AI shopping assistants don't browse your store the way a human does. They pull structured data, check signals, and either recommend you or move on. Most Shopify stores fail that check before a single shopper ever sees them.

We've run AI readiness audits on hundreds of Shopify stores. The same five failure points keep showing up. Here's what they are and what to do about each one.

1. Missing Product Schema Markup

AI shopping tools — ChatGPT Shopping, Google AI Overviews, Perplexity — rely on Schema.org Product markup to understand what you sell, how much it costs, and whether it's in stock. Without it, they're guessing. They don't guess — they skip.

Shopify's default themes generate basic Product schema, but "basic" isn't enough. You need price, priceCurrency, availability, brand, sku, gtin, and ideally aggregateRating. Missing two or three of those fields and your product falls below the threshold AI systems use to surface recommendations confidently.

The fix: Audit your pages using Google's Rich Results Test. For most Shopify stores, a metafield-powered schema snippet added to the product template fills the gaps in under an hour. If you're on a third-party theme, verify the vendor has updated schema since 2024 — many haven't.

2. No Ratings or Review Data

Social proof isn't just a conversion tool. It's a ranking signal for AI. ChatGPT Shopping explicitly surfaces aggregateRating data when comparing products. Stores without reviews get deprioritized — not penalized, just passed over in favor of stores that have them.

According to PowerReviews' 2024 state of ratings report, 45% of consumers won't buy a product with zero reviews. AI models are trained on human behavior. They've absorbed that same bias.

The fix: If you're using a review app (Okendo, Judge.me, Yotpo), verify it's outputting aggregateRating schema on the product page — not just displaying stars visually. Many apps render the visual widget but drop the schema. Open your product page source, search for aggregateRating. If it's not there, the AI can't see your stars.

3. Thin Product Descriptions

AI shopping assistants answer natural language questions. "What's a good running shoe for wide feet under $120?" To surface your product as an answer, the model needs enough text on your product page to match that intent.

I pulled a sample of 200 Shopify stores across apparel, home goods, and sporting equipment. More than 60% had product descriptions under 80 words. That's not enough for an AI to confidently match your product to a specific buying question — especially a long-tail one.

The fix: Start with your top 20 products — the ones driving 80% of revenue. Aim for 150–300 words covering: who the product is for, specific use cases, dimensions or fit, and materials. Write the description like you're answering questions a shopper would ask an AI. Because you are.

4. Missing or Buried Policy Pages

Before an AI assistant recommends your store to a real person, it needs to trust you. Trust signals — return policies, shipping timelines, contact information — aren't just for nervous human shoppers. They're part of what AI systems check when deciding whether a store is safe to recommend.

Google's Merchant Center guidelines list return policy and shipping information as required for full product eligibility in AI-powered shopping features. If those pages don't exist, aren't linked in the footer, or aren't machine-readable, you're invisible to an entire category of AI recommendations.

The fix: Add MerchantReturnPolicy and OfferShippingDetails schema to your product pages. Link policy pages directly from the footer — not buried in an accordion or behind a login wall. Shopify's built-in policy pages handle most of this if they're actually filled out and published.

5. Broken or Uncrawlable Product Feeds

AI shopping platforms — especially those connected to Google Shopping, Microsoft Bing Shopping, and now ChatGPT's shopping index — pull product data from structured feeds, not just live page crawls. If your Google Merchant Center feed has errors, stale pricing, or out-of-sync inventory, AI tools using that feed will surface bad data or skip you entirely.

Google's own Merchant Center documentation notes that feeds with more than 20% item disapprovals can suppress an entire account's eligibility for AI-powered features. That's a real threshold. Most store owners don't check their feeds more than once at setup.

The fix: Log into Google Merchant Center and open the Diagnostics tab. Look for disapproved products, missing attributes, and price mismatches. Most common culprits: missing GTINs for branded products, prices that don't match the live site, and images under minimum size requirements. Fix the top errors first. Then set a monthly calendar reminder — feeds decay as your catalog changes.

How We Chose This List

These five issues came directly from AI readiness audits across hundreds of Shopify stores over the past 12 months. We looked at what signals ChatGPT Shopping, Google AI Overviews, and Perplexity use when surfacing product recommendations — then measured how often those signals were missing or broken.

We left out generic issues like slow site speed or bad photos. Those hurt conversions. These five specifically prevent a store from appearing in AI shopping responses. Every item on this list has a free or low-cost fix. None require a developer retainer.

Frequently Asked Questions

Does Shopify automatically add product schema?

Shopify's default themes add basic Product schema, but it's often incomplete. It typically includes name, price, and availability — but misses brand, GTIN, and aggregateRating. Use Google's Rich Results Test to see exactly what's present on your pages.

How many reviews do I need before AI will recommend my products?

There's no published minimum, but products with no aggregateRating schema are consistently deprioritized. Getting schema in place matters more than raw review count. Start there.

My store has all of this. Why am I still not showing up in ChatGPT Shopping?

Schema and feeds get you eligible, not guaranteed. ChatGPT Shopping also factors in brand mentions across the web and whether your product pages specifically answer the questions shoppers ask. If the technical layer is clean and you're still invisible, the gap is usually content depth or off-site authority.

Do I need to submit my store directly to ChatGPT or Perplexity?

Not directly. ChatGPT Shopping pulls from Bing's shopping index, which feeds from your Google Merchant Center feed. Perplexity crawls the web directly. Fix your schema, fix your feed, keep your pages indexable. That covers most major AI platforms without separate submissions.

How often should I audit my AI readiness?

Quarterly at minimum. Feeds change every time you update inventory. Schema breaks when themes update. Build it into regular store maintenance — not as a one-time project.

What to Do Next

Start with the schema audit — it's free, takes 20 minutes, and usually reveals two or three issues you didn't know were there. Then check your Merchant Center feed errors. Those two steps alone will move most stores from invisible to eligible.

If you want to see exactly where your store stands across all the signals AI shopping assistants check, we built a tool for that.

Get your AI Commerce readiness audit →