Skip to main content

​
Overview

This glossary provides definitions for all metrics available in MentionLab. Understanding these metrics is essential for interpreting your brand’s visibility and performance across AI platforms.
Metrics are organized into six categories: Project Parameters, Mention Metrics, Source Metrics, Sentiment Metrics, Shopping Metrics, and Fan-Out Metrics.

​
Project Parameters

These metrics describe the structure and scope of your monitoring project.
Type: NumberThe total number of questions configured in your study. Each query represents a prompt that will be sent to AI platforms to monitor how they respond and mention brands.
More queries provide broader coverage but also consume more of your monthly quota.
Type: NumberThe number of tags used to categorize and organize queries in your study. Tags help you segment and analyze results by topic, intent, or any custom grouping.
Type: NumberThe total count of brands being tracked in your study. This includes your primary brand and any competitors you’re monitoring.
Type: NumberThe number of distinct query templates in your project. When running queries in multiple languages, this represents the unique questions before translation.
If you have 10 queries running in 3 languages, you have 10 unique queries but 30 total queries.
Type: NumberThe cumulative number of query executions, calculated as:Total Queries=Number of Queries×Number of Runs\text{Total Queries} = \text{Number of Queries} \times \text{Number of Runs}This represents your total API usage over the measurement period.
Type: NumberThe number of distinct sources that AI platforms referenced when generating responses to your queries. This indicates the breadth of information sources influencing AI responses about your brand.

​
Mention Metrics

These metrics measure how often and where your brand appears in AI-generated responses.
Type: NumberThe count of times a brand is mentioned in an AI response. A single response can contain multiple mentions if the brand name appears several times.
A mention is counted each time the brand name (or its aliases) appears in the response text.
Type: NumberThe average position where your brand appears in AI responses. A lower number indicates your brand is mentioned earlier, typically suggesting higher prominence.
Brands mentioned in the first position (rank 1) are often perceived as the primary recommendation.
RankingInterpretation
1-2Top recommendation
3-5Strong presence
6-10Moderate visibility
10+Low prominence
Type: PercentageThe proportion of all brand mentions that belong to your brand, calculated as:Share of Voice=Number of MentionsTotal Brand Mentions×100\text{Share of Voice} = \frac{\text{Number of Mentions}}{\text{Total Brand Mentions}} \times 100
Share of Voice helps you understand your competitive position relative to other brands mentioned in AI responses.
Type: PercentageThe percentage of AI responses where your brand was mentioned at least once, calculated as:Brand Visibility=Responses Containing BrandTotal Responses×100\text{Brand Visibility} = \frac{\text{Responses Containing Brand}}{\text{Total Responses}} \times 100
A brand visibility of 75% means your brand appeared in 75 out of every 100 AI responses.
Type: NumberThe average number of times your brand name appears within a single AI response when it is mentioned. A higher value indicates stronger prominence.

​
Source Metrics

These metrics analyze the sources (websites, articles, documents) that AI platforms use to generate their responses.
Type: PercentageThe percentage of AI responses that include at least one source citation, calculated as:Citation Rate=Responses With CitationsTotal Responses×100\text{Citation Rate} = \frac{\text{Responses With Citations}}{\text{Total Responses}} \times 100
Citation rate measures how often AI platforms back up their responses with references. A higher citation rate means the platform is more likely to cite sources when discussing your queries.
Type: NumberThe total number of times a website or source was referenced by AI platforms when constructing their responses. Citations indicate which sources are influencing AI-generated content.
Type: NumberThe number of citations from sources where your tracked brand is mentioned. This measures how often AI platforms cite content that includes your brand.
High brand citations indicate your content is being used as a reference by AI platforms.
Type: PercentageThe proportion of total citations that come from a specific source, calculated as:Share of Citations=Site CitationsTotal Citations×100\text{Share of Citations} = \frac{\text{Site Citations}}{\text{Total Citations}} \times 100
Type: PercentageThe proportion of brand-related citations attributed to a specific source, calculated as:Share of Brand Citations=Site Brand CitationsTotal Brand Citations×100\text{Share of Brand Citations} = \frac{\text{Site Brand Citations}}{\text{Total Brand Citations}} \times 100
Type: NumberThe count of unique pages (URLs) that AI platforms cited across all responses. This measures the diversity of sources being referenced.
Type: NumberThe number of unique cited pages that contain mentions of your tracked brand. These are sources that both:
  1. Were cited by AI platforms
  2. Include your brand in their content
Type: PercentageThe ratio of brand pages to total pages, indicating how much of the cited content includes your brand:Presence=Brand PagesPages×100\text{Presence} = \frac{\text{Brand Pages}}{\text{Pages}} \times 100
Type: PercentageThe proportion of a site’s total citations that include your brand, calculated as:True Reach=Site Brand CitationsSite Total Citations×100\text{True Reach} = \frac{\text{Site Brand Citations}}{\text{Site Total Citations}} \times 100
True Reach is only valid for entire domains, not individual URLs. It measures how well your brand is represented across a site’s cited content.
A True Reach of 40% means that 40% of the times a site is cited by AI platforms, the cited content mentions your brand.

​
Sentiment Metrics

These metrics analyze how AI platforms describe your brand — positively, neutrally, or negatively.
Type: Number (0–100)A numeric score representing the overall sentiment of AI responses mentioning your brand. The score maps to three categories:
Score RangeCategory
0–33Negative
34–66Neutral
67–100Positive
Type: PercentagesThe distribution of AI responses across positive, neutral, and negative sentiment categories, calculated as:Category %=Responses in CategoryTotal Mention Count×100\text{Category \%} = \frac{\text{Responses in Category}}{\text{Total Mention Count}} \times 100
Type: ComparisonA per-entity comparison of sentiment scores, letting you see how AI platforms describe your brand versus competitors.

​
Shopping Metrics

These metrics track your brand’s presence in AI-generated shopping and product recommendations.
Type: PercentageThe percentage of AI responses that include shopping product recommendations, calculated as:Shopping Coverage=Responses With ProductsTotal Responses×100\text{Shopping Coverage} = \frac{\text{Responses With Products}}{\text{Total Responses}} \times 100
Type: NumberThe total number of distinct brands detected in shopping results across all AI responses.
Type: NumberThe total number of distinct products detected in shopping results across all AI responses.
Type: NumberThe total number of distinct merchants (retailers) detected in shopping results across all AI responses.
Type: PercentagesThe proportion of product recommendations attributed to each merchant, showing which retailers AI platforms favor.

​
Fan-Out Metrics

These metrics measure the follow-up questions that AI platforms suggest after responding to your queries.
Type: NumberThe total count of all follow-up queries suggested by AI platforms across all responses.
Type: NumberThe deduplicated count of distinct follow-up queries. Multiple AI responses may suggest the same follow-up question.
Type: NumberThe average number of follow-up queries suggested per AI response, calculated as:Avg Per Response=Total Fan-OutsDistinct AI Responses\text{Avg Per Response} = \frac{\text{Total Fan-Outs}}{\text{Distinct AI Responses}}
Type: PercentageHow often a specific follow-up query is suggested, relative to total AI responses. Helps identify the most common follow-up topics.