How to add FAQ and schema markup to many pages

By Ranjit Bhinge, founder of Blur Studio · Updated · 10 min read

Short answer: To add FAQ blocks and schema markup (code that labels your page facts for search engines) to many pages, keep one data table with a row per page, write one page template that prints both the visible text and a block of JSON-LD from that same row, test five pages, then roll out in batches. Know one change first: Google ended the FAQ rich result on 7 May 2026, so FAQ blocks are now for readers, not for a special search display. Ploy, a software subscription where an AI agent builds and edits websites, can generate pages from a data table, but a template in your own site works too.

Ranjit Bhinge

You have dozens or hundreds of service, location or product pages and no wish to paste code into each one by hand.

I'm Ranjit, and I run Blur Studio: 15+ years building software, 120 apps taken live, and clients who have gone on to raise $10M+. I build AI systems and processes that help businesses do a lot more with a lot less time. I also work differently from most of what else is out there, more on that further down this page. Say hello.

The safe way to add FAQ blocks and schema to many pages is to generate both from the same data. One table row holds the facts for one page: the name, address, phone, hours, price and three questions with answers. One template prints the visible page and the JSON-LD from that row, so the code can never say something the page does not. Then you test a handful of pages with Google's tools, publish in batches, and watch Google Search Console for errors.

This page is for owners, marketers and agencies who run template-based sites. It covers what Google still supports, what to put in the table, a copy-ready JSON-LD example, the mistakes that cause trouble at scale, how to test, and where Ploy fits among the options.

This page may contain affiliate links from which Blur Studio earns commissions.

What schema is, and what changed for FAQ

Schema markup, also called structured data, is code that tells search engines what a page is about in a fixed format: this is a business, this is its phone number, this is a price. JSON-LD is the code format. It sits in a script block in the page and does not change how the page looks. Google's introduction to structured data says it recommends JSON-LD if your setup allows it, and calls it the easiest to implement and maintain at scale. Google can also read JSON-LD that JavaScript adds to the page.

The FAQ change matters. Google's documentation says the FAQ rich result, the expandable questions that used to appear under a listing, no longer appears in Google Search starting 7 May 2026. The FAQ page of its docs was removed on 15 June 2026. Many guides online still teach FAQ schema as a way to get rich results. That advice is out of date.

So should you still add FAQ blocks? Yes, for readers. A short FAQ answers the follow-up questions a buyer has and gives AI tools clear, answer-first text to read. But do not expect a special display in Google. FAQPage markup may still be valid under the schema.org vocabulary, but I would add it only if it costs nothing, and never in place of good visible text.

Also note what Google says about AI search. Its AI features page says there are no additional requirements to appear in AI Overviews or AI Mode, and that you need no special schema or new files. Schema is not a trick for AI answers. It is a way to state facts clearly, and Google says markup does not guarantee any feature will show.

Google's rules for structured data, in its own words

  • "Don't mark up content that is not visible to readers of the page."
  • Markup must be a true representation of the page content.
  • Do not mark up irrelevant or misleading content, such as fake reviews.
  • Ratings not given by actual users can lead to a manual action, which removes rich result eligibility. Ranking in web search is unaffected by a structured data manual action, per Google's policies page.

These rules decide your design. At scale, the main risk is not a typo but a template that says things the page does not, or things that are no longer true.

Which schema types fit which pages

Page typeSchema typeWhat Google's docs require
Your business, location or service-area pageLocalBusiness (or a subtype such as Dentist or Store)Required: name and address. Recommended: geo with at least 5 decimal places, openingHoursSpecification, priceRange (under 100 characters), telephone with country and area code, url
A single product pageProductRequired: name, plus at least one of review, aggregateRating or offers. An Offer needs a price. Recommended: availability, priceCurrency, priceValidUntil
Category or listing pagesDo not use ProductGoogle says product markup is for a single product (or its variants), not for category pages
Articles and guidesArticleCheck Google's Article docs for current fields
Site navigation trailBreadcrumbListCheck Google's docs for current fields

The schema.org site describes LocalBusiness as a particular physical business or branch of an organization, with about 35 subtypes. Google says it does not guarantee that features using structured data will show up in results. Google's Product docs also say a price-valid-until date in the past may stop a display.

Step 1: build the data table

One row per page. Name the columns so a template can use them. Example columns for a service business with location pages:

ColumnExampleUsed for
slugemergency-plumber-tulsaPage address
business_nameExample PlumbingPage text and JSON-LD
street, city, region, postcode, country1 Main St, Tulsa, OK, 74101, USAddress in both
phone_intl+1-918-555-0100Telephone with country code
hoursMon-Fri 08:00 to 17:00Visible hours and openingHoursSpecification
price_textCall-out fee from $90 (made-up)Visible price line
unique_factsThree real facts about this location or serviceThe part that makes the page different
faq1_q, faq1_a, faq2_q, faq2_a, faq3_q, faq3_aReal questions with real answersVisible FAQ block
reviewed_by, reviewed_onName, dateYour own control: who checked this row and when

The columns unique_facts and the FAQ answers are where thin pages are avoided. If a column is the same on every row, it adds nothing, so write FAQ answers that fit the location or product. Google's spam policies target many pages "generated for the primary purpose of manipulating search rankings", no matter how they are created. Our guide to making hundreds of pages from a spreadsheet goes deeper on which rows deserve a page.

Step 2: write one template that prints both

The template reads a row and prints the visible block and the JSON-LD from the same fields. Here is the JSON-LD half for a business page, using placeholders in double braces. The values come from the table above, and the example is a pattern to adapt, not a promise of any result.

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "Plumber",
  "name": "{{business_name}}",
  "url": "https://www.example.com/{{slug}}",
  "telephone": "{{phone_intl}}",
  "priceRange": "{{price_range}}",
  "address": {
    "@type": "PostalAddress",
    "streetAddress": "{{street}}",
    "addressLocality": "{{city}}",
    "addressRegion": "{{region}}",
    "postalCode": "{{postcode}}",
    "addressCountry": "{{country}}"
  }
}
</script>

Rules for the template:

  1. Build the JSON with code, not by pasting text. A stray quote mark in a business name or answer breaks the whole block. Let your template language escape the values.
  2. Print only fields that also appear on the page. If the hours are in the code, show them in the text.
  3. Leave a field out if the row is empty. Do not print empty strings or "N/A".
  4. Use one address per page. A location page marks up that location, not the head office.
  5. Do not copy ratings sitewide. Only mark up reviews that are real, from real users, and shown on that page.
  6. Check dates. An expired price-valid-until date is a known cause of lost displays.

Where FAQs go: print the questions and answers as normal visible text, and if you also add FAQPage code, use the same words. A hidden answer in the code that is not on the page breaks Google's visibility rule.

Mistakes that hurt at scale

  • Text in code but not on the page, such as hidden FAQ answers or offers nobody can see.
  • Stale values: prices, availability and hours that were right once and are wrong now. A template makes one stale value appear on a hundred pages.
  • One sitewide rating copied to every page.
  • Reviews not from real users. Google says this can lead to a manual action.
  • Phone numbers without a country code.
  • Product markup on category pages.
  • Broken JSON from unescaped quotes in templated text.
  • Near-identical pages that differ only by a place name. That is a content problem no markup can fix.

Step 3: test plan with pass and fail marks

Test in this order before you roll out.

  1. Pilot five pages that differ: one short row, one long row, one with an apostrophe or ampersand in the name, one with an empty optional field, one with a long FAQ.
  2. Run the Rich Results Test (Google's tool, which shows which rich results a page can generate, from a URL or pasted code). It tests rich-result eligibility, not every part of schema.org.
  3. Run the Schema Markup Validator (schema.org's general validator) for syntax and vocabulary problems.
  4. Read the page against the code. Every value in the code must be visible and correct on the page.
  5. Publish a batch of 20 pages, then watch the enhancement and rich result reports in Search Console and URL Inspection for errors.
  6. Repeat with the next batch if the first is clean.
CheckPassFail
ValidityNo errors in either tool on all five pilot pagesAny error, or a warning you cannot explain
MatchEvery code value is visible on the pageAny value only in code
AccuracyA person checked phone, hours, price and addressAny unchecked value
UniquenessEach page has facts that differ from its neighboursOnly a name swapped
Search Console after 2 to 4 weeksNo new errors in reportsErrors or manual action message

Do not judge the effect on rankings in the first weeks. Google says changes can take a few hours to several months to show, and it does not say valid schema ranks pages higher. Judge on correctness, then on clicks and enquiries over time.

Your options, from least to most effort

OptionCostBest forWatch for
Hand-edit each pageYour timeUnder 10 pagesPages drift apart; typos
A WordPress plugin or theme featureCheck the plugin's pageSites already on WordPressDefault settings may mark up things you do not show; check each output
A template in your own site code or CMSA developer's timeLarge catalogs and many location pagesNeeds someone who can test it
Semrush or a similar site audit toolCheck the vendor pageFinding pages with missing or broken markupAn audit finds problems; it does not write the template
A freelancer or agencyQuote only, or hourlyOwners with no technical helpAsk for a sample on five pages first
An AI agent such as PloySee belowGenerating pages and markup from a table, with approvalCredits, human review, no guaranteed outcome

Where Ploy fits, as one option of several

Ploy's built-in database, called PloyDB, is a spreadsheet-style table where one template turns every row into a page. Add a row and a page is published, and edits reach visitors usually within a minute. Ploy lists limits of 97 columns per table, 128 KB per text value and a 32 MB CSV import. That maps well onto the table design above. Ploy's comparison pages say its agent handles meta tags, schema, the sitemap and internal linking, and its AEO docs describe an agent readiness audit that scores a site from 0 to 5 and fixes gaps it can. Those are Ploy's own descriptions, so test them on five pages and read the output yourself.

What its docs do not describe: a check for duplicate or thin rows, or a way to leave a single row out of Google's index. Noindex is documented only as a manual setting for staging pages. Ploy's own comparison with Webflow lists heavy programmatic SEO at scale under "pick Webflow if". So your row-level quality checks have to come from you.

Plan facts from Ploy's pricing page in November 2026: Free is $0 (500 credits a day, up to 2,000 in total), Starter is $50 a month (4,000 credits), Pro is $300 a month (24,000 credits), Enterprise is custom. Credits do not roll over. Ploy does not publish what a page or a markup run costs in credits. AI-written FAQ answers and pages need a person to review them before publishing. See our PloyDB guide and Ploy pricing breakdown.

Who should skip this

  • Sites with fewer than about ten pages. Edit by hand and check each one.
  • Anyone chasing FAQ rich results. They ended on 7 May 2026.
  • Anyone whose rows hold no real facts. Fix the data first. Markup will not add substance. Our local SEO pages guide shows what real local facts look like.
  • Skip Ploy if your site lives on a platform you do not plan to leave, or if you need a developer-controlled template with unit tests. A template in your own code is better then.

For how AEO and schema fit together, see what AEO is. For docs, pricing and comparison pages, see pages that rank for software companies.

How I'd approach this

I would start small and let the table do the work. Schema is easy to get slightly wrong on a hundred pages at once.

First-week plan:

  1. Day 1: List your page types and pick one. Write its columns. Fill 10 real rows.
  2. Day 2: Write the template that prints visible text and JSON-LD from the same row. Escape every value.
  3. Day 3: Build five pilot pages, including the awkward ones. Run both validators and read each page against its code.
  4. Day 4: Fix what failed. Have a person check every phone number, price and hour.
  5. Day 5: Publish a batch of 20 and set a reminder to check Search Console in two weeks.

Decision rule: if you already have a CMS with templates, add the JSON-LD there. If you have no tooling and want an agent to generate the pages from a table, try Ploy on the Free plan with 10 rows and see what the output and the credit use look like before you pay. If you have hundreds of rows and need custom checks, a developer-built template is safer. Blur Studio builds custom content systems and looks after the pages AI tools read: see what Blur Studio builds.

A closely related page covers the same ground from another angle: SEO for Business Owners With No Time: Options and Costs.

Related guide: Stuck on Page Two? How to Push Key Pages Up.

Questions people ask

Does FAQ schema still get rich results in Google?

No. Google's documentation says the FAQ rich result stopped appearing in Google Search starting 7 May 2026. You can still add FAQ blocks for readers, but do not expect a special display.

Is JSON-LD or microdata better?

Google recommends JSON-LD if your setup allows it and calls it the easiest to implement and maintain at scale. It also accepts microdata and RDFa. Google can read JSON-LD that JavaScript adds to the page.

Do I need schema to appear in AI Overviews?

No. Google says there are no additional requirements to appear in AI Overviews or AI Mode, and no special schema is needed. Schema helps state facts clearly, but it does not guarantee a feature. See what AEO is.

What tool tests my schema?

Use Google's Rich Results Test for rich-result eligibility and schema.org's Schema Markup Validator for general syntax. After publishing, watch Search Console reports and the URL Inspection tool.

Can I mark up an answer that is not shown on the page?

No. Google says not to mark up content that is not visible to readers, and the markup must be a true representation of the page. Print the same text in both places.

Can AI write the FAQ answers for all my pages?

It can draft them, but a person should check each answer, because AI output can be wrong. Answers that are the same on every page add little. Make them specific to the page. See making many pages from a spreadsheet.

Does Ploy add schema automatically?

Ploy's comparison pages say its agent handles schema, meta tags, the sitemap and internal linking, and its docs describe an agent readiness audit. Treat those as Ploy's own claims, test five pages, and validate the output with Google's tool.

How many pages should I roll out at once?

Pilot five varied pages, then publish in batches of about 20. Check Search Console between batches. A single template error is copied to every page that uses it.

More in this topic: Websites: SEO routines and search rankings (4 guides) →

Sources

Ready for the next step?

A 30 minute call. If you do not need me, I will tell you.

Ranjit Bhinge You talk to me, Ranjit. I reply within a few hours.

Not sure what you need? Tell me what is going on. I will point you the right way, even if it is not me.

Blur Studio · 30 minute call

Book a 30 minute call

You talk to me, Ranjit. If you do not need me, I will tell you.

  • We talk about your business and what is going on.
  • You leave with a clear next step, even if it is not me.
  • I reply to every message within a few hours.
Loading the calendar…