AEO and GEO for Shopify stores: the settings, data and pages that get you cited

Most advice on answer engine optimisation (AEO) and generative engine optimisation (GEO) is written for any website. A Shopify store is not any website. Your theme writes some of your structured data for you, Shopify manages your robots.txt, your product categories come from a fixed taxonomy, and a handful of admin settings decide which AI channels see your catalogue at all.

This guide walks through AEO and GEO for a Shopify store in the order an AI engine meets it: can it reach your pages, can it read them, does it trust the facts, and which channels carry your catalogue directly. Every Shopify, Google and OpenAI detail below comes from their own documentation, checked on 27 October 2026.

The short version

• Check crawl access first. OAI-SearchBot decides whether your pages can show up in ChatGPT search. GPTBot is only about model training. You can allow one and block the other in robots.txt.liquid.

• Your theme probably outputs product markup already. Shopify's structured_data Liquid filter turns a product into schema.org Product or ProductGroup JSON-LD. Check that it is there and that it matches the visible page.

• Pick a standard product category for every product. Category metafields, which Shopify calls product attributes, are the structured facts such as fabric, size and neckline that engines compare.

• Keep Merchant Center current. Google says there is no special markup for AI Overviews or AI Mode, but it does recommend up-to-date Merchant Center information. The Google & YouTube app syncs it from Shopify.

• Measure weekly, by prompt. AI answers move week to week. Judge progress over several weekly runs, not one answer.

How an AI engine reads a Shopify store

An engine that answers "best linen shirt under ₹3,000" has two ways to learn about your products. It can crawl and index your public pages like a search engine, or it can receive your catalogue as structured data through a feed or a channel such as Shopify Catalog. Most engines use both. That gives a Shopify store four layers to get right, and a gap in any one of them shows up in the answer.

The rest of this guide takes the layers in order. If you want the wider strategy first, read our guide to how answer engines find, choose and quote products.

Layer 1: crawl access and robots.txt.liquid

Every Shopify store gets a default robots.txt. Shopify's help centre says it allows public content such as products, collections, pages, blogs and policies, and disallows admin, cart, checkout, account and order pages, along with filtered and sorted collection URLs that duplicate content. For most stores that default is right for AI engines too.

You can change it by adding a robots.txt.liquid template under Online Store > Edit code. Shopify recommends adding or removing rules with Liquid rather than replacing the file with plain text, so Shopify can keep updating the defaults. It also warns that Shopify Support cannot help with edits to this file and that incorrect use can lose all of your traffic. One more detail matters here: an older robots.txt.liquid template can keep outdated rules such as Disallow: /search, and deleting the template restores the current default.

For ChatGPT, know which crawler does what. OpenAI's crawler documentation says:

• OAI-SearchBot is used to surface websites in ChatGPT's search features. Sites that opt out are not shown in ChatGPT search answers, though they can still appear as navigational links.

• GPTBot crawls content that may be used to train OpenAI's foundation models. Blocking it does not remove you from ChatGPT search.

• ChatGPT-User visits a page when a person asks ChatGPT to, and is not used to decide what appears in search.

OpenAI says a robots.txt change can take about 24 hours to reach search. If your store sits behind Cloudflare or another proxy, check its bot rules as well: Google's guidance for AI features asks site owners to allow crawling in robots.txt and in any CDN or hosting setup. Shopify also notes that blocking AI crawlers in robots.txt does not stop Shopify Catalog from sending product data to agentic storefronts you have turned on, so robots.txt and channel settings are two separate switches.

Layer 2: product markup your theme already writes

Shopify's structured_data Liquid filter converts a product or article object into schema.org JSON-LD. Shopify's developer documentation says a product with no variants is output as a schema.org Product, and a product with one or more variants as a ProductGroup. Many current themes call this filter on product pages, so a store can have valid markup without anyone writing it.

Three checks are worth five minutes:

1. Is there exactly one product block? Open a product page, view the source and search for application/ld+json. An old app and a theme can both print product markup, which gives engines two versions of the same price.

2. Does it match the page? Google's AI features guidance recommends that structured data match the visible text. A sale price in the markup and a full price on the page is the kind of mismatch AI engines repeat.

3. Are shipping and returns covered? Google's merchant listing documentation lists shippingDetails and hasMerchantReturnPolicy as recommended offer properties, and recommends a single store-wide shipping and returns policy under Organization markup, with offer-level details only for exceptions.

Google's merchant listing documentation also requires a name, an image and an offer with a price above zero and a currency, and recommends brand, description, GTIN, SKU, availability and reviews. For variants, it recommends isVariantOf pointing to the ProductGroup. If your theme's markup is missing these, the fix is usually in the product data, not in code: the filter can only print what the product record holds.

Layer 3: product data, categories and attributes

This is where most Shopify stores lose AI answers. An engine comparing two linen shirts needs to know that one is 100% linen, relaxed fit and machine washable. If that lives only in a product photo or a long paragraph, the engine may not find it, or may fill the gap with a guess.

Shopify gives you a structured place for these facts. Every product can have one category from Shopify's Standard Product Taxonomy, such as Apparel & Accessories > Clothing > Clothing Tops > Shirts. Each category comes with category metafields, which the taxonomy calls product attributes: for a shirt, size, neckline and fabric. Shopify's help centre says these attributes help shoppers and search engines discover products, and that the category is also what channels such as Google and Facebook use for a standard product type.

A practical order of work:

• Assign a category to every active product. Shopify lets you filter the product list by category in the admin, which makes the uncategorised ones easy to find.

• Fill the attributes shoppers ask about. Material, size range, fit, colour and care for fashion; ingredients and skin type for beauty; metal, purity and weight for jewellery.

• Write titles that say what the product is. "Aria" tells an engine nothing. "Aria 100% linen relaxed shirt, sage" does.

• Put answers on the page as text. Size guides, delivery times and return windows as text, not inside images. Google lists making important content available in text among its recommended practices for AI features.

Our post on how to get your products recommended by ChatGPT covers product page copy in more depth.

Layer 4: the channels that carry your catalogue

Crawling is only half of it. Shopify's help centre describes agentic storefronts as the way shoppers discover and buy your products in AI channels, active by default for eligible stores. It lists ChatGPT, where shoppers check out on your own online store, plus Google AI Mode and Gemini, Microsoft Copilot and Meta surfaces through Shopify's direct checkout if you turn it on. Products reach these channels through Shopify Catalog. The same page says a knowledge base supplies store information such as policies and brand details, so agents can answer shoppers' questions.

For Google, products can also arrive through the Google & YouTube sales channel or Merchant Center. Shopify says the Google & YouTube app syncs your products and store information to Google Merchant Center so you can manage listings from the Shopify admin, and that it is available only in certain countries and currencies. Google's own guidance for AI Overviews and AI Mode says there are no extra requirements to appear, no special schema to add and no new AI text files to create, but recommends keeping Merchant Center and Business Profile information up to date.

Two settings to check in the admin, under Sales channels:

• Agentic: which AI channels are on, and whether your policies and brand details are filled in for the knowledge base.

• Google & YouTube: whether products are approved in Merchant Center, and any disapproved items with their reasons.

What about llms.txt and agents.md?

Shopify now serves an /agents.md file for every store, written for shopping agents, and its developer documentation says most stores need nothing more than the managed default. You can customise it, or /llms.txt and /llms-full.txt separately, with Liquid templates added under Edit code. These templates run in a restricted context: store data comes from an agents object that exposes values such as the store URL, MCP endpoint, currency and sitemap.

Treat these files as a short, accurate summary of the store, not as a ranking lever. Google says plainly that you do not need AI text files to appear in its AI features. Open your own /agents.md, read it, and make sure nothing in it contradicts your policies.

A 30-minute Shopify AEO check

1. Open yourstore.com/robots.txt. Confirm OAI-SearchBot and Googlebot are not blocked, and that there is no stale Disallow: /search.

2. View the source of your three best-selling product pages. Confirm one product JSON-LD block each, with the same price, stock and currency as the page.

3. List products without a standard category. Count them.

4. Open one product in each top category and count the empty attributes.

5. Read your shipping, returns and COD policies. Are delivery times and return windows written as plain sentences an engine can quote?

6. Check Sales channels > Agentic for which AI channels are on, and Google & YouTube for disapproved products.

7. Ask ChatGPT and Google AI Mode two questions a shopper would ask in your category. Note who is named, what is said about you, and which pages are cited.

Step 7 is the one most stores skip, and it is the only one that shows the result. One answer is noise. Run the same questions every week for a month before deciding what moved. Our five-part AEO strategy explains how to choose which prompts to track.

Where Eldor fits

The checks above tell you whether AI engines can read your Shopify store. They do not tell you what the engines actually say. Eldor installs as a native Shopify app and closes that loop. It runs the prompts your shoppers ask on each AI engine every week, records whether your brand is named, linked or recommended and which sources are cited, and compares every claim about price, stock and policies with your live store. Then it fixes the data behind the wrong answers, filling missing attributes and rewriting thin product pages, and checks schema, feeds and quotable policies as part of agent readiness. Tracking can be split by market (US, Europe, India) and by plan (logged out, free and paid), in clean sessions with no memory.

The quickest way to see your starting point is the free Eldor AI audit: 25 shopper prompts on ChatGPT and Google AI Overviews, 100 of your products checked against the live store, and a fix plan ranked by revenue at stake. For how to run your own version first, see our free AI visibility audit guide.

FAQ

Do Shopify stores need special code for AEO or GEO?

Usually not. Many Shopify themes already output product JSON-LD through the structured_data Liquid filter, and Google says there is no special schema needed for AI Overviews or AI Mode. The bigger gains come from complete product data, categories and attributes, and policies written as text.

Should a Shopify store block GPTBot?

That is a training choice, not a visibility one. OpenAI says GPTBot crawls content for model training, while OAI-SearchBot decides whether pages can appear in ChatGPT search. You can block GPTBot and still allow OAI-SearchBot in robots.txt.liquid.

Does Shopify's structured data include variants?

Yes. Shopify's documentation says the structured_data filter outputs a product with no variants as a schema.org Product and a product with variants as a ProductGroup.

What are category metafields in Shopify?

They are the attributes that come with a product category from Shopify's Standard Product Taxonomy, such as size, neckline and fabric for a shirt. Shopify says they help shoppers and search engines discover products.

Does a Shopify store need llms.txt to show up in AI answers?

No. Shopify already serves an agents.md file for every store, and Google says you do not need AI text files to appear in AI Overviews or AI Mode. Keep the file accurate, but put the effort into product data.

How often should I check my Shopify store's AI visibility?

Weekly. AI answers change from week to week, so run the same prompts every week and judge the trend over several runs rather than reacting to one answer.

Keep reading

• AEO for D2C brands: how answer engines find, choose and quote your products

• How to get your products recommended by ChatGPT

• AEO for jewellery brands

Sources: Shopify Help Center, "Editing robots.txt.liquid", "Product category" and "Agentic storefronts"; Shopify Help Center, "Google & YouTube"; Shopify developer documentation, "structured_data" Liquid filter, "agents.md.liquid" and "llms.txt.liquid"; OpenAI developer documentation, "Overview of OpenAI crawlers"; Google Search Central, "AI features and your website" and "Merchant listing (Product, Offer) structured data". All checked 27 October 2026.