How tools calculate it
Peec AI defines its visibility score as the percentage of AI responses that mention your brand. Profound divides the runs that mention your brand by the runs that mention any brand, works that out per model and averages the models with equal weight. Semrush scores 0 to 100 from two factors, how many topics include your brand in AI answers and how consistently it is mentioned within them, without publishing the weighting.
Otterly.AI publishes no single score: its brand report places brands on two axes, the share of prompts that mention you and a likelihood-to-buy figure worked out from your average position. HubSpot’s glossary describes a brand visibility score as, typically, the percentage of tracked prompts whose answers mention the brand.
How AEOHIT calculates its score
AEOHIT’s AI Visibility Score combines five readings from the answers it captures on seven engines: presence (how often you appear), prominence (how early), share of voice (you against your competitors), sentiment, and citations (whether engines cite your site as a source). Each engine is weighted by its real-world referral share, and the weights are recalibrated as the market shifts. The methodology page names the source of every answer.
Why scores from different tools do not compare
Each tool asks different questions, on different engines, from different countries, on different days, and divides by something different: all responses, responses that mention any brand, or topics. A 40 in one tool and a 25 in another can describe the same week. Compare like with like: your own score over time, and your competitors’ scores on the same questions in the same tool.
How to read yours
Watch the trend over several weeks rather than one reading, because answers vary by day, location and phrasing. Look at the per-engine breakdown, since one engine can name you while another never does. Read it with share of voice: a rising score while a competitor rises faster is still a lost position. A low first score is common, and it is the baseline your fixes are measured against.
What it cannot tell you
A score says how often you are named, not why. It does not show which page an engine read, whether what it said about you was right, or what the customer did next. The answers themselves do: the claims engines make about you, the sources they cite and the competitors named beside you are where the fixes come from.
How to raise it
Find the questions where a competitor is named instead of you and the sources the engines cite for them. Then answer those questions plainly on your own pages, add structured data, make sure AI crawlers can read your site, keep your business details consistent, and get listed where the engines already look. Re-measure the same questions after each change.
Sources
- Peec AI docs: visibility
- Profound docs: how metrics are calculated
- Semrush knowledge base: AI visibility data
- Otterly.AI help centre: brand report KPIs
- HubSpot glossary: brand visibility score
Checked on .