Full Audit — extended AI visibility
This is your paid Full Audit (extended depth): a deeper version of the free mini-audit, run across more buying questions and more AI platforms — see Results by platform below for every engine's score.
Executive summary
Citation Share — whose sources AI points to
0% you · 11% competitors · 89% third-party
In 3 of the 9 AI answers we collected for your buying questions, AI named ahrefs.com instead of you — a signal of opportunity going elsewhere.
Start here — the action that closes the gap fastest
- Build authority on the third-party sources AI trusts for your category — get listed on, and link to, the relevant directories and review sites (see Source presence).
Open your live dashboard → Your results in one place: every quote and source AI cited, your score over time, and (with monitoring on) automatic re-checks. Sign in with this email.
These scores are directional, not precise: each is measured across 3 repetitions per question, and AI answers vary between runs.
What AI actually said about you
AI did not cite your company in the tested questions — the full transcript is in the appendix below.
This score isn’t one AI’s opinion of you. It’s measured: your buyers’ real questions, run across Perplexity, Gemini and Google AI Overviews — 3× each — counting how often AI cited you versus everyone else. Reproducible, not a vibe.
We show what AI actually does today. What you do with it is yours to decide — including nothing. We sell measurement and, if you want it, the work — never guaranteed citations.
A low score isn’t a verdict — it’s a starting line. The action above is where the gap closes fastest.
Appendix — full data
The complete dataset behind the summary above: every quote, every source AI cited, the competitors recommended in your place, your funnel-stage visibility, the SEO-vs-AI comparison and the full technical audit.
Results by platform
| Platform | Score | Citation frequency |
|---|---|---|
| Perplexity | 0/100 | 0% of queries cited your site (in the answer or its sources) |
| Gemini | 0/100 | 0% of queries cited your site (in the answer or its sources) |
| GPT-4o | 0/100 | 0% of queries cited your site (in the answer or its sources) |
| Claude | not measured | all queries failed (rate-limit/outage) — excluded from the score |
| Grok (xAI) | Scale & Protect | Only on Scale & Protect monitoring — continuous Grok tracking, not a one-time audit engine. |
What AI says about your company (verbatim)
None of the AI platforms named or quoted your company in the tested queries.
Hallucinations — inaccurate information
No hallucinations detected — AI’s information about your company appears consistent with your website.
Fact Integrity — is your approved truth reflected everywhere?
Where AI’s answers, your own pages, or third-party listings diverge from the facts you approved — worst first, with the one thing to fix. We measure consistency with your registry, not objective truth.
Who is recommended instead of you
| Domain | Citations |
|---|---|
| ahrefs.com | 3/9 queries |
Source Intelligence — where AI gets its information
When buyers ask AI in your category, it tends to name ahrefs.com instead of you. Those answers are built from the sources below — and your own site is only 0% of what AI cited. The lever: get present and credible on the third-party sources AI actually reads (marked Third-party). We can't guarantee AI will cite you — this raises the odds, and answers shift between runs.
Citation Share
Across the 132 sources AI engines cited for your 5 buying questions: 0% you, 11% competitors, 89% third-party.
Denominator by engine: Perplexity 88 · Gemini 44. Citation Share is your share of the sources AI cited — not merely whether you appeared (presence ≠ share). The count above includes every source each engine actually returned for your questions. Perplexity, Gemini and Google AIO surface sources on almost every answer; ChatGPT and Claude surface them only when they run a live web search, so their per-engine counts vary from run to run.
| Source domain | Citations | Cited by (platforms) | Whose |
|---|---|---|---|
| reddit.com | 7 | GeminiPerplexity | Third-party |
| youtube.com | 4 | GeminiPerplexity | Third-party |
| ahrefs.com | 3 | GeminiPerplexity | Competitor |
| qoulomb.com | 3 | Gemini | Third-party |
| adobe.com | 2 | Gemini | Third-party |
| alhena.ai | 2 | Perplexity | Third-party |
| builtin.com | 2 | Gemini | Third-party |
| dageno.ai | 2 | Perplexity | Third-party |
These are the domains AI relies on for your category. Being present and accurate on the high-frequency third-party sources increases how often AI can verify and cite you.
The questions we asked — exactly what we tested
Your AI Reputation score reflects these real buying questions your customers ask AI. We generate them from your site and industry, then freeze the set so every re-run is a like-for-like comparison. Each is tagged by funnel stage.
- how do I audit how often my brand is being cited by AI tools like ChatGPT Awareness
- best tool to check if my company is showing up in AI-generated answers Consideration
- how can I see if AI assistants are mentioning my competitors more than me Awareness
- is there a way to audit my visibility in large language model responses Awareness
- how do I find out what AI chatbots are saying about my business Consideration
5 distinct buying questions — each one asked across every AI engine and repeated for stability, which is why the total queries tested is higher.
Traditional SEO vs AI visibility
Search is shifting from ranked links to AI answers. Here is where you stand on both — for the same buying questions:
Traditional Google SEO · external estimate
0/100
Top 10 for 0 of 5 buying queries.
AI visibility — the growing channel
0/100
How often AI assistants cite you for the same questions.
The Google number is a directional estimate from a live Google check (Serper), not Search Console. Google's own guidance is to judge SEO by Search Console trends over time, not a single rank.
You don't own your brand SERP yet. Google's top results for your own brand name don't point to you — typically a young domain and/or a same-named site competing for the term. Read this number as "not indexed for your brand yet," not "invisible." Owning your brand SERP is the first, highest-leverage fix.
Businesses Google ranks for these queries: triplewhale.com, mindstudio.ai, campaigncreators.com, partnerstack.com, duaneforresterdecodes.substack.com.
Google still drives traffic today, but its share is declining while AI answers grow. A gap between the two is the early warning: the channel that's growing isn't recommending you yet.
Off-domain presence
AI corroborates a business across third-party sources — and joins those sources to you through the links on your own site. The more of your real profiles you link, the more of your authority AI can connect.
Linked from your site — AI can already connect these to you:
- LinkedIn Company Page
This checks which profiles you link from your own site. We don't assume you're absent from a platform we can't verify — it's about connecting the profiles you already have so AI joins the dots.
Technical audit
- robots.txt — OAI-SearchBot access (ChatGPT Search)
- robots.txt — PerplexityBot access
- robots.txt — ChatGPT-User access
- robots.txt — Perplexity-User access
- robots.txt — ClaudeBot access
- llms.txt (informational — not scored)
- sitemap.xml
- Schema.org JSON-LD
- Meta tags (title + description)
- HTTPS
- Content available without JavaScript
- LinkedIn Company Page link on your site
Detected Schema.org types: Organization, WebSite, Product, SoftwareApplication, BreadcrumbList, FAQPage
| ✓ | llms.txt present | 15 pts |
| ✓ | Content readable without JS | 20 pts |
| ✓ | Machine-readable pricing file (/pricing.md) | 20 pts |
| ✓ | Price in structured data or server-rendered HTML | 25 pts |
| ✓ | Pricing not hidden behind "contact sales" | 10 pts |
| ✓ | Agent instructions (AGENTS.md) | 10 pts |
Would an AI agent make it all the way to you?
| 1 | ✓ | Discover you | Server-rendered content with a sitemap/llms.txt — the agent can crawl you. |
| 2 | ✓ | Read your offering | Machine-readable pricing/offering (pricing.md / structured data / visible price). |
| 3 | ✓ | Compare you | Structured data lets the agent line you up against alternatives cleanly. |
| 4 | ✓ | Act on you | AGENTS.md gives the agent explicit instructions to act on your behalf. |
Funnel from your machine-readable signals; verdict from a live reasoning pass over your signals + competitors.
Approve your pricing, policies and key facts in CitePulse and we publish them as a signed, versioned feed plus a Model Context Protocol (MCP) endpoint. AI agents — ChatGPT-class answer engines and MCP clients like Claude, Cursor and Codex — then read your facts directly and verify the signature offline, no scraping. The signature attests you approved each value — not that it's objectively true.
| Signed feed | https://api.citepulse.ai/.well-known/brand-facts/citepulse.ai.json |
| MCP endpoint | https://api.citepulse.ai/mcp/citepulse.ai |
These endpoints are already live for citepulse.ai — empty until you approve your facts. Approve them in CitePulse and agents get a verifiable answer.
Action plan
Do this first
Build authority on the third-party sources AI trusts for your category — get listed on, and link to, the relevant directories and review sites (see Source presence).
What an AI assistant needs before it can cite you
- ✓ Reach & read you. AI answer-crawlers (ChatGPT, Perplexity, Claude) are allowed and your content is readable without JavaScript.
- ✓ Understand what you do. structured data (Schema.org) and a clear title/description let AI parse who you are.
- ✓ Corroborate you elsewhere. you’re linked from at least one trusted third-party profile AI can cross-check.
These are prerequisites, not guarantees: they let AI cite you — being the clearest, best-corroborated answer to your buyers’ questions is what makes it choose to.
Your technical foundation is solid
Schema, robots.txt and llms.txt are all in place. Your gap isn't technical: AI doesn't cite you because it doesn't yet connect you to the third-party sources it trusts. That's a visibility and authority problem, not a code problem.
Keep building presence on — and links to — the third-party sources AI trusts for your category (see Source presence above).
As you build that presence, CitePulse monitoring will catch the moment AI starts citing you — or a competitor like ahrefs.com takes the spot. Monitoring is available on paid plans; we deliver monitoring, not guaranteed AI-citation outcomes.
Methodology & disclaimer
How we measured this. We generated the real buying questions your customers ask, then ran 5 of them across 4 AI platforms — 3 repetitions each — and counted how often AI cited your company versus everyone else. The technical checks read your live site the way an AI crawler does. Scores are directional, not precise: AI answers vary between runs, so treat these as a reliable signal of your standing, not a guaranteed constant.
Correlation, not causation. When AI names a competitor on your buying questions, we report it as an observed signal of opportunity going elsewhere — not proof of lost revenue or that the competitor is a direct rival. We do not, and cannot, attribute a dollar figure to it.
What we deliver. Measurement, the specific fixes your audit surfaced, and
ongoing monitoring on paid plans. We do not guarantee that any AI assistant will
cite or recommend your business. llms.txt is agent-readiness hygiene for the
emerging agentic web — an emerging convention, not a ranking or citation factor. Sentiment
(how AI talks about you) and hallucinations (where AI states something untrue) are distinct
and reported separately.
These results are informational and reflect AI responses observed at the time of testing. Not professional advice.