FAQ schema on Shopify: useful answers first, no rich-result promise

Do not spend your next product-page edit chasing FAQ rich results

Google's documentation changelog states that FAQ rich results stopped appearing in Google Search on May 7, 2026. The FAQ feature documentation was subsequently removed. See the official deprecation entry.

That changes the practical recommendation for a Shopify merchant. Keep useful customer answers. Do not install a schema app or edit theme code solely to obtain an expanded Google FAQ result.

---

Separate three different things

Item	What it does	What it does not establish
Visible product FAQ	Gives a shopper an answer on the page	That Google or an AI engine will cite it
FAQPage structured data	Describes question-and-answer content in a machine-readable vocabulary	Eligibility for the retired Google FAQ rich result
Product structured data	Describes product facts for supported product-search uses	A guaranteed ranking, rich result, or AI recommendation

Google's AI optimization guide says structured data is not required for its generative AI features and there is no special schema markup to add for them. That is Google's guidance, not a claim about every third-party system.

---

Three official checks replace the old copy-and-paste advice

1. Check feature support: Google's changelog confirms the FAQ display retirement. A historical tutorial is not current eligibility guidance
2. Check what the platform generates: Shopify documents a structured_data Liquid filter for product and article data. Inspect your own theme and apps before adding another implementation
3. Check the relevant type: Use Google's Product documentation for product-search requirements, rather than assuming an FAQ test covers product data

Do not paste a script into a generic “Edit HTML” field and assume Shopify will publish it correctly. The appropriate implementation depends on the theme, app, and template. If code changes are needed, use a duplicate theme and a qualified implementer.

---

Write an answer a customer can actually use

A care question is a better starting point than a promise that schema will increase citations. On Brooklinen's care guide, Classic Percale has specific wash and dry instructions, while Mulberry Silk has different instructions, including a delicate cycle and laying flat to dry.

The lesson is not to copy another brand's washing advice. It is to connect each answer to the correct product and an approved source.

Review field	Completed example from the inspected source
Product	Brooklinen Classic Percale
Customer question	How should I wash and dry it?
Source answer	Machine wash cold with like colors; tumble dry low, remove promptly
Boundary	Applies to that fabric's listed care instructions, not the brand's silk products
Source	Brooklinen care guide, checked September 14, 2026

Use those five fields to review one question on your own PDP. If the product-specific source is missing, get the fact approved before publishing an answer. Do not ask AI to invent a plausible care instruction.

---

Three mistakes in our own published tutorial

We checked the earlier Bulti page on September 14, 2026. Its advice failed in three different ways, not just because one date was old.

Earlier instruction on this page	What was wrong	Replacement decision
“use the page settings' Edit HTML or script insert area”	It implied a uniform Shopify insertion control without establishing the actual theme/app implementation	Identify the existing generator and edit in a duplicate theme only if there is a supported use case
“A green check means success”	Syntax or tool output cannot establish correct product facts or eligibility for a retired feature	Check the visible fact, its structured equivalent, and current feature support separately
“FAQ is the easiest place to start and has the biggest overlap with GEO”	We did not establish that priority or an AI citation benefit	Fix missing customer answers and supported product data before adding FAQ markup

These are our own editorial failures, not defects we are attributing to a customer's store. The original workflow could have sent a merchant into unnecessary code work while leaving the product answer unchanged.

---

If FAQ markup is already on your site

Do not rush to delete it simply because Google retired the display feature. First identify who maintains it and whether another integration actually uses it.

• Read the visible answer and compare it with the structured answer
• Fix conflicting or obsolete product facts in their source of truth
• Check whether the theme and an app generate inconsistent copies
• Use a schema validator for vocabulary and syntax, while checking Google's current documentation for supported rich-result types
• Verify the rendered page after any change; a validator cannot prove that the answer is correct or that it will be cited

A read-only inspection you can hand to your developer

Open the live product page, open browser developer tools, and run this snippet in the Console. It reads JSON-LD already present in the document; it does not edit the page, submit data, or make network requests.

const found = [];
function inspect(value, scriptNumber) {
  if (Array.isArray(value)) {
    value.forEach(item => inspect(item, scriptNumber));
  } else if (value && typeof value === 'object') {
    const types = [].concat(value['@type'] || []);
    if (types.some(type => ['Product', 'ProductGroup', 'FAQPage'].includes(type))) {
      found.push({ scriptNumber, types: types.join(', '),
        id: value['@id'] || '', name: value.name || '', data: value });
    }
    Object.values(value).forEach(item => inspect(item, scriptNumber));
  }
}
document.querySelectorAll('script[type="application/ld+json"]').forEach((script, index) => {
  try { inspect(JSON.parse(script.textContent), index + 1); }
  catch { console.warn('Invalid JSON-LD in script', index + 1); }
});
console.table(found.map(({ data, ...summary }) => summary));
console.log(found);

Compare the objects with the visible product and variant. Multiple objects are not automatically an error: a ProductGroup and its variants can be legitimate. Look for conflicting facts about the same item. This snippet does not inspect Microdata, and an empty result is not proof of absent structured data or a crawl problem.

Then ask the implementer to identify each script's source in the theme or installed apps. The DOM shows the output; it does not reliably identify which app generated it. Do not uninstall apps based on the count alone.

Use this handoff row: page URL / selected variant / script number and type / conflicting field / visible value / markup value / generating theme or app / proposed correction.

The priority is avoiding stale or contradictory information, not adding markup for its own sake.

---

Correction to the earlier version

This article previously promoted a simple FAQ JSON-LD insertion workflow, cited an unattributed monthly keyword volume, and treated a Rich Results Test check as the endpoint. That was not a sound recommendation for a current US Shopify store.

The revised task is smaller: publish one correct, useful answer; identify existing markup; only change code when there is a supported use case. For deciding where that answer belongs, read product page vs blog.

FAQ

Q. Is FAQPage markup itself forbidden? No. Retirement of Google's FAQ rich result is not a ban on the vocabulary. It does remove that Google display feature as a reason to implement it.

Q. Does FAQ schema make an AI cite my store? There is no such guarantee. Google says no special schema is required for its generative AI features. Helpful visible answers remain worth maintaining for shoppers.

Q. What should a Shopify merchant fix first? Correct the product facts and visible answers. Review existing product markup against current official requirements before adding more code or apps.