The Shopify Agentic Commerce Readiness Checklist: What We Test in the First Hour

November 13, 2026

By Steve Merrill · November 13, 2026

What Is a Shopify Agentic Commerce Readiness Checklist?

It's a short set of tests that show whether an AI shopping agent can find, understand and buy from your store. We run six of them in the first hour of every audit: crawler access, product copy, structured data, variants and inventory, checkout, and policies.

Each test has a time box. If you can't finish one in the time given, that tells you something too.

Why Test Before You Change Anything?

Because most stores guess wrong about what's broken. Owners assume it's the theme or the ad spend. In the audits we run, the cause is usually sitting in robots.txt or in a product description that says almost nothing.

OpenAI has published the Agentic Commerce Protocol, built with Stripe, to let merchants sell inside ChatGPT. Shopify has been building toward the same idea on its side. Agents are going to read your catalog whether you prepare for them or not.

Test 1: Can AI Crawlers Reach Your Store? (10 Minutes)

Open yourstore.com/robots.txt and read it. Shopify generates a default file, and you can customize it through the robots.txt.liquid template documented in the Shopify Help Center.

Look for rules that block GPTBot, OAI-SearchBot, PerplexityBot or Googlebot on product and collection paths. OpenAI lists its crawlers in its bot documentation, and Perplexity does the same in its crawler guide.

I always run this one first. A blanket disallow added by an old app or a well-meaning developer ends the whole conversation before it starts.

Test 2: Do Your Product Descriptions Say Enough? (15 Minutes)

Pick ten products. Copy each title and description into a plain text file and read it with no images.

Can you tell what it is, who it fits, what it's made of, and when someone would use it? A shopper asking an agent for "a waterproof daypack that fits a 15-inch laptop" needs those words to exist on your page.

Thin copy is the most common failure we see. Titles like "Classic Tee – Black" give an agent nothing to match against.

Test 3: Is Your Product Structured Data Valid? (10 Minutes)

Run three product URLs through Google's Rich Results Test. Most Shopify themes output Product markup, but the quality varies by theme and by installed apps.

Check for price, priceCurrency, availability, brand and a GTIN or MPN where one applies. Google's Product structured data documentation lists which properties are required and which are recommended. The vocabulary itself lives at schema.org/Product.

Watch for duplicate Product blocks. Two apps writing two versions of the same price is a real problem.

Check the Offer block too. If availability says InStock on a sold-out product, an agent will believe it and send a shopper to a dead end.

Test 4: Do Variants and Inventory Match Everywhere? (10 Minutes)

Pick a product with several sizes or colors. Compare what the storefront shows, what the structured data says, and what your Google Merchant Center feed reports.

Any mismatch is a trust problem. An agent that quotes a price or an in-stock status that turns out to be wrong will stop recommending you. Fix the source, then let the other channels catch up.

Test 5: Can an Order Complete Without a Human? (10 Minutes)

Place a real test order from a product page. Count the interruptions: forced account creation, newsletter pop-ups, shipping that needs a manual quote, custom fields that need explanation.

Agent-led checkout works best on standard flows. Shopify's own checkout is strong here, so the trouble usually comes from add-ons bolted on top. Remove what you can.

Test 6: Are Your Policies Specific Enough to Quote? (5 Minutes)

Find your shipping, returns and contact pages. They should be linked in the footer and written in plain sentences.

"Returns accepted within 30 days of delivery, unused, in original packaging" is something an agent can repeat. "Contact us for return information" is not. Vague policies cost you the sale when the shopper asks a direct question.

How Do You Score the Results?

Give each test a pass, partial or fail. Anything that fails on crawler access or checkout goes to the top of the list, because those two stop the process entirely.

Partial scores on copy and structured data come next. They take more work, but they're where recommendation quality is decided.

Want a second set of eyes on your results? See how WRKNG Digital approaches agentic commerce for Shopify stores.

Frequently Asked Questions

How long does a Shopify agentic commerce readiness check take?

The first pass takes about an hour: crawler access, product copy, structured data, variants and inventory, checkout, and policy pages. Fixing what you find takes longer, but the audit itself fits in one sitting.

Do I need to change my theme to be ready for AI shopping agents?

Usually not. Most readiness problems live in product data, robots.txt rules, and policy pages rather than in the theme. Theme changes only matter if your product markup is missing or broken.

Should I block AI crawlers on my Shopify store?

If you want products recommended or purchased through AI assistants, blocking their crawlers works against you. Both OpenAI and Perplexity document their bots and the robots.txt rules that control them, so you can choose what to allow.

What is the most common failure in a readiness check?

Thin product descriptions. When a title and a two-line description are all an agent has to work from, it can't match your product to a detailed shopper request.

Back to Blog