ChatGPT Prompts for Schema Markup: 10 JSON-LD Prompts

ChatGPT prompts for schema markup help you write schema.org JSON-LD for a page, but only from information a visitor can actually see on that page. Paste the visible content, name the page type, and ask ChatGPT to map each property to where it appears, mark what is missing, and never guess addresses, prices, or ratings. The 10 prompts below cover choosing a type, writing the code, fixing errors, and testing.

Google recommends JSON-LD and says not to mark up content that is not visible to users. Google no longer shows FAQ or HowTo rich results, so do not add that markup just to win them, and always test with Google's own tools.

Plan the markup

1. Pick the schema type

Here is my page content [paste]. Which schema.org type fits this page best, and which more specific subtype should I use? List types I should not use here and why.

2. Property map from visible content

For a [type] page, map each property Google supports to the exact text on my page [paste]. Mark each one required or recommended, and write MISSING ON PAGE where the value is not visible.

3. Site-wide markup plan

My site has these page types: [list]. Suggest which markup belongs on which page type (for example Organization on the home page, Product on product pages) and what should not repeat on every page.

Write JSON-LD

4. Local business block

Write LocalBusiness JSON-LD, using the most specific subtype, for [business] from this contact page text [paste]: name, address, phone, hours, and URL. Leave out anything not on the page.

5. Product block

Write Product JSON-LD for this product page [paste]: name, image URLs, description, brand, offer price, currency, and availability exactly as shown. No ratings unless reviews are shown on the page.

6. Article block

Write Article JSON-LD for this blog post [paste]: headline, author name and URL, datePublished and dateModified with time zone, and images. Use the dates printed on the page.

7. Breadcrumb block

Here is my breadcrumb trail as shown on the page [paste with URLs]. Write BreadcrumbList JSON-LD with the same names, order, and URLs.

Fix and test

8. Fix validation errors

The Rich Results Test shows these errors and warnings [paste] for this code [paste]. Explain each one in plain words and give the corrected code. Do not add values that are not on my page.

9. Find conflicts with my plugin

My page already outputs this markup from a plugin [paste], and I want to add this block [paste]. Find duplicates, mismatched values, or conflicting types, and tell me which to keep.

10. Testing checklist

Write a testing checklist for new markup on [page]: Rich Results Test, Schema Markup Validator, URL Inspection after publishing, and which Search Console report to watch, with what to do if errors appear.

Write a full markup plan in one run

The flagship Schema Markup Prompt: JSON-LD From What Is on Your Page picks the type, maps every property to visible text, writes one JSON-LD block, flags conflicts with existing markup, lists page fixes, gives CMS steps, and ends with a test plan. Its example writes Bakery markup for a neighborhood bakery's contact page.

Rules that keep markup safe

Mark up only what visitors can see, use the required properties from Google's documentation for each feature, and never add ratings, reviews, or prices that are not on the page. Valid markup makes a page eligible for rich results; it does not guarantee them.

Related prompts

For local businesses, see ChatGPT prompts for local SEO. For a site-wide check, use the technical website SEO audit and action plan. To improve the page content itself, try the GEO/AEO content pass for a product page and Shopify collection SEO copy from attributes.

Find more in the SEO category, look through Browse, or post your own on Share.