Measure · Visibility
Is your brand in the AI answer?
We ask ChatGPT, Gemini and Perplexity the questions your customers ask, and record whether your brand appeared, in what position, and with which source. On Agency and Enterprise, Claude can be added as an option.
| ChatGPT | Gemini | Perplexity | |
|---|---|---|---|
| 1. | 3. | 2. | |
| Not mentioned | 1. | Not measured | |
| 2. | Not mentioned | 1. | |
| 3. | 2. | Not mentioned | |
| Not mentioned | Not measured | 3. |
How is it measured?
- 01
You choose the prompts.
Questions your customers really ask. We suggest starter prompts for your industry; you pick.
- 02
We ask the engines.
Each prompt goes to every engine you chose, with web search on. Depending on your plan, the same prompt is repeated up to 3 times.
- 03
We read the answer.
Did you and your competitors appear, in what position, which sites were cited, and was the tone positive or negative.
- 04
We calculate the score.
Results are stored with their sample and dates. We don't keep the raw answer, only the short excerpt where the brand appears.
Not just a number, but the reason behind it.
Mention rate
How many of our prompts mentioned your brand.
Position
If it appeared, in what position. Being first is not the same as being fifth.
Competitive share
Your share of mentions against your competitors on the same prompts.
Citations
Which sites the AI cited: yours, a complaints site, or a competitor.
Sentiment
Whether your brand was described positively, neutrally or negatively.
Trend
Same prompts, same engines, over time. When it dropped and when it rose.
How we calculate the score is no secret.
Visibility score = mention rate x 0.40 + position score x 0.35 + competitive share x 0.25
No measurement, no score; we never fill gaps with guesses. Every score shows how many prompts, engines and repeats, and which date range, it's based on.
The visibility score is separate from your readiness score. One measures your site, the other measures AI answers.
How does AI describe your brand?
We ask different engines the same four questions: do they know you, what do they say, who do they consider your competitors, and does it match your site. It's not a score, it's a snapshot. The first snapshot per site is free.
- RecognizesPartially
- SentimentNeutral
- Competitors named3
- ConsistencyConflicts (founded 2012 on the site, 2015 in the answer)
Which engines do we measure?
| Engine | How we ask | Status |
|---|---|---|
| ChatGPT | Web search on | Measured |
| Gemini | Google Search grounding on | Measured |
| Claude | Web search on | Optional on Agency and Enterprise |
| Perplexity | Its own search model | Measured |
Product names are used only to describe which services we measure. We are not affiliated with or endorsed by these companies.
AI answers vary by person, location and time. We measure the answer we got at a specific moment under specific conditions; we don't claim it's the exact answer any given user will see.
On the Agency plan you can measure with each engine's most capable model (16 credits for ChatGPT and Claude, 4 for Gemini, 2 for Perplexity; two to four times the standard rate).
One prompt asked once on one engine costs credits by engine: ChatGPT 4, Gemini 2, Perplexity 1, Claude 8. Example: 10 prompts x 3 core engines (ChatGPT, Gemini, Perplexity) = 70 credits. Web search and sentiment analysis included.
Measure your first 5 prompts today.
A free account comes with 100 credits. Measuring 5 prompts on 3 engines (ChatGPT, Gemini, Perplexity) costs 35.