Your Shopify store isn't showing up in AI search — what to check

You typed "best wool base layers in Sweden" into ChatGPT, or asked Perplexity which store sells the thing you sell. It answered with three confident recommendations and a tidy list of links. None of them were you. You know your range is better, your prices are fair, your reviews are real. The model has simply never heard of you.

This is becoming the new "page two of Google", except worse, because there is no page two. The AI gives one answer. If your Shopify store isn't in it, you don't exist for that shopper. The good news is that the reasons a store goes missing are a short, concrete list, and most of them are things you can check yourself this afternoon without touching a line of code.

Why Shopify stores go invisible to AI

Five things keep a store out of AI answers. Usually it's one or two of them, not all five.

1. Your product facts are trapped in JavaScript or theme widgets. Many Shopify themes render the price, variant options or full description with JavaScript, or stuff key information into a slider, tab or image. AI crawlers are not browsers. Most of them fetch your raw HTML and don't run scripts. If the price only appears after the page "loads" in your eyes, the crawler saw a blank space where the price should be. This is a rendering problem, and it's the single most common reason an otherwise good store is invisible.

2. Your Product schema is missing or half-built. Structured data is the labelled summary that tells a machine "this is a product, it costs 499 SEK, it's in stock, it has 4.6 stars". Some Shopify themes ship complete Product and Offer markup. Plenty ship none, or ship it with the price and availability fields left empty. AI answer engines lean hard on this markup to pull a clean, quotable fact. No schema, nothing clean to quote.

3. You're blocking the AI crawlers. A robots.txt rule, or an app you installed in a "block the bots" panic, may be turning away the exact crawlers that build AI answers, like GPTBot, OAI-SearchBot, PerplexityBot and ClaudeBot. If the crawler can't fetch the page, it can never cite it. The same blocked-bot mistake is one of the four reasons in why ChatGPT isn't citing your website, which is worth a read if your store sits behind one of these rules.

4. Your descriptions are too thin to answer real questions. Shoppers ask AI specific things. "Is this jacket warm enough for skiing?" "Does it run small?" "Can I machine wash it?" If your product page is three lines of marketing fluff and a spec table, there's no sentence for the model to lift in answer to those questions. Thin copy is invisible copy.

5. The store simply isn't indexed or ranking yet. AI search retrieves from conventional search indexes underneath. A new domain with little authority that doesn't yet rank for "merino base layer Stockholm" won't get pulled into the candidate set the model reads from. No ranking, no retrieval, no mention. This one takes the longest to fix and there's no shortcut around it.

Check your store in five minutes

You don't need a developer for any of this. Open one of your best-selling product pages and work through it.

Try to select the price and description with your cursor. Click and drag across the price, then across the description, as if highlighting text to copy. If it highlights like normal text, good, a crawler can read it. If it won't select, or it behaves like a picture, the information is probably an image or JavaScript-only, and the crawler sees nothing there.

View the raw page source. Right-click the product page and choose "View Page Source" (not "Inspect"). Press Cmd+F or Ctrl+F and search the source for your product's price and a sentence from your description. If they're in there, the crawler can read them. If the source is mostly empty <div> tags and your content is nowhere, it only appears after JavaScript runs, and most AI crawlers never see it.

Run the page through a structured-data validator. Paste the product URL into Google's Rich Results Test or schema.org's validator. Check that it finds a Product with a filled-in Offer, including price, priceCurrency and availability. Empty or missing fields are your answer.

Read your robots.txt. Go to yourstore.com/robots.txt and search the text for GPTBot, OAI-SearchBot, PerplexityBot and ClaudeBot. A Disallow: / under any of those user-agents means you've shut that engine out. Our robots.txt explainer covers which bots do what.

Ask the AI to recommend a store like yours. In ChatGPT with browsing on, and in Perplexity, type the question your customer would type: "best [your product category] in [your country]". See who gets named. If it's three competitors and not you, you now know your content exists in the model's reach and you're simply not in the shortlist, which points at causes 4 and 5. Perplexity even lists the sources it leaned on, so why Perplexity cites your competitors but not you is a useful way to read exactly what those rival stores did better.

The fixes

Work top to bottom. The early ones are quick and clear the way for the rest.

1. Put real, selectable text on the page. Make sure the price, variant info and the heart of your description live in plain HTML, not in an image, a PDF, or a widget that only fills in after scripts run. In Shopify this often means editing the description in the product editor rather than relying on a theme app to inject it, and avoiding image-based "spec sheets". The rule of thumb: if you couldn't select it with your cursor in the check above, the crawler couldn't read it either. The companion piece on making content legible to AI goes deeper on this.

2. Get complete Product and Offer schema on every product page. Your theme may already do this. If the validator showed gaps, fix the source. Some themes expose schema settings; others need a schema app or a small theme.liquid edit. If you want to see what a full, correct Product block looks like before you wire it up, our schema generator builds one you can compare against. Make sure price, priceCurrency, availability and aggregateRating are actually populated, not left as empty placeholders.

3. Allow the AI crawlers. If your robots.txt blocks the bots that build answers, remove those disallows. You can keep blocking a training-only bot like GPTBot if you object to your catalogue training models, but understand that's a separate decision from being cited. The retrieval bots, OAI-SearchBot, PerplexityBot and ClaudeBot, are the ones that put you in answers. Don't lock them out by accident.

4. Write descriptions that answer real buyer questions. Go back to the questions shoppers actually ask and answer them in the copy, in plain sentences. Sizing, materials, care, what it's good for, what it isn't. A description that says "runs true to size, machine washable at 30°C, warm enough for autumn hiking but not deep winter" gives the model something to quote. A wall of adjectives gives it nothing.

These compound. Real text and schema make a page readable; allowing the bots lets them in; rich descriptions give them something worth repeating. Ranking and authority build over months on top of that, but the first three you can finish this week.

A faster way to find the gap

Doing this by hand, page by page, across ChatGPT, Perplexity, Claude and Google's AI answers, then re-checking after every theme update, is the part nobody keeps up with. Rankport's AI visibility checker runs the whole check on your store automatically. It reads your robots.txt for each crawler, confirms your product pages are server-rendered and your text is selectable, validates your Product and Offer schema, and shows which engines mention you and which don't, with the fix for each spelled out in plain English. If your Shopify store is missing from AI search, it tells you exactly why.