Find Duplicate Schema Errors Killing Rich Results
Scan a page for the #1 cause of Rich Results validation failures: the same data expressed in BOTH visible microdata (itemscope/itemtype/itemprop) AND JSON-LD. Returns a clear list of which microdata attributes to remove from the visible HTML so JSON-LD can stand alone — plus a check for orphaned aggregateRating, broken @id references, and 5 other common Rich Results killers.
Pro tip
When duplicate FAQ data appears in both microdata and JSON-LD on the same page, Google's Rich Results Test rejects the entire FAQPage. We've hit this twice on AlignAI's own properties. The pattern: visible <details> blocks marked up with itemtype="https://schema.org/FAQPage" while the same Q&A pairs also live in JSON-LD. Pick JSON-LD, strip the microdata.
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 Rich Results debugger. I''ll paste a page''s HTML. Scan it for the #1 cause of Rich Results validation failures — the same data expressed in BOTH visible microdata (itemscope, itemtype,...You are a Rich Results debugger. I''ll paste a page''s HTML. Scan it for the #1 cause of Rich Results validation failures — the same data expressed in BOTH visible microdata (itemscope, itemtype, itemprop) AND JSON-LD on the same page — plus 7 other common Rich Results killers. Tell me exactly what to remove or fix.
<page_context>
Page URL: [https://YOUR-DOMAIN.COM/...]
Have you run Google''s Rich Results Test against this URL? [yes / no — if yes, paste the error messages]
</page_context>
<page_html>
[PASTE FULL HTML SOURCE — view source, select all, copy. Don''t strip anything.]
</page_html>
<duplication_check_workflow>
Walk through the page systematically:
1. List every JSON-LD block (script type="application/ld+json") and the schema type each one declares.
2. Scan the visible HTML for any element carrying itemscope, itemtype, or itemprop attributes.
3. For each microdata element, identify which JSON-LD block (if any) expresses the same entity.
4. Where a duplication exists → flag it. JSON-LD must win; microdata must be stripped from the visible HTML.
</duplication_check_workflow>
<other_killers_to_check>
Also flag every instance of these 7 common Rich Results killers:
A. **Orphaned AggregateRating** — AggregateRating present without backing Review entities on the same page or referenced via @id.
B. **Broken @id references** — an @id referenced from another schema block (e.g. itemReviewed @id) that doesn''t exist in any other block.
C. **Wrong @type for the content** — e.g. <script> declares @type Article but the visible content is a step-by-step tutorial (should be HowTo); declares Product but no offers/price; declares LocalBusiness but no PostalAddress.
D. **Required fields missing** — for each detected @type, list any Schema.org REQUIRED fields that are empty/missing.
E. **Placeholder values** — any field still containing "Lorem ipsum", "Your Company Name", "[BRAND]", "example.com", "TODO", etc.
F. **Invalid JSON syntax** — trailing commas, unescaped quotes, unbalanced braces. If any block fails to parse, all schema on the page is ignored by Google.
G. **Conflicting nested schemas** — e.g. an Organization with a different name than the WebSite''s publisher reference.
</other_killers_to_check>
Output exactly four parts:
**PART 1 — JSON-LD Inventory**
Markdown table: Block # · @type · Validates as JSON? (y/n) · Required fields complete? (y/n + which missing) · Placeholder values? (y/n + which) · @id used (if any).
**PART 2 — Microdata vs JSON-LD Duplication Map**
Markdown table: Microdata location (line or section) · @type expressed · Duplicates which JSON-LD block? (block # or "no duplicate") · Action (REMOVE microdata / KEEP — no duplication).
For every REMOVE row, paste the exact attributes to strip (e.g. itemscope itemtype="https://schema.org/FAQPage", itemprop="acceptedAnswer", etc.) so I can find-and-replace them out of the HTML.
**PART 3 — Other Rich Results Killers Found**
For each killer A through G that appears, one section showing:
- Killer letter + name
- Where it appears (block # or HTML location)
- What''s wrong in plain English
- The exact fix (paste-ready if possible)
**PART 4 — Verdict + Re-Test Plan**
- Will this page pass Google''s Rich Results Test after fixes? Yes / Maybe / Unlikely + one-sentence why.
- Rich Results Test URL to re-verify: https://search.google.com/test/rich-results?url=[PAGE-URL]
- Schema.org JSON validator URL: https://validator.schema.org
- Order of fixes — start with the highest-leverage one and explain why (usually: invalid JSON > missing required fields > duplication > other).
If you find no issues, say so plainly. Don''t fabricate problems.
Notes: Claude is the strongest variant for the duplication map — it's the most rigorous about line-by-line microdata-vs-JSON-LD matching. We've found this prompt catches issues Google's own Rich Results Test sometimes misses on first pass.
ChatGPT
GPT-5
FRESH APR 2026
Act as a Rich Results debugger. I''ll paste a page''s HTML. Find the #1 cause of Rich Results validation failures — the same data in BOTH visible microdata (itemscope/itemtype/itemprop) AND JSON-LD...Act as a Rich Results debugger. I''ll paste a page''s HTML. Find the #1 cause of Rich Results validation failures — the same data in BOTH visible microdata (itemscope/itemtype/itemprop) AND JSON-LD on the same page — plus 7 other common Rich Results killers.
Page context:
- URL: [https://YOUR-DOMAIN.COM/...]
- Have you run Google''s Rich Results Test? [yes / no — if yes, paste error messages]
Page HTML (paste full source — view source, select all, copy):
[PASTE]
Workflow:
1. List every JSON-LD block (script type="application/ld+json") and the @type each declares.
2. Scan visible HTML for itemscope, itemtype, itemprop attributes.
3. For each microdata element, identify which JSON-LD block (if any) expresses the same entity.
4. Where duplication exists → flag. JSON-LD wins; microdata gets stripped.
Also check these 7 killers:
A. Orphaned AggregateRating (no backing Review entities or @id refs).
B. Broken @id references (@id used but not defined elsewhere).
C. Wrong @type for the content.
D. Required Schema.org fields empty/missing per @type.
E. Placeholder values ("Lorem ipsum", "Your Company Name", "[BRAND]", "example.com", "TODO").
F. Invalid JSON syntax (trailing commas, unescaped quotes, unbalanced braces).
G. Conflicting nested schemas (e.g. Organization name ≠ WebSite publisher name).
Output:
1. JSON-LD Inventory table: Block # · @type · Valid JSON? · Required fields complete? · Placeholders? · @id used.
2. Microdata vs JSON-LD Duplication Map table: Microdata location · @type · Duplicates which block? · REMOVE / KEEP. For every REMOVE, paste exact attributes to strip.
3. Other killers found — one section per killer A–G that appears: location, plain-English problem, exact fix.
4. Verdict: pass Rich Results after fixes? Yes / Maybe / Unlikely + why. Re-test URL. Schema.org validator URL. Order of fixes (invalid JSON > missing required > duplication > other).
If no issues found, say so. Don''t fabricate problems.
Notes: GPT-5 with browsing can fetch the URL directly. The duplication map is reliable; double-check Section 3 manually for false positives on edge cases.
Gemini
Gemini 2.5 Pro
FRESH APR 2026
You are a Rich Results debugger. I''ll paste a page''s HTML. Find the #1 cause of Rich Results validation failures — same data in BOTH visible microdata (itemscope/itemtype/itemprop) AND JSON-LD —...You are a Rich Results debugger. I''ll paste a page''s HTML. Find the #1 cause of Rich Results validation failures — same data in BOTH visible microdata (itemscope/itemtype/itemprop) AND JSON-LD — plus 7 other common Rich Results killers.
PAGE CONTEXT:
• URL: [https://YOUR-DOMAIN.COM/...]
• Rich Results Test errors (if you''ve run it): [PASTE or NONE]
PAGE HTML (paste full source):
[PASTE]
WORKFLOW:
1. List every JSON-LD block + the @type each declares.
2. Scan HTML for itemscope, itemtype, itemprop attributes.
3. For each microdata element, identify which JSON-LD block expresses the same entity.
4. Where duplication exists → flag. JSON-LD wins; microdata gets stripped.
ALSO CHECK 7 KILLERS:
A. Orphaned AggregateRating.
B. Broken @id references.
C. Wrong @type for content.
D. Required Schema.org fields missing per @type.
E. Placeholder values ("Lorem ipsum", "Your Company Name", "[BRAND]", "example.com", "TODO").
F. Invalid JSON syntax.
G. Conflicting nested schemas.
OUTPUT:
1. JSON-LD Inventory table: Block # | @type | Valid JSON? | Required fields complete? | Placeholders? | @id used.
2. Microdata vs JSON-LD Duplication Map table: Microdata location | @type | Duplicates which block? | REMOVE / KEEP. For every REMOVE, paste exact attributes to strip.
3. Other killers found — one section per A–G that appears: location, plain-English problem, exact fix.
4. Verdict: pass Rich Results after fixes? Yes / Maybe / Unlikely + why. Re-test URL. Schema.org validator URL. Fix order: invalid JSON > missing required > duplication > other.
If no issues, say so. Don''t fabricate.
Notes: Gemini 2.5 Pro is excellent at killer F (invalid JSON syntax) — catches subtle issues other models miss.
Copilot
Copilot (GPT-5)
FRESH APR 2026
Find Rich Results validation killers on a page. I''ll paste a page''s HTML. Detect duplicate microdata + JSON-LD (the #1 cause of failures) plus 7 other common killers.
Page context:
• URL:...Find Rich Results validation killers on a page. I''ll paste a page''s HTML. Detect duplicate microdata + JSON-LD (the #1 cause of failures) plus 7 other common killers.
Page context:
• URL: [https://YOUR-DOMAIN.COM/...]
• Rich Results Test errors (if run): [PASTE or NONE]
Page HTML (paste full source):
[PASTE]
Workflow:
1. List every JSON-LD block + its @type.
2. Scan HTML for itemscope / itemtype / itemprop attributes.
3. For each microdata element, identify which JSON-LD block expresses the same entity.
4. Duplication → flag. JSON-LD wins; strip the microdata.
7 other killers:
A. Orphaned AggregateRating.
B. Broken @id references.
C. Wrong @type for content.
D. Required Schema.org fields missing per @type.
E. Placeholder values.
F. Invalid JSON syntax.
G. Conflicting nested schemas.
Output:
1. JSON-LD Inventory table: Block # · @type · Valid JSON? · Required fields complete? · Placeholders? · @id used.
2. Microdata vs JSON-LD Duplication Map table: Location · @type · Duplicates which block? · REMOVE / KEEP. For every REMOVE, paste exact attributes to strip.
3. Other killers: per A–G that appears — location, plain-English problem, exact fix.
4. Verdict: pass Rich Results after fixes? Yes / Maybe / Unlikely + why. Re-test URL. Validator URL. Fix order.
If no issues, say so. Don''t fabricate.
Notes: Copilot is reliable for this prompt — particularly strong on killer C (wrong @type) since it cross-references against Bing's structured-data inspector logic.
Frequently Asked Questions
What does the Find Duplicate Schema Errors Killing Rich Results prompt do?
Scan a page for the #1 cause of Rich Results validation failures: the same data expressed in BOTH visible microdata (itemscope/itemtype/itemprop) AND JSON-LD. Returns a clear list of which microdata attributes to remove from the visible HTML so JSON-LD can stand alone — plus a check for orphaned aggregateRating, broken @id references, and 5 other common Rich Results killers.
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.