What the number actually represents
A Visibility Score exists to answer a simple question in a single glance: is a brand's overall AI presence improving or declining. Underneath that one number sits a combination of the more specific metrics this glossary covers separately, most commonly how often the brand is mentioned, how often it gets cited with a link, and whether the tone of those mentions is favorable. Different tracking tools weigh these inputs differently, and none of them has settled on one universal formula, so a Visibility Score from one tool is not directly comparable to a Visibility Score from another, even if both display a number between 0 and 100. The score is best treated as a trend line for a single brand tracked by a single method over time, not as an absolute figure that means something on its own or against a score produced by a different tool. Some tools also factor in position within the answer, or the number of assistants a brand appears on, which further explains why the same underlying data can produce different scores depending on which tool calculated them.
How it works
Building a Visibility Score starts with the same underlying data every other metric on this page draws from: a defined prompt set run repeatedly across the AI assistants a brand's buyers use, with every mention, citation and sentiment reading logged. Each of those readings gets converted into its own sub-score, often a percentage, and the sub-scores are then combined using a weighting formula the tracking tool defines, for example giving citation a heavier weight than a plain mention, since a citation demonstrates a stronger result. That weighting choice is exactly why scores are not portable between tools: two tools measuring the identical brand on the identical prompts can output different composite numbers if one weighs sentiment more heavily than the other. Because the underlying inputs already require repeated runs to be reliable on their own, a Visibility Score inherits that same requirement. A score calculated from a single check carries all the same instability as the individual metrics it is built from, just hidden behind one tidier-looking number. The score becomes genuinely useful only once it is tracked on a fixed schedule, so a team can see the line move, rather than treating any single reading as a verdict on where the brand stands.
Why it matters for AI visibility
A single composite number is easier to put in front of a leadership team than five separate metrics, which is the main reason a Visibility Score is useful at all: it turns a technical tracking exercise into a figure that can sit on a dashboard next to traffic and conversion numbers. The risk sits in the same place as the convenience. A rising score can hide a falling citation rate if mention volume happens to be climbing faster, and a team that watches only the composite number can miss exactly which underlying driver is moving. The score works best as a headline metric that prompts a deeper look at its components when it moves, not as a replacement for tracking those components individually. Reporting the trend alongside at least one or two of the underlying drivers, rather than the composite number alone, keeps the score honest about what is actually changing.
Good practices
- Track the underlying metrics, mention rate, citation rate, sentiment, alongside the composite score, not instead of it.
- Compare your score against the same tool's score for named competitors, never against a score reported by a different tool.
- Ask any tracking tool what weighting formula it uses before treating the number as a precise measurement rather than a directional signal.
- Watch the trend over weeks or months rather than reacting to a single reading.
- Investigate a moving score by checking which underlying metric actually shifted, before assuming the whole picture changed evenly.
- Report the score alongside its components when sharing results internally, so a drop is explainable rather than just alarming.
Common mistakes
- Comparing a Visibility Score from one tracking tool against a score from another, when the two use different formulas and inputs.
- Treating the composite score as precise to the point, when it is a directional summary built from several noisier metrics.
- Reacting to a single reading instead of watching the trend across repeated, scheduled checks.
- Reporting the score without any of its underlying components, which hides which specific metric actually caused a change.
Related terms
- Share of Voice in AI: one of the metrics a Visibility Score commonly folds in as an input.
- Citation Rate: the linked-mention metric that often carries the heaviest weight inside a composite score.
- Sentiment in AI Answers: the tone reading that adjusts a composite score beyond a raw mention count.
- Brand Mention: the raw counted event several of a Visibility Score's inputs are built from.
Frequently asked questions
Is a Visibility Score of 80 good?
It depends entirely on the tool that produced it and the competitors it is compared against. A score of 80 from one tool is not equivalent to 80 from another, since each defines and weighs its inputs differently. The only reliable read is the same tool's score for your brand over time, against the same tool's score for your named competitors.
Why did my Visibility Score drop even though mentions increased?
A composite score usually weighs more than raw mention count. If citations dropped or sentiment turned less favorable while mentions rose, the combined score can fall even as one input improves. Checking the underlying components separately is the only way to see which one actually moved and caused the change.
Can two competitors have very different scores with similar mention rates?
Yes. If one competitor's mentions come with more citations or more favorable sentiment, its composite score can sit well above a brand with a similar raw mention count but weaker results on those other inputs. Mention rate alone rarely explains the full gap between two scores.
How often should a Visibility Score be recalculated?
On the same schedule as the underlying metrics it is built from, typically weekly, so the composite figure reflects current data rather than an outdated snapshot. Recalculating less often risks reporting a score that no longer matches what AI assistants are actually saying about the brand right now.
See your own AI visibility
Truffle combines mention rate, citation rate and sentiment into one visibility score, tracked over time against your named competitors. See where your brand stands today.
Start free trial See how it worksFrequently asked questions
Is a Visibility Score of 80 good?
It depends entirely on the tool that produced it and the competitors it is compared against. A score of 80 from one tool is not equivalent to 80 from another, since each defines and weighs its inputs differently. The only reliable read is the same tool's score for your brand over time, against the same tool's score for your named competitors.
Why did my Visibility Score drop even though mentions increased?
A composite score usually weighs more than raw mention count. If citations dropped or sentiment turned less favorable while mentions rose, the combined score can fall even as one input improves. Checking the underlying components separately is the only way to see which one actually moved and caused the change.
Can two competitors have very different scores with similar mention rates?
Yes. If one competitor's mentions come with more citations or more favorable sentiment, its composite score can sit well above a brand with a similar raw mention count but weaker results on those other inputs. Mention rate alone rarely explains the full gap between two scores.
How often should a Visibility Score be recalculated?
On the same schedule as the underlying metrics it is built from, typically weekly, so the composite figure reflects current data rather than an outdated snapshot. Recalculating less often risks reporting a score that no longer matches what AI assistants are actually saying about the brand right now.
