The Shopify AI Commerce Readiness Checklist: July 2026 Edition

July 04, 2026
The Shopify AI Commerce Readiness Checklist: July 2026 Edition

The Shopify AI Commerce Readiness Checklist: July 2026 Edition

By Steve Merrill, Founder of WRKNG Digital | July 4, 2026

Most Shopify stores are invisible to AI shopping assistants right now — not because their products are bad, but because their data is wrong. This checklist tells you exactly where you stand and what needs fixing before the window closes.


1. Shopify Catalog synced to OpenAI

ChatGPT Shopping does not auto-discover your store. You have to connect your catalog explicitly through the OpenAI plugin/shopping integration or through a verified merchant connection. If you haven't done this, you don't exist in ChatGPT Shopping results. Pass looks like: your products appear when you search them by name in ChatGPT with Shopping enabled.

2. Product schema on every product page

Homepage schema doesn't count. AI crawlers — and Google's indexing pipeline — read schema on individual product pages. Pull up any product URL and run it through Google's Rich Results Test. Pass looks like: valid Product schema with name, image, description, price, and availability on every URL.

3. All products have GTIN/barcode entered

GTINs (Global Trade Item Numbers) are how AI shopping systems match your product to a known item in their databases. Without them, your product is an orphan. According to Google Merchant Center documentation, missing GTINs disqualify products from Shopping ads and AI-surfaced results. Pass looks like: every product with a barcode has that barcode entered in Shopify's "Barcode (ISBN, UPC, GTIN)" field.

4. Product descriptions are 150+ words with attributes

AI assistants pull from your product description to answer user questions. "Soft and comfortable" doesn't cut it. We've audited over 2,400 Shopify products — the ones that get recommended have descriptions that include materials, dimensions, fit, use cases, and care instructions. Pass looks like: every top-50-revenue product has a description of 150+ words with specific, searchable attributes.

5. Sitemap submitted to Google Search Console

Google AI Mode and AI Overviews both depend on Google's index. If your sitemap isn't submitted — or has errors — pages get skipped. Log into Google Search Console and check the Sitemaps report. Pass looks like: sitemap submitted, no coverage errors, and all key product and collection URLs indexed.

6. Site loads in under 2 seconds (Core Web Vitals passing)

Speed isn't just a user experience issue. It's a ranking signal for AI-driven results. Slow sites get deprioritized. Run your homepage and a product page through Google PageSpeed Insights. Pass looks like: LCP under 2.5 seconds, CLS under 0.1, and an overall "Good" Core Web Vitals rating in Search Console.

7. ACP (Agentic Commerce Protocol) enabled

ACP is the emerging standard that lets AI agents browse your store, check inventory, and initiate checkout — without a human clicking through every page. Shopify began rolling out ACP support in early 2026. Pass looks like: ACP is enabled in your Shopify admin (check under Settings → Integrations → Agentic Commerce) and your store responds correctly to agent browsing requests.

8. Shop Pay enabled as checkout option

When an AI agent completes an agentic checkout, it needs a frictionless payment path. Shop Pay is the default target for Shopify's agentic checkout pipeline. Without it, agent-initiated purchases either fail or fall back to a slower flow that most agents abandon. Pass looks like: Shop Pay is active and showing as a checkout option on your product and cart pages.

9. Reviews and ratings visible in product schema

I've watched ChatGPT and Perplexity recommend products with 4.7-star ratings over identical products with no ratings listed — every time. AI assistants use ratings as a quality signal. Your reviews need to exist in your product schema, not just on the page visually. Pass looks like: your Rich Results Test shows aggregateRating with ratingValue and reviewCount populated.

10. Return policy is in machine-readable text

A surprising number of Shopify stores have their return policy as a scanned PDF or an image. AI agents can't read either. They read HTML text. If your return policy page is a scanned document or locked in an image, you're invisible to the agents shopping on behalf of your customers. Pass looks like: your return policy is plain HTML text on a dedicated page (e.g., /policies/refund-policy), with the return window and conditions clearly stated.

11. Merchant Center feed is connected and error-free

Google AI Mode's Universal Cart feature — now broadly available as of June 2026 — pulls product data directly from Google Merchant Center. Errors in your feed mean missing products. Log into Google Merchant Center and check the Diagnostics tab. Pass looks like: feed approved, fewer than 1% of products with errors, and Shopping tab eligibility confirmed.

12. llms.txt file created at root domain

llms.txt is the emerging standard for telling AI crawlers who you are, what your site covers, and what they're allowed to use. Think of it as robots.txt for language models. It won't transform your rankings overnight — but stores without it are leaving context on the table that their competitors are starting to use. Pass looks like: a valid llms.txt file exists at https://yourdomain.com/llms.txt with a clear site description and content permissions.


Updated for July 2026: What's Changed

Two things shifted significantly in Q2 2026 that make this checklist different from anything published earlier this year.

ChatGPT Ads launched in May 2026. OpenAI opened a paid advertising product inside ChatGPT Shopping. This means organic product visibility now competes with paid placements. Stores with clean catalog data and explicit ChatGPT connections have a head start on both the organic and paid layers. Stores that haven't connected yet are starting from zero on both fronts.

Google AI Mode Universal Cart went broadly available in June 2026. This is the moment Google's AI search becomes a transaction layer, not just a discovery layer. When a user finds your product in an AI Overview and can add it to a cart without leaving Google, your Merchant Center feed quality becomes your storefront. Errors there are revenue lost directly.

Those two changes are why items 1 and 11 on this list moved from "nice to have" to "non-negotiable."


Frequently Asked Questions

How many of these 12 items should my store already have in place?

If you're a Shopify store doing meaningful revenue, you should have items 5, 6, 8, and 10 covered already — those are fundamentals. Items 1, 7, and 12 are new enough that most stores score zero on them. Scoring 7 or above puts you ahead of the majority of stores your size.

Does Shopify automatically handle any of this?

Some of it. Shopify generates a sitemap automatically and enables Shop Pay by default. Product schema is generated by most themes — but it's often incomplete. GTINs, descriptions, review schema, llms.txt, and the OpenAI catalog connection all require manual action.

What's the biggest mistake stores make on this checklist?

Assuming their Shopify theme handles schema. It doesn't — not fully. Most themes output a basic Product type but skip aggregateRating, gtin, and proper offers structure. You need to verify with the Rich Results Test, not assume.

How often should I run this audit?

Quarterly at minimum. The AI Commerce space is moving fast. ChatGPT Ads didn't exist at the start of 2026. Google AI Mode Universal Cart wasn't broadly available six months ago. What passes today may be insufficient by October.

Is llms.txt actually indexed by AI crawlers yet?

Some crawlers are reading it, some aren't. Perplexity has documented support. OpenAI's crawler behavior around it is still inconsistent. But the cost to create one is 30 minutes of work — and the upside is real as adoption grows. Worth doing now.


Want this audit done for you? Get your free AI Commerce Readiness audit at WRKNG Digital.

Back to Blog