How to Get Your Shopify Products Recommended by ChatGPT

Shopify gets a lot right out of the box. Most themes output product data that AI tools can read, and Shopify fills in a page description for you if you leave it empty. So it's tempting to assume a Shopify store is "AI-ready" by default.

Our numbers say: partly. The technical basics are usually there. The words often aren't. And words are what ChatGPT matches a shopper's question against.

This guide covers what AI shopping assistants read in a Shopify store, what we found in 541 UK Shopify stores in October 2026, and what you can fix yourself.

What we checked, and how

Between 9 and 11 October 2026 we ran a live audit on UK online shops. For each Shopify store we read the public product list (up to 1,000 products), then opened up to 25 product pages spread across the catalogue and looked at what's in the HTML: product data, meta descriptions, descriptions and names.

This article uses the 541 Shopify stores where the audit was reliable (at least 20 product pages read, not blocked): 257,633 products and 13,438 product pages. They're independent UK shops from our own outreach list, mostly 10 to 2,500 products, so read the numbers as "small and mid-sized UK Shopify stores", not every store on the platform. The full write-up, with WooCommerce for comparison, is in our UK data report.

What we found:

• 79.9% of Shopify stores had product data AI can read on every product page we opened. That's the good news
• But 15.2% had none on any page we opened
• 32.5% had thin descriptions (under 150 characters) on at least 15% of their products, and 13.9% on at least half
• 19.3% of all products had descriptions that thin
• 23.8% of stores had at least one product named with a stock code
• Only 6.3% of product pages had no meta description, because Shopify fills one in automatically

Compared with WooCommerce, Shopify stores were in better technical shape. Only about a third (35.7%) had at least one big gap we look for, against 60.5% of WooCommerce stores. When a Shopify store does have a gap, it's usually in the words.

How ChatGPT picks products

When someone asks ChatGPT, Google AI Mode or Perplexity "what's a good gift for a coffee lover under £40?", the assistant looks for products that clearly match, then picks the ones it can describe with confidence.

Matching happens on words: the title, the description, tags and specs. Confidence comes from facts it can check: price, stock, materials. Shopify themes handle a lot of the facts. Nobody handles the words for you.

If you're new to all this, our explainer on what AI commerce is covers the background.

Step 1: Write product titles that say what the product is

In 23.8% of the Shopify stores we checked, at least one product title started with an internal code. That's lower than WooCommerce, but it still happens a lot with imported or wholesale catalogues.

A title AI can match says what it is, plus the details people search by:

• "MUG-12 Speckle" becomes "Handmade Speckled Stoneware Mug, 350ml"
• "TEE-BLK-L" becomes "Organic Cotton T-Shirt - Black, Large"

Keep codes in the SKU field on the variant, where Shopify expects them.

You can do this yourself: in Products, use bulk edit to change titles in a spreadsheet-style view.

Step 2: Give every product a real description

This is the main Shopify gap. A third of the stores we checked (32.5%) had at least 15% of products with descriptions under 150 characters, roughly one sentence.

One sentence is enough for a person who can see the photo. It isn't enough for an AI that needs to answer "is it dishwasher safe?" or "does it come gift-wrapped?" before it recommends anything.

A good description covers:

1. 1What it is and who it's for, in the first sentence
2. 2Materials, size, capacity, fit, and anything people compare
3. 3How and where it's used
4. 4Care, delivery or gifting details

Write it like you'd explain the product to a customer in person. A few short paragraphs is plenty.

You can do this yourself. For the first ten best sellers, you should. Past that, it's a lot of hours.

Step 3: Use tags and product types properly

Shopify tags and product types are easy to ignore, but they help both your own store search and the tools that read your catalogue. Add the words shoppers actually use: the occasion ("gift for him"), the material ("linen"), the style ("minimalist").

You can do this yourself with bulk edit.

Step 4: Check your product data

Most Shopify themes output product data in the page (Schema.org Product markup, as JSON-LD) with the title, price, currency and stock. That's why 79.9% of stores in our sample had it on every page we checked.

The 15.2% that had none usually fall into one of these groups: a heavily customised or older theme, a page builder app that replaces the product template, or a headless setup. To check your store:

1. 1Open a product page
2. 2Paste the URL into Google's Rich Results Test (free)
3. 3Look for a "Product" result with a price and availability

If it's missing, try previewing a current Shopify theme such as Dawn on a test product, or ask your theme developer. Several SEO apps in the Shopify App Store also add or extend product data.

This one lives in your theme. Pickastor doesn't change your theme or add markup to your pages, but the free AI Score checks your product pages and tells you whether the data is there.

Step 5: Write SEO titles and meta descriptions on purpose

Shopify auto-fills a meta description from the start of your product description when you leave the "Search engine listing" fields empty. That's why only 6.3% of product pages we checked were missing one.

But an automatic excerpt of a thin description is still a thin summary. If your description starts with "Free UK delivery on orders over £50", that's what search engines and AI tools see as the page summary.

You can do this yourself: on each product, scroll to "Search engine listing" and write a page title (about 60 characters) and description (about 150 characters) that say what the product is and why someone would want it.

Step 6: Let AI crawlers in

Shopify's default robots.txt is reasonable, but check two things:

• Visit yourstore.com/robots.txt and make sure AI crawlers such as OAI-SearchBot and GPTBot (OpenAI), PerplexityBot and Googlebot aren't blocked. Some apps and custom robots.txt edits add blocks
• Password protection. A store that's still behind a password page is invisible to every crawler

You may also hear about llms.txt, an optional file that gives AI tools a short summary of your site. It's a nice extra, not a fix for anything above.

Step 7: Get your catalogue into product feeds

Feeds hand your whole catalogue to shopping platforms in one go. On Shopify, the free Google & YouTube app connects your products to Google Merchant Center, which powers Google Shopping and Google's AI features. OpenAI has also opened a way for merchants to share product feeds with ChatGPT; check chatgpt.com/merchants for the current status. Shopify itself has been adding ways to share catalogues with AI assistants, so check your admin for new sales channels.

Feeds deliver whatever is in your product data. Clear titles and real descriptions make every feed better at the same time.

What you can do yourself, and where a tool helps

Quick to do yourself:

• Check product data with the Rich Results Test
• Check robots.txt and password protection
• Install the Google & YouTube app and connect Merchant Center
• Write proper SEO titles and descriptions for your top products

Slow by hand (it grows with your catalogue):

• Renaming coded titles
• Rewriting thin descriptions
• Adding useful tags
• SEO titles and meta descriptions for every product

The median store in our Shopify sample had about 420 products. Even at 15 minutes a product, rewriting the thin ones is days of work.

How Pickastor fits in

Pickastor does the slow list. You connect your store through Pickastor's approved Shopify app, choose products, and Pickastor rewrites each one: a clear title, a detailed description, tags, product attributes and the phrases shoppers type into AI tools, plus an SEO title and meta description. You review every change side by side, and nothing changes in your store until you push it. On Shopify the push updates the title, description, tags and the SEO title and meta description.

Every optimized product also goes into an AI product feed for your store (XML in Google Merchant Center format, plus JSON).

What it doesn't do: it doesn't edit your theme or add Schema.org markup, it doesn't create llms.txt or change robots.txt, and it doesn't touch image alt text, brands, reviews or barcodes. The steps above for those are yours, and most of them are quick.

Your first 3 products are free; after that you pay once by product count, no subscription. See pricing, or run the free AI Score first. If you also run a WooCommerce store, here's the WooCommerce version of this guide.

A one-hour plan

1. 1Run the free AI Score
2. 2Test one product page in the Rich Results Test
3. 3Check yourstore.com/robots.txt
4. 4Fix the titles of your five best sellers
5. 5Rewrite your best seller's description properly, and its search engine listing

Then work down the list, best sellers first. AI recommendations move like SEO: changes take weeks to show, and nobody can guarantee a result. But a Shopify store with clear titles and real descriptions gives ChatGPT something to work with, and most of your competitors haven't done it yet.