The AI Visibility Score Checker is a free tool that measures how visible your brand is when people ask AI assistants for recommendations. It asks 25 realistic buyer questions to 4 independent AI engines, verifies every brand mention against the raw answer text, and turns the results into a single 0 to 100 score you can track over time.
This documentation explains exactly how that score is produced: which engines are tested, how the queries are designed, what the scoring weights are, and how to read every section of your report. There is no black box. The same formula runs for every brand.
Why AI visibility matters
A growing share of product discovery now happens inside ChatGPT, Gemini, Perplexity, and Google AI Overviews instead of classic search results. When someone asks an AI assistant for the best tool, agency, or product in your category, the assistant names a handful of brands and the conversation usually ends there. Unlike a search results page, there is no page two. You are either named in the answer or you are invisible.
Traditional SEO tools cannot see this layer. Rank trackers tell you where you stand in Google's blue links, but nothing about whether GPT or Gemini recommends you. The AI Visibility Score exists to make that layer measurable, repeatable, and comparable across brands and across time.
What a scan gives you
- An overall score from 0 to 100 with a tier label from Invisible to Leader, built from up to 100 real AI answers.
- A per-engine mention matrix showing which engines know your brand and which ones never mention you.
- Share of voice and a competitor table: every brand named in every answer is tallied, so you see who AI engines recommend instead of you.
- A content gap roadmap generated from the queries where you did not appear, with recommended topics, formats, and priorities.
- Evidence for every claim: each counted mention ships with the exact excerpt from the AI answer where your brand appears.
How a scan works, at a glance
| Step | What happens |
|---|---|
| 1. Query generation | 25 buyer queries are generated for your brand and keywords across 5 intent categories, from direct brand checks to comparison shopping. |
| 2. Engine panel | Each query is asked independently to live models from OpenAI, Google, and DeepSeek, plus a web-grounded AI search leg that works like Perplexity. |
| 3. Mention extraction | Every answer is analyzed for whether your brand appears, its position, how strongly it is recommended, and the sentiment. Claimed mentions are re-verified verbatim against the raw text. |
| 4. Scoring | Each query and engine pair becomes one scored sample. Your overall score is the average across all samples, using five published weights. |
The details live in the next three chapters: Engines and queries, How the score is calculated, and Accuracy guardrails.
Running your first scan
- Open the AI Visibility Score Checker.
- Enter your brand name exactly as it is written publicly. The verification layer only counts literal matches, so "Acme Analytics" and "AcmeHQ" are different brands to the checker.
- Add 2 to 5 keywords that describe your category, the way a customer would say them. These drive the category, problem, and long-tail queries.
- Optionally add your website URL. This unlocks the source authority signal, which checks whether your own domain shows up in the live web sources AI engines draw from.
A scan takes a couple of minutes because it waits on up to 100 real model answers. Results are kept for 24 hours per brand and website, so repeat checks within a day return the same report. For tracking progress we recommend re-running monthly, or after major content launches and PR coverage.