Explain Your Structured Data in Plain English
Paste any page's HTML and get back a plain-English translation of every JSON-LD block on the page — what each one tells Google, ChatGPT, and Perplexity about your business, what's missing, and which fixes would have the biggest impact. Built for non-developers who inherited a website and have no idea what schema is already there.
Pro tip
Most small business owners have schema on their site they've never seen — added by a theme, a plugin, or a previous developer. Half of it is wrong, half is incomplete, and a small fraction is actively hurting Rich Results. Run this once before adding any new schema; you'll often find you don't need to add anything, just fix what's already there.
How to use this prompt
- Pick your AI model. Choose the tab for Claude, ChatGPT, Gemini or Copilot — each variant is tuned for that model.
- Copy the full prompt. Click Copy Full Prompt to copy the text to your clipboard.
- Paste into your AI tool. Open your chosen model and paste the prompt into a new chat.
- Replace the
[placeholders]. Swap any bracketed fields for your company name, audience, product or tone. - Run and refine. Review the output. If anything is off, ask the AI to tighten tone, length or format.
Prompt Variants by Model
Claude
Claude 4.x
FRESH APR 2026
You are a structured-data translator. I''m going to paste a page''s HTML. Find every JSON-LD block on the page and translate each one into plain English a small business owner can understand — what...You are a structured-data translator. I''m going to paste a page''s HTML. Find every JSON-LD block on the page and translate each one into plain English a small business owner can understand — what it tells Google, ChatGPT, and Perplexity about my business, what''s missing, and what to fix first. I''m not a developer.
<my_context>
My business: [NAME] — [WHAT YOU DO IN ONE SENTENCE]
Page URL: [https://YOUR-DOMAIN.COM/...]
What I want this page to be cited for in AI answers: [VERBATIM QUERY or topic]
</my_context>
<page_html>
[PASTE THE FULL HTML SOURCE — view source on the page, select all, copy. Strip nothing.]
</page_html>
For every <script type="application/ld+json"> block you find on the page, produce this exact structure:
**Block [N] — [@type detected]**
**What it says (plain English):**
2–4 sentences. No jargon. Pretend you''re explaining to a smart friend who runs a business but has never heard of schema. Translate the actual content of the block — names, addresses, dates, descriptions — not just the schema type.
**What this tells AI engines:**
2–3 bullets. What signal does this give Google, ChatGPT, Perplexity? E.g. "Tells search engines this page is a tutorial that takes 30 minutes" or "Tells Google your business is located in Austin and serves dental patients."
**What''s missing or wrong:**
Bulleted list. Be specific:
- Required fields that are empty or missing (name them).
- Fields filled with placeholder text (e.g. "Lorem ipsum", "Your Company Name").
- Wrong @type for the content (e.g. "marked as Article but it''s a how-to — should be HowTo").
- Outdated information (dates from years ago, replaced product names).
- Schema fighting itself (e.g. "duplicate FAQ data also expressed as visible microdata — Rich Results error").
**Fix priority:**
Pick one: 🔴 Fix today / 🟡 Fix this month / 🟢 Optional improvement.
One sentence on why.
After all blocks, end with three meta sections:
**OVERALL — What''s on This Page**
One paragraph: how many JSON-LD blocks were found, what types, and whether the overall picture an AI engine would form of this page matches what I told you I want it to be cited for.
**OVERALL — The 3 Highest-Leverage Fixes**
Numbered list. Each: what to fix, why it matters most, written for a non-developer. If a fix involves new schema, point me to the AlignAI prompt that generates it (e.g. "Use the ''Write Your Business''s Organization Schema'' prompt").
**OVERALL — Things That Are Actually Fine**
A short note on what''s already working — non-developers tend to over-edit. Tell me what to leave alone.
If you find ZERO JSON-LD blocks on the page, say so plainly and recommend the 3 schema types this page should have based on its content + the query I want it cited for.
Notes: Claude is the strongest variant for the plain-English translation step. Particularly good at spotting placeholder text and outdated values.
ChatGPT
GPT-5
FRESH APR 2026
Act as a structured-data translator. I''ll paste a page''s HTML. Find every JSON-LD block and translate each into plain English a small business owner understands — what it tells Google, ChatGPT, and...Act as a structured-data translator. I''ll paste a page''s HTML. Find every JSON-LD block and translate each into plain English a small business owner understands — what it tells Google, ChatGPT, and Perplexity about my business, what''s missing, and what to fix first. I''m not a developer.
My context:
- Business + what we do (one sentence): [NAME — DESCRIPTION]
- Page URL: [https://YOUR-DOMAIN.COM/...]
- What I want this page to be cited for in AI answers: [QUERY or TOPIC]
Page HTML (paste full source — view source, select all, copy):
[PASTE]
For every <script type="application/ld+json"> block, produce:
**Block [N] — [@type detected]**
What it says (plain English): 2–4 sentences, no jargon, translate the actual content (names, addresses, dates, descriptions) — not just the schema type.
What this tells AI engines: 2–3 bullets. What signal does this give Google, ChatGPT, Perplexity?
What''s missing or wrong: bulleted list — required-but-empty fields, placeholder text, wrong @type for the content, outdated info, schema fighting itself (e.g. duplicate FAQ data in microdata + JSON-LD — Rich Results error).
Fix priority: 🔴 Fix today / 🟡 Fix this month / 🟢 Optional improvement. One sentence on why.
After all blocks:
OVERALL — What''s on This Page: one paragraph — block count, types, whether the picture an AI engine would form matches what I want this page cited for.
OVERALL — The 3 Highest-Leverage Fixes: numbered. Each: what to fix, why it matters most, non-developer language. If a fix needs new schema, point to the matching AlignAI prompt.
OVERALL — Things That Are Actually Fine: short note on what to leave alone.
If ZERO JSON-LD blocks found: say so and recommend 3 schema types this page should have based on its content + the citation goal.
Notes: GPT-5 with browsing can fetch the URL directly — replace the paste step if browsing is enabled.
Gemini
Gemini 2.5 Pro
FRESH APR 2026
You are a structured-data translator. I''ll paste a page''s HTML. Find every JSON-LD block and translate each into plain English a small business owner understands — what it tells Google, Gemini,...You are a structured-data translator. I''ll paste a page''s HTML. Find every JSON-LD block and translate each into plain English a small business owner understands — what it tells Google, Gemini, ChatGPT, and Perplexity about my business, what''s missing, and what to fix first. I''m not a developer.
MY CONTEXT:
• Business + what we do: [NAME — SENTENCE]
• Page URL: [https://YOUR-DOMAIN.COM/...]
• What I want this page cited for in AI answers: [QUERY or TOPIC]
PAGE HTML (paste full source):
[PASTE]
For every <script type="application/ld+json"> block, produce:
**Block [N] — [@type detected]**
What it says (plain English): 2–4 sentences, no jargon, translate the actual content (names, addresses, dates, descriptions).
What this tells AI engines: 2–3 bullets — signals to Google / Gemini / ChatGPT / Perplexity.
What''s missing or wrong: bulleted — empty required fields, placeholder text, wrong @type for content, outdated info, schema fighting itself (duplicate microdata + JSON-LD = Rich Results error).
Fix priority: 🔴 today / 🟡 this month / 🟢 optional. One-sentence why.
After all blocks:
OVERALL — What''s on This Page: one paragraph — block count, types, whether the AI-engine picture matches my citation goal.
OVERALL — The 3 Highest-Leverage Fixes: numbered. Each: what to fix, why, non-developer language. Point to matching AlignAI prompts.
OVERALL — Things That Are Actually Fine: short note — non-developers over-edit; tell me what to leave alone.
If ZERO JSON-LD blocks: say so and recommend 3 schema types this page should have.
Notes: Gemini 2.5 Pro is strong for the OVERALL summary — it tends to spot mismatch between schema and citation goal.
Copilot
Copilot (GPT-5)
FRESH APR 2026
Translate the structured data on a page into plain English. I''ll paste a page''s HTML. Find every JSON-LD block and explain — for a non-developer — what each tells Google, ChatGPT, Perplexity, and...Translate the structured data on a page into plain English. I''ll paste a page''s HTML. Find every JSON-LD block and explain — for a non-developer — what each tells Google, ChatGPT, Perplexity, and Bing Copilot about my business, what''s missing, and what to fix first.
My context:
• Business + what we do: [NAME — SENTENCE]
• Page URL: [https://YOUR-DOMAIN.COM/...]
• What I want this page cited for in AI answers: [QUERY or TOPIC]
Page HTML (paste full source):
[PASTE]
For every <script type="application/ld+json"> block:
**Block [N] — [@type detected]**
What it says (plain English): 2–4 sentences, no jargon, translate the actual content.
What this tells AI engines: 2–3 bullets — signals to Google / ChatGPT / Perplexity / Bing Copilot.
What''s missing or wrong: bulleted — empty required fields, placeholders, wrong @type, outdated info, schema fighting itself (duplicate microdata + JSON-LD = Rich Results error).
Fix priority: 🔴 today / 🟡 this month / 🟢 optional + one-sentence why.
After all blocks:
OVERALL — What''s on This Page: one paragraph — block count, types, AI-engine picture vs citation goal.
OVERALL — The 3 Highest-Leverage Fixes: numbered, non-developer language, point to matching AlignAI prompts.
OVERALL — Things That Are Actually Fine: short note on what to leave alone.
If ZERO JSON-LD blocks: say so and recommend 3 schema types this page should have.
Notes: Copilot is reliable for this prompt. Particularly clear in the priority-coded fix list.
Frequently Asked Questions
What does the Explain Your Structured Data in Plain English prompt do?
Paste any page's HTML and get back a plain-English translation of every JSON-LD block on the page — what each one tells Google, ChatGPT, and Perplexity about your business, what's missing, and which fixes would have the biggest impact. Built for non-developers who inherited a website and have no idea what schema is already there.
Which AI models is this prompt tested on?
This prompt is field-tested on Claude, ChatGPT, Gemini and Copilot. Each model has its own optimized variant above.
Do I need a paid AI account to use this prompt?
No. This prompt is written to run on the free tier of Claude, ChatGPT, Gemini and Copilot. Paid tiers simply give you longer context windows and faster responses.
Can I customize this prompt for my business?
Yes. Any text inside square brackets is a placeholder you replace with your own business details, such as company name, audience, product or tone. You can also ask the AI to adjust format, length or style after the first output.
When was this prompt last verified?
Each model variant above shows its own freshness stamp. AlignAI re-verifies every prompt at least monthly and rebuilds when a major model changes.
Don’t see what you need? tailored to your use case.