> ## Documentation Index > Fetch the complete documentation index at: https://promptwatch.com/docs/llms.txt > Use this file to discover all available pages before exploring further. # Visibility score > The Visibility Score measures how prominently your brand appears in AI answers on a 0-100 scale: every response is scored on placement, share of attention, and depth, then averaged, with answers that don't mention you counting as zero. ### What is the Visibility Score? The Visibility Score measures how prominently your brand appears when AI models answer your tracked prompts, on a 0-100 scale. Every analyzed response gets its own score, and the number on your dashboard is the average across all of them, with responses that don't mention you counting as zero, so one score reflects both how often you appear and how well you appear when you do. The Brand Visibility chart on the Promptwatch project dashboard, with the brand list beside it showing your brand's and each competitor's average visibility score. Visibility measures prominence, not favorability. An answer built entirely around problems with your product is highly visible, and it scores that way; how positively you're described is the [sentiment score](/docs/academy/sentiment-score)'s job, and the two are scored independently so neither can hide the other. ### How a single response is scored During analysis, every brand a response substantively discusses gets a score from 1 to 100 for how prominently it's presented as an answer to the prompt. The score is based on placement, share of attention, depth, repetition, structural emphasis, and relevance: * **Placement sets the baseline**: the sole recommendation scores 100 and a dominant one among several 95-99, the first entry in a ranked list lands at 90-94, a major secondary focus 81-89, second place 65-80, third 50-64, and fourth or later 25-49. A brief passing mention scores 10-24, and a minimal one 1-9. * **Attention and depth adjust it**: a dedicated heading, repeated mentions, or substantial comparison detail push the score above the placement baseline, while being merely enumerated, hedged, or overshadowed by another brand pulls it below. List position is a starting point, not the whole story, so an answer can name you first and still score you modestly if it spends three paragraphs on a competitor. * **Criticism doesn't lower it**: a response devoted primarily to severe problems with your product scores in the 90s on visibility despite the negative tone, because your brand dominates the answer. That combination, high visibility with low sentiment, is exactly the pattern worth catching. * **Absent means zero**: if the response doesn't substantively mention your brand, it scores 0. Statements like "I don't have information about your brand" count as not mentioned, so a raw mention count would mislead you: a brand can technically appear in every answer and still have almost no real presence. Names appearing only in cited URLs or source titles don't count either, the same rule that separates [citations from mentions](/docs/academy/citations-vs-mentions). ### How the number you see is calculated The score you see is the sum of all response scores divided by the total number of analyzed responses in your selection, zeros included. Dividing by all responses rather than just the ones that mention you is deliberate: it makes the score reflect your presence across everything AI is being asked, so there are exactly two ways to move it, appear in more answers, or appear more prominently in the answers you're already in. The chart plots this average per day, and analysis scores every brand in every answer, not just yours, so the brand list next to the chart shows your competitors' averages on the same scale. The gap between your average and the leader's tells you whether your problem is coverage or prominence. ### Visibility vs. the neighboring metrics * **[Share of Voice](/docs/academy/share-of-voice)**: how often you appear relative to competitors, presence only. A footnote mention counts the same as the headline pick, so read it next to visibility to separate "we don't appear" from "we appear badly". * **[Position in answer](/docs/academy/position-in-answer)**: where you land in the order of brands, counted from 1. Position is the starting point of the visibility score, which then adds share of attention and depth on top. * **[Sentiment score](/docs/academy/sentiment-score)**: how positively you're described, scored separately so prominence never leaks into tone or vice versa. For how all six core metrics fit together, see the [Metrics overview](/docs/academy/metrics-overview). ### Where to find it The Brand Visibility chart sits on your project dashboard, on each monitor page, and on each prompt page, with chart-type toggles for line and bar views of visibility plus Share of Voice and position views. You can slice it by date range, AI model, prompt type, and topic; filtering to a single model is usually the most revealing cut, because a strong blended score can hide one platform where you barely exist. Every individual response also shows its own score on the Responses page, so any average can be traced back to the exact answers behind it. See [Understanding responses](/docs/academy/understanding-responses). ### What moves it Because zeros count, coverage usually moves the score more than polish: getting into answers that currently skip you outweighs improving answers that already feature you. If your score is stuck at or near zero, work through [Why is my visibility zero?](/docs/academy/visibility-zero) first. For a score that dropped after being healthy, [Why did my score drop?](/docs/academy/why-did-my-score-drop) walks the likely causes, and the [competitor heatmap](/docs/academy/competitors-and-heatmap) shows the prompts where competitors appear and you don't.