Methodology
How VisiScan measures AI visibility
TL;DR: VisiScan asks real buyer questions to up to four AI engines, repeats those questions, detects business and competitor mentions, records citations, and turns the observed answers into a transparent score.
What AI visibility is
AI visibility is whether an answer engine — ChatGPT, Claude, Perplexity, or Gemini — names a business when a buyer asks it for a recommendation. It is a different measurement from search rank: a page can rank first in Google and still never be named in the AI answer shown above those results, and a business with no first-page ranking can be the one the engine recommends.
The practice of improving it is called generative engine optimization, or GEO. VisiScan does not do the optimizing; it does the measurement — a repeatable, dated reading of what the engines actually say, so any change you make has a before and after instead of a guess.
Every figure in a report is an observation of a changing system on a specific date, tied to the exact question, engine, sample, and timestamp that produced it. It is not a ranking and not a promise of future recommendations.
What a scan measures
- Whether the scanned business is named in answers from ChatGPT, Claude, Perplexity, and Gemini.
- Which competitors are named instead, including the position and context observed in the answer.
- Which websites and domains the engines cite as supporting sources.
- Paid reports add Brave grounding, competitor citation gaps, cited-answer verification, and applicable local profile evidence. These are evidence surfaces, not a fifth primary AI engine.
- More than 30 on-site and off-site AI visibility signals used for readiness recommendations.
How a scan works
- VisiScan profiles the business website and identifies its market and buyer context.
- It generates 5 buyer-intent questions for the free scan or 12 for the full report.
- Each question is sent to up to four engines (ChatGPT, Claude, Perplexity, and Gemini). Full reports run 3 samples per question and engine.
- The returned answers are checked for the business name, competitors, and cited sources.
- The report combines those observations with readiness checks and cites the underlying evidence.
How the score is calculated
Visibility is the weighted percentage of tested question-and-engine cells where the business is named. Trust and comparison questions are weighted 1.5x because they are closer to a purchase decision. Best-near-me, emergency, and price questions use 1x weight. Repeated samples are majority-voted per question and engine, and confidence intervals are shown where sample size matters.
The separate paid Citation Score weights query coverage at 50%, reciprocal rank at 20%, competitive share at 20%, and surface coverage at 10%. If a component fails or does not apply, it is excluded and the remaining weights are renormalized. Brave cited-answer verification is additional evidence and does not change the four-engine Visibility Score.
Data provenance and freshness
Every answer has an engine, sample, timestamp, and measurement mode. Live mode uses configured provider APIs; replay or unavailable-provider data is labeled in the report. Results are observations of a changing AI response, not a permanent ranking or a guarantee of future recommendations.
Plans
| Plan | Coverage | Price |
|---|---|---|
| Free scan | 5 questions × up to 4 engines | $0 |
| Full report | 12 questions × up to 4 engines × 3 samples | $49 one-time |
| Monitor | Automatic weekly re-scans and trends | $29/month |
Frequently asked questions
What does VisiScan measure?
VisiScan measures whether ChatGPT, Claude, Perplexity, and Gemini name a business in buyer-intent answers, which competitors they mention instead, and which sources they cite.
How is the Visibility Score calculated?
The score is a weighted named-rate across tested questions and AI engines. Trust and comparison questions receive 1.5x weight; best-near-me, emergency, and price questions receive 1x weight.
Are VisiScan results real AI answers?
Live scans query configured AI providers and record the returned answer, engine, sample, and provenance. A report identifies its measurement mode so replay or unavailable-provider results are not presented as live data.
How is the paid Citation Score calculated?
The paid Citation Score weights query coverage at 50%, reciprocal rank at 20%, competitive share at 20%, and surface coverage at 10%. Failed or inapplicable components are excluded and the remaining weights are renormalized.
Is AI visibility the same as SEO?
No. SEO measures where a page ranks in the list of blue links. AI visibility measures whether an answer engine names the business in the answer it writes, which is decided by retrieval, relevance, authority, and the engine's own generation step. The two can disagree in both directions.
How often should a business re-scan?
AI answers change week to week as engines update. A one-time scan is a baseline; the Monitor plan re-runs the full scan weekly so a change in named-rate can be separated from normal sample-to-sample variation.
Can VisiScan guarantee my business will be recommended?
No. A scan reports what the engines currently say and which readiness gaps are behind it. Whether a fix changes the answer depends on each engine's retrieval and generation, which no external tool controls.