> ## 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. # Position in answer > Position in answer records where your brand lands in the order of brands within an AI response, counted from 1, because the first name in an answer gets most of the attention. ### What is position in answer? Position in answer is the order in which your brand first appears among the brands in a single AI response, counted from 1. If an answer to "best CRM for startups" discusses HubSpot, then Pipedrive, then you, your position in that answer is 3. The metric exists because AI answers are read top-down, and most readers never make it past the first recommendation. Two brands can both be "mentioned" in the same answer while one is the headline pick and the other is the ninth row of a comparison table. Mention counts and Share of Voice treat those identically; position is the metric that doesn't. ### How position is measured During analysis, Promptwatch reads each response in natural order, top to bottom and left to right, across prose, headings, lists, and comparison tables. Every brand the answer substantively discusses gets a position based on its first qualifying appearance: * **First appearance decides.** If you are named in the opening sentence and again at the end, your position comes from the opening sentence. Repeats and later references don't change it. * **Positions count brands, not words.** Position 2 means one brand was discussed before you, not that you appeared as the second word. Positions run 1, 2, 3 through the number of brands in the answer, with no gaps. * **Only substantive appearances count.** Your name inside a cited URL or a source title does not establish a position, the same rule that governs what counts as a [mention](/docs/academy/citations-vs-mentions). * **Some responses have no position.** When an answer's structure makes the order genuinely impossible to determine, no position is recorded, and that response simply doesn't feed the average. For each day, Promptwatch averages your positions over the responses that mention you and have a recorded position, which is why the brand list shows values like "#2.3": on average, slightly worse than second brand named. ### Position vs. the neighboring metrics * **Visibility Score**: position is the starting point of the visibility score, which then adds share of attention and depth of coverage. An answer can name you first but spend three paragraphs on a competitor; position says #1, visibility says otherwise. Read position when you specifically care about ordering, visibility for the full picture. See [Visibility Score](/docs/academy/visibility-score). * **Citation rank**: position orders brands in the answer text; [citation rank](/docs/academy/citation-rank) orders sources in the citation list. An answer can name you first and cite you last. ### What moves it Position improves when models consider your brand the most relevant answer to the prompt, not just a valid one. In practice: * **Category association**: models lead with the brand they most strongly associate with the query. Broad third-party coverage that pairs your brand with the category ("Acme, the project management tool for agencies") pushes you up the order. * **Prompt specificity**: you often hold position 1 on niche prompts ("best CRM for dental practices") while sitting at 5 or 6 on the broad ones. Filtering by prompt or topic shows where your ordering is strong. * **Competitor moves**: your position can fall with nothing changing on your side, because a competitor became the new default first answer. Compare your average with theirs before assuming you did something wrong. ### Where to find it Open the Brand Visibility chart on your dashboard, a monitor page, or a prompt page, and switch to the position tab (the numbered-list icon). The chart ranks each day's brand averages against each other, so the lines move in #1, #2, #3 order rather than plotting decimals, and the brand list next to it shows the latest day's average, like "#1.8 avg. position". Lower is better, and #1 sits at the top of the chart, so a line moving up is good news. The Promptwatch Brand Visibility chart on the position tab, with average positions in the brand list. Slice by AI model before drawing conclusions: it is common to be the first answer on one platform and fifth on another, and the blended average shows you neither. For how position fits into the full metric set, see the [Metrics overview](/docs/academy/metrics-overview).