MACHINE 404 / Shopify

Is your Shopify store readable by AI?

Shopify starts higher than anything else we measure, and it still loses the same points as everyone else. The difference is that on Shopify the fixes are mostly settings and theme edits rather than development, which makes it the cheapest platform to put right.

What you already have, for free

Product pages emit structured data with a price and an availability by default. That is the heaviest check in the scan, worth 12 points, and Shopify hands it to you without anybody doing anything. Themes also emit breadcrumbs, and the platform produces a sitemap and a valid robots.txt on its own.

There is also an endpoint an agent can transact against. Most stores have never switched it on, because nothing in the admin tells you it matters. That is 8 points sitting behind a toggle.

What Shopify never does for you

It describes products. It does not describe the company. Legal name, contact point, area served and the profiles that prove you are one business rather than three similar ones: none of that appears unless somebody adds it. That is the single biggest identity block in the rubric and the most common gap we measure on otherwise healthy stores.

Nor does it answer questions. A FAQ written into a page as prose is invisible; the same questions as data are quotable back to a customer, and they are worth 6. And the AI crawler policy is a decision nobody makes: the default robots.txt names none of them.

The apps question

There is an app for most of this, and some are good. Two warnings from measuring real stores. First, two apps writing structured data produces two conflicting descriptions of the same shop, which is worse than one missing description. Second, an app that injects its markup with JavaScript is invisible to the crawlers that do not run it, which is most of them. Check the page source, not the app's dashboard.