Skip to content
English
Contact us
  • There are no suggestions because the search field is empty.

How is my visibility score (and Share of Voice) calculated?

GetMint turns thousands of AI answers into a few clear numbers — here's what each one means.

Visibility is the percentage of relevant AI responses that mention your brand naturally — when someone asks a question without naming any brand (e.g. "best project management tool for startups"). It's your core score: how often AI surfaces you unprompted.

Share of Voice (SoV) compares you to your competitors for those same questions. If you're mentioned in 30% of answers and your three competitors split the rest, SoV shows whether you're winning or losing the answer. Example: if a brand is named in 50 of 100 responses and there are 500 total brand mentions across everyone, that brand's share is 50 / 500 = 10%.

Sentiment scores how positively AI describes you, roughly: (positive mentions − negative mentions) ÷ total responses.

Brand Alignment measures whether AI describes you accurately — your positioning, pricing, and key features.

The GetMint Score at the top of your dashboard is a single 0–100 health metric that aggregates Visibility, Sentiment, and Alignment into one headline number. Use it for the big picture; open each module to see what's driving it.

Two things worth knowing:

  • Because models are non-deterministic, every score is a probability from repeated simulations, not a fixed ranking — so it moves a little week to week.
  • Each engine (ChatGPT, Perplexity, Gemini) pulls from different sources, so your scores won't match across them. That's expected and useful.

A low or zero score isn't a failure — it's your baseline. The point is to move it over time.

Key takeaways

  • Visibility counts brand mentions in AI answers; Share of Voice compares you to competitors; Sentiment scores how positively AI describes you; Brand Alignment measures accuracy.
  • The GetMint Score (0–100) rolls Visibility, Sentiment, and Alignment into a single headline number.
  • Every score is a probability from repeated simulations — it moves week to week. Read the trend.
  • Scores differ across engines because each engine consults different sources. Treat each as its own surface.