Skip to content

Platforms: Shopify

Best SEO and GEO agency
for Shopify brands.

Shopify gets the technical basics right. What it leaves to you decides who gets recommended: words on your collection pages, complete product data, reviews engines can read and a catalog AI shopping agents trust. Passionfruit does that part.

How Shopify stores get found now

  1. AI agents read your catalog, not just your pages

    Shopify can send product data through Shopify Catalog to AI shopping channels such as ChatGPT, Perplexity and Microsoft Copilot. Thin titles and missing attributes there mean a thin answer, however good the storefront looks.

  2. The shortlist forms before the click

    Questions like “best waterproof hiking boots under $200” are answered in ChatGPT and Google’s AI Overviews before anyone opens a results page. The stores named tend to be the ones whose pages answer the question in plain text.

  3. Reviews and FAQs have to be in the HTML

    The main AI crawlers studied, including GPTBot and PerplexityBot, do not run JavaScript. Review widgets and FAQ apps that load in the browser leave those crawlers reading a product page without its reviews or answers.

What Shopify does for search, and what it leaves to you

Shopify handles more of the technical basics than most platforms. These are the areas we check on every store: what Shopify covers, and what we add.

URLs

Out of the box

Product, collection, page and blog URLs carry a fixed prefix: /products/, /collections/, /pages/ or /blogs/. Only the handle after it can change.

What we add

Handles that match how people search, a 301 for every handle we change, and no time lost fighting prefixes that cannot be removed.

Duplicate product paths

Out of the box

Products also load inside collections, at /collections/name/products/name, with a canonical back to /products/. Many themes still link the longer path.

What we add

Theme links changed to the canonical product URL, so crawlers spend their visits on pages that can rank.

Sitemap

Out of the box

Generated for you, with separate files for products, collections, pages and blogs, and entries per language on Markets.

What we add

A check that every page that should rank is in it, and that filtered, sold-out and duplicate pages are not.

robots.txt

Out of the box

A sensible default that keeps cart, checkout, account and filtered pages out. It can be edited with robots.txt.liquid, which Shopify calls an unsupported customization.

What we add

An audit of any custom file, since older themes can still block pages that should be crawled, and explicit rules for the AI search crawlers you want in.

Structured data

Out of the box

Themes built on Shopify’s structured_data filter output Product markup (ProductGroup for products with variants) and Article markup for blog posts.

What we add

One complete Product graph per page, with brand, GTIN, ratings, shipping and returns merged in, and the duplicate graphs review apps add removed.

AI crawlers and agents

Out of the box

The default robots.txt has no rules aimed at AI crawlers, so they fall under the general ones. Shopify serves an llms.txt for you and opens store data to agents through Shopify Catalog and a storefront MCP endpoint.

What we add

Crawler access checked in the live robots.txt and server responses, and the catalog attributes agents read filled in from your product data.

International

Out of the box

With Markets set up, Shopify handles hreflang, canonicals and a sitemap for each market.

What we add

Titles, descriptions and collection copy written for how each market searches, rather than translated word for word.

Collection pages

Out of the box

A product grid, a title and an optional description field.

What we add

An answer-first introduction, comparison tables and FAQs in the HTML: the copy that collection searches and AI answers draw on.

See where your Shopify site stands.

Book a call and we will check these on your own site, live.

Book a call

How AI engines read a Shopify store

An AI answer about your products can draw on five places at once. We make each of them say the same, complete thing.

  1. Shopify Catalog

    Agentic storefronts send your products to AI channels from the Shopify admin. That route is separate from robots.txt: blocking crawlers does not stop it, and allowing them does not fill it. We fill titles, attributes and metafields so the catalog version of a product is as complete as the page.

  2. Your storefront HTML

    Crawlers such as OAI-SearchBot and PerplexityBot read the page as the server sends it. Anything an app adds in the browser, from reviews to size guides, is missing for them. We move what matters into the theme.

  3. Google’s index

    AI Overviews and AI Mode are built on Google’s own crawl, so classic Shopify SEO still feeds them: indexable collections, clean canonicals and fast pages.

  4. Google Merchant Center

    Free listings and Google’s shopping results read your Merchant Center feed for brand, price and availability. We keep it complete and in step with the page.

  5. Reviews and forums off your site

    Engines lean on Reddit threads, review sites and roundups when they recommend a store. We track what they say about you and earn a place in them.

Search problems Shopify stores face

The issues we find on almost every Shopify store we audit, each with the fix.

  1. 01Collection grids that link the wrong URL

    Product cards point to /collections/name/products/name, so crawlers and links land on a duplicate. We change the theme to link the canonical product URL.

  2. 02Two Product schemas on one page

    The theme outputs one, a review app adds another, and neither is complete. We keep one graph with ratings, brand, GTIN, shipping and returns.

  3. 03Product pages missing their reviews for AI

    Reviews, FAQs and size guides loaded by apps in the browser never reach crawlers that skip JavaScript. We render them in Liquid or from metafields, so they are in the HTML.

  4. 04Collections with no words

    A grid of products cannot answer “which one is best for wide feet”. We add answer-first copy, comparison tables and FAQs to the collections that carry revenue.

  5. 05A blocked crawler nobody noticed

    An old robots.txt.liquid or a proxy in front of the store can turn away AI search crawlers, and Shopify itself advises against a proxy such as Cloudflare in front of a store. We check the live rules and server responses and let in the crawlers you want.

Find out which of these is costing you most.

Book a call and we will rank them by the revenue at stake on your own site.

Book a call

The apps you run, and where they stop

Most Shopify brands already run a few of these. Each does its job; none of them covers the whole answer.

All-in-one SEO apps

Their gap
They fix meta tags, alt text and broken links in bulk and add schema templates inside the store. Most cannot tell you whether ChatGPT recommends you, or who it names instead.
Your opening
We track that across ChatGPT, Gemini, Perplexity, Claude and Google’s AI answers and work on the pages and sources that change it.

Review apps

Their gap
They collect reviews well, but many load them in the browser and add their own Product markup. That leaves AI crawlers with no reviews and Google with two schemas.
Your opening
We put the reviews in the HTML and merge the ratings into one graph.

Feed apps

Their gap
They push products to Google and marketplaces, and a feed is only as good as the titles and attributes in it.
Your opening
We rewrite those around how people actually ask for the product, and keep feed and page in step.

Agentic storefronts

Their gap
Turning on an AI channel in the admin is one setting. Being the product the agent picks depends on data and reputation the setting does not create.
Your opening
We build both: complete attributes in Catalog, and the reviews, threads and press engines check before they recommend.

The Shopify program, part by part

Each part of the program is built for how Shopify works: the theme, its apps, its fixed URLs and the catalog it sends to AI channels.

Content

Buying guides and comparisons on the blog, linked to the collections they support. Shopify’s blog sits on the same domain as your products, so a guide that gets cited sends its authority, and its readers, straight to the collection.

More on Content
01 / 08

Why Shopify brands work with Passionfruit

Shopify solves the parts of SEO that used to take a developer: sitemaps, canonicals, fast hosting and a default robots.txt. What is left is the work that wins the recommendation, and it sits across the theme, the apps, the catalog and the rest of the web.

Passionfruit Intelligence shows where your store appears in ChatGPT, Gemini, Perplexity, Claude and Google’s AI answers, who is named instead, and which pages and sources are behind it. Our team then ships the fixes: theme changes, collection copy, schema, feeds and the off-site work engines trust.

We work inside your store with your team’s approval, and every change is measured against revenue, not just rankings.

What we measure

  • Revenue from organic search and AI referrals, by collection
  • How often AI answers name your store for your top products
  • Indexed product and collection pages versus live catalog
  • Catalog completeness for the attributes AI channels read

Questions

Is Shopify good for SEO?

Yes, for the technical basics. Shopify generates sitemaps, canonicals and a sensible robots.txt, and its standard themes output Product markup. What it leaves to you is collection copy, complete product data, reviews in the HTML and a catalog AI shopping agents can trust, which is what decides who gets recommended.

Can I remove /products/ and /collections/ from Shopify URLs?

No. Product, collection, page and blog URLs carry a fixed prefix, and only the handle after it can change. We write handles that match how people search and add a 301 for every handle we change, rather than fighting the prefix.

Does blocking AI crawlers in robots.txt keep my products out of ChatGPT?

Not on its own. Shopify can send product data to AI channels through Shopify Catalog, which is separate from robots.txt and managed in the admin. Crawlers such as OAI-SearchBot still read your pages directly, so we check both.

Why can’t AI crawlers see my reviews and FAQs?

The main AI crawlers studied do not run JavaScript, so reviews, FAQs and size guides that apps load in the browser never reach them. We render that content in Liquid or from metafields so it is in the HTML.

Do Shopify SEO apps cover AI search?

They fix meta tags, alt text and schema templates in bulk, but most cannot tell you whether ChatGPT recommends you or who it names instead. Passionfruit tracks that across ChatGPT, Gemini, Perplexity, Claude and Google’s AI answers and works on the pages and sources that change it.

Do you work with headless Shopify and Hydrogen?

Yes. We check that canonicals, schema and content are in the server response rather than added after the page loads in a browser.

More platforms

Get found and chosen in AI answers.

Book a call and we will show you where Google and AI answers leave you out today, and what we would ship first to get you in.