What this prompt does
This prompt generates complete Schema.org JSON-LD for a technical page. You specify the [page_type] and [topic], and the AI returns a primary schema with recommended properties plus the supporting types that fit your page: breadcrumb schema matching [url_structure], author schema with sameAs links to [author_profiles], organization schema, FAQ schema when you have [faq_count] questions, and conditionally HowTo, SoftwareApplication, Video, and aggregate-rating schema. It outputs everything as a single script tag ready to paste into the page head.
The structure works because correct, nested structured data is one of the highest-leverage, lowest-effort SEO wins available, and the failure mode is almost always malformed nesting or missing required properties. By enumerating each schema type as a conditional and asking the model to follow Google's guidelines and Rich Results Test recommendations, the prompt produces markup built to validate rather than just look plausible. Treating each supporting type as a conditional matters because adding schema for content a page does not contain is a guideline violation, not a bonus. The [url_structure] and [author_profiles] variables ground the breadcrumb and author entities in your real site, so the entities reference your actual paths and verifiable profiles rather than placeholders.
When to use it
- You are publishing a tutorial, FAQ, or tool page and want rich-result-eligible markup.
- You need breadcrumb schema that mirrors your actual
[url_structure]. - You want author schema with sameAs links to establish E-E-A-T signals.
- You have a page with
[faq_count]questions and want valid FAQPage markup. - Your content is step-by-step and would benefit from HowTo schema.
- You discuss a tool or product that warrants SoftwareApplication schema.
Example output
Expect a single <script type="application/ld+json"> block containing the primary schema for your [page_type], nested with breadcrumb, author, and organization entities, plus any conditional types your page qualifies for. The markup is structured to pass Google's Rich Results Test, with required properties populated and entities referenced correctly.
Pro tips
- Set
[url_structure]to your real canonical path so breadcrumb items resolve correctly rather than to placeholder URLs. - Populate
[author_profiles]with genuine sameAs links — GitHub, LinkedIn, and similar — because empty or fake profiles weaken the E-E-A-T signal you are trying to build. - Only claim conditional schema your page actually has; adding FAQ or aggregate-rating markup for content that isn't present is exactly the kind of mismatch Google penalizes.
- Always run the generated block through Google's Rich Results Test before deploying — the prompt builds to pass, but validation catches edge cases.
- Match
[faq_count]to the real number of questions on the page so the FAQPage entity is accurate. - For tutorial content, lean into the HowTo schema and make sure each step in the markup mirrors a visible step on the page.
- When iterating, ask the model to flag which properties are required versus recommended so you know what is safe to trim.
