Back to AI Search & SEO

Diagnose Why AI Engines Aren't Citing You

You're ranking on Google but ChatGPT, Perplexity, and Google AI Overviews never mention you. This agentic workflow walks the four-layer diagnostic — crawl access, schema coverage, content extractability, and entity authority — and tells you which layer is the bottleneck plus the next 3 actions to unblock it. Designed for businesses that already have basic SEO in place.

Expert MULTI-STEP WORKFLOW Revenue-driving
Pro tip

AI engines fail to cite you for one of four reasons, in order of frequency: (1) they're blocked from crawling, (2) you have no extractable answer on the page, (3) your schema is invalid or absent, (4) you're not recognized as a credible entity. Fix in that order — fixing schema while you're still blocked from crawling wastes weeks. This prompt finds the bottleneck before you spend.

diagnostic citations aeo audit ai-search authority

How to use this prompt

  1. Pick your AI model. Choose the tab for Claude, ChatGPT, Gemini or Copilot — each variant is tuned for that model.
  2. Copy the full prompt. Click Copy Full Prompt to copy the text to your clipboard.
  3. Paste into your AI tool. Open your chosen model and paste the prompt into a new chat.
  4. Replace the [placeholders]. Swap any bracketed fields for your company name, audience, product or tone.
  5. 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 senior AI-search diagnostician. My site ranks fine on Google but I''m not getting cited by ChatGPT, Perplexity, Google AI Overviews, or Claude. Walk the four-layer diagnostic, identify the...
You are a senior AI-search diagnostician. My site ranks fine on Google but I''m not getting cited by ChatGPT, Perplexity, Google AI Overviews, or Claude. Walk the four-layer diagnostic, identify the...

Unlock this prompt + all 92 in the library — free, updated monthly, zero spam.

Your email unlocks all 92 prompts instantly. No password, no signup — just your email. 

You are a senior AI-search diagnostician. My site ranks fine on Google but I''m not getting cited by ChatGPT, Perplexity, Google AI Overviews, or Claude. Walk the four-layer diagnostic, identify the bottleneck, and tell me exactly what to fix first. This is an agentic workflow — work through every layer in order; don''t skip ahead.

<my_situation>
Business: [NAME] — [WHAT YOU DO IN ONE SENTENCE]
Top 3 queries I''d expect AI engines to cite us for: [Q1], [Q2], [Q3]
Where I tested (paste verbatim what each engine said when I asked these queries):
  - ChatGPT response: [PASTE the full text of the AI''s response, including any citations]
  - Perplexity response: [PASTE]
  - Google AI Overview: [PASTE — or "didn''t appear" if no AI Overview triggered]
  - Claude response: [PASTE]
My homepage URL: [https://YOUR-DOMAIN.COM]
A representative money page (the page I most want cited): [https://YOUR-DOMAIN.COM/...]
Monthly Google organic traffic order of magnitude: [<1k / 1k–10k / 10k–100k / >100k]
Domain age: [N years]
Have I run Google''s Rich Results Test on the money page? [yes — paste result / no]
</my_situation>

<diagnostic_workflow>
Walk through these four layers in this exact order. Do not skip ahead — at each layer, reach a verdict (PASS / FAIL / UNKNOWN) before moving on.

**LAYER 1 — Crawl Access**
Can AI engine crawlers actually fetch my pages?

Check these signals from what I gave you:
- Is robots.txt fetchable at the domain? Does it explicitly allow GPTBot, OAI-SearchBot, ChatGPT-User, PerplexityBot, Perplexity-User, ClaudeBot, Claude-Web, anthropic-ai, Google-Extended, Applebot-Extended, CCBot? Or is it default-allow with no explicit listing?
- Is there a sitemap.xml at the domain referenced from robots.txt?
- Does the homepage and money page return 200 to a normal user-agent fetch?
- Are there any subtle blocks: <meta name="robots" content="noindex">, X-Robots-Tag headers, Cloudflare bot-fight mode, JS-only content with no SSR fallback?

If you can''t verify a signal from what I gave you, ask me to fetch it (provide the exact curl command or URL).

VERDICT for Layer 1: PASS / FAIL / UNKNOWN + 1-sentence reasoning.

If FAIL → STOP. Don''t assess Layers 2–4 yet. The other layers don''t matter until crawlers can reach you.

**LAYER 2 — Content Extractability**
Even if crawlers reach the page, can they extract a citable answer?

Check the money page''s visible content (use the rendered text, not just HTML):
- Does the first paragraph directly answer one of my target queries in ≤50 words?
- Is there a sentence on the page that''s 20–40 words, contains the primary entity in the first half, includes a specific number/date/methodology, and is self-contained (no anaphora)?
- Are there 5+ FAQs phrased as real user queries (how/what/can/is/why/when)?
- Is the answer buried under brand-affirmation copy ("Welcome to AcmeCo, your trusted partner...")?

VERDICT for Layer 2: PASS / FAIL / UNKNOWN.

If FAIL → STOP HERE. Adding schema while content is unextractable is wasted effort.

**LAYER 3 — Schema Coverage & Validity**
Is the structured data telling the right story to AI engines?

Check (using anything I''ve given you and what you can request):
- Does the homepage have valid Organization + WebSite JSON-LD?
- Does the money page have at least one page-specific schema (FAQPage / HowTo / Article / Product / LocalBusiness)?
- Any Rich Results errors on these pages?
- Any duplicate microdata + JSON-LD?
- Any orphaned AggregateRating?

VERDICT for Layer 3: PASS / FAIL / UNKNOWN.

**LAYER 4 — Entity Authority**
Does the open web recognize us as a real, credible entity in our domain?

Check from what I gave you (and ask me for what''s missing):
- Wikipedia / Wikidata entry for the business or its founders? (search wikipedia.org and wikidata.org)
- Press / podcast mentions linking to us from ≥5 reputable domains in our industry?
- Founders'' LinkedIn URLs surfaced from the About page?
- Sameas references from authoritative directories (Crunchbase, industry-specific equivalents)?
- Domain age + organic traffic order of magnitude (small + new = harder to break in)?

VERDICT for Layer 4: PASS / FAIL / UNKNOWN.
</diagnostic_workflow>

After walking all four layers, produce these final outputs:

**PART 1 — Layer Verdicts**
Markdown table: Layer · Verdict · One-sentence reasoning.

**PART 2 — Bottleneck Identification**
Which is the FIRST failing or unknown layer, in order? That''s the bottleneck. One paragraph explaining why this is what''s blocking citations.

**PART 3 — Cited-Competitor Pattern**
From the AI responses I pasted: which 2–3 competitors are AI engines actually citing for my queries? What pattern do they share that I don''t? (Schema present? Wikipedia entry? Specific data on the page? News mentions?)

**PART 4 — The Next 3 Actions**
Numbered. Each: what to do, expected impact, time to do it, which AlignAI prompt to use if applicable. ONLY actions that unblock the identified bottleneck. Do not list actions for layers below the bottleneck.

**PART 5 — Re-Test in 30 Days**
The exact set of queries to re-run on each engine after the actions ship, so I can measure whether the bottleneck cleared.
Notes: Claude is the strongest variant for the multi-layer reasoning. The "STOP if FAIL" pattern is critical — without it, models tend to dump a kitchen-sink fix list and miss the bottleneck.

Frequently Asked Questions

What does the Diagnose Why AI Engines Aren't Citing You prompt do?

You're ranking on Google but ChatGPT, Perplexity, and Google AI Overviews never mention you. This agentic workflow walks the four-layer diagnostic — crawl access, schema coverage, content extractability, and entity authority — and tells you which layer is the bottleneck plus the next 3 actions to unblock it. Designed for businesses that already have basic SEO in place.

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.

More in AI Search & SEO

AI Search Readiness Audit & Fix Plan

Build an agentic workflow that ingests your homepage HTML, scores AI-search readiness on a...

Intermediate Revenue-driving

Generate FAQs Your Customers Actually Search For

Generate 10 FAQ Q&A pairs phrased the way real customers type into ChatGPT, Google and voice search...

Beginner Revenue-driving

Rewrite Your Homepage Hero for AI Answer Engines

Restructure your homepage's first 50 words so AI answer engines (ChatGPT Search, Perplexity, Google...

Intermediate Revenue-driving

Don’t see what you need? tailored to your use case.