assistant needs is missing, blocked or vague. That is a lower bar than it sounds, and most stores fail it on things that take an afternoon in Shopify admin to fix. The order above is the order to fix them in, because the first two are what the assistant is trying to read and the rest decide whether it can and whether it should.
Check your own store
The free check reads a Shopify store the way an assistant does and gives five answers in a few seconds: whether the catalogue can be read, whether a product page carries structured data and a GTIN, which AI crawlers are blocked, and whether the policy pages are linked. No login, no email needed for the answers.
The full audit is the same reading done properly: the whole catalogue, fifteen product pages, policies, speed and eight buyer searches, ranked into a fix list with examples from your own products. £99, within 24 hours.