Visibility
The Visibility page (Insights → Visibility in the sidebar) is your central view of how your brand’s AI visibility is changing over time. It’s organised into eight tabs, each covering a different dimension of your performance. The page header has All Topics and All Groups pickers to narrow every tab to certain topics or monitor groups, a time window picker for the date range, and View by to change how finely the charts group time.

Dashboard Tabs
1. Overview Tab
Your top-level visibility health at a glance: eight KPI cards up top, followed by trend charts.
KPI Cards
Eight cards, each with a current value and a gauge:
| KPI | What It Measures |
|---|---|
| Visibility | Your overall visibility score, 0-100 (a score, not a percent) |
| Presence Rate | Share of responses that mention you at all |
| SOV | Share of Voice: your share of all mentions of you and your tracked competitors |
| Citation Rate | Share of responses that cite a page on your own domain |
| Sentiment | Average tone of your mentions, -100 to +100 |
| Avg Position | Where in the answer you’re typically mentioned, as a percentile (lower is earlier) |
| Site Health | Your site’s average AEO Audit score, from AEO Audit. Point-in-time, not tied to the date range |
| Hero Rate | Share of your mentions that land in the first third of the answer |
For how each one is calculated, see Scoring Basics.
Trend Charts
Four charts, in a 2x2 grid:
- Visibility Score Trend - your overall score over time
- Presence Rate - how often you’re mentioned at all, over time
- Sentiment Trend - average tone over time, across mentioned runs only
- Share of Voice (100% Stacked) - your share of mentions vs each tracked competitor, stacked over time
| Pattern | What It Means |
|---|---|
| Steady upward slope on Visibility | Content investment is working |
| Sudden drop | Investigate immediately - check Responses for the affected prompts |
| Flat despite publishing | New content isn’t earning AI citations yet. Allow 4-6 weeks |
| Gradual decline | Competitors are gaining ground. Check the Competitors tab |
2. Competitors Tab
Where you stand against each tracked competitor, three ways.
Share of Voice - Totals
A bar chart of each entity’s share of all mentions in the window - you and each tracked competitor, side by side. Hover a bar to see its mention count.
| Share of Voice | What It Means |
|---|---|
| Rising | You’re getting mentioned more relative to competitors |
| Falling | Competitors are taking more of the AI conversation |
| Flat | Holding position - not gaining or losing ground |
Win Rate vs Competitors (Head-to-Head)
For each tracked competitor, a count of responses in four columns - Both, Brand Only (Win), Competitor Only (Loss), and Neither - plus a Win Rate: of the responses that mention you or that competitor, the share that mention you. Click any count to open the responses behind it.
Use this to find which competitors you’re consistently losing to, then focus content on those specific comparison and alternative prompts.
Position Battle
Shown as Mention Order vs Competitors · Position Battle: Avg Position %, Best Position %, and Mentions for you and each tracked competitor. Lower percentile means earlier in the response, which reads as a stronger recommendation.
3. Answer Engines Tab
How your visibility breaks down across the AI platforms GrackerAI monitors: ChatGPT, Perplexity, Google AI Mode, Google AI Overview, Copilot, Grok, Gemini, DeepSeek, Claude, and Brave Leo. Only the platforms your monitor groups run on show data.
Performance by Answer Engine
A trend chart with one line per platform. Switch between Visibility and Sentiment.
By Answer Engine
A table with Visibility Score, Presence Rate, Sentiment, and Runs per platform. Click a Runs count to open the underlying responses.
| Gap Pattern | What It Usually Means |
|---|---|
| Strong Perplexity, weak ChatGPT | Mainstream content gap - get into broader publications and directories |
| Strong ChatGPT, weak Gemini | Community/video content gap - Reddit, YouTube, forums |
| Weak across all surfaces | Fundamental authority gap - start with foundational content |
| Weak on Copilot only | Enterprise content gap - case studies, compliance docs, analyst coverage |
Answer Engine x Entity Mentions
A matrix with one column per AI platform, showing mention counts for you and each competitor, with a total column - so you can see at a glance which platform is driving a competitor’s lead.
4. Demographics Tab
How your visibility breaks down by market: country and language.
Performance by Country
A trend chart with one line per country. Switch between Visibility and Sentiment.
By Country
Visibility Score, Presence Rate, Sentiment, and Runs per country your monitor groups target. Useful when you run monitors for multiple markets and want to see which geographies are driving score changes.
Language Coverage
One row per language and answer engine, with Presence Rate, Visibility Score, and Runs - so you can spot a language your monitoring hasn’t covered well yet.
5. Topics Tab
Share of Voice by Topic is a heatmap table: rows are your brand’s topics (plus “Uncategorized”), columns are you and your tracked competitors. Each cell shows that entity’s share of all mentions within that topic - each row adds up to roughly 100%. Hover a cell for its mention count.
Green means 70% or more of a topic’s mentions, amber 25-70%, red under 25%.
6. Prompts Tab
Tracks visibility at the individual prompt level, split into three lists.
Weakest Prompts
Your bottom 10 prompts by visibility score. Start here to find your biggest content gaps.
Strongest Prompts
Your top 10 prompts by score (where you have a score above zero) - the prompts already working, worth defending.
Lost Prompts
Prompts where a competitor is mentioned but your brand never was, over the selected window. A competitor has already earned AI’s trust for that query and you haven’t yet. These are your highest-value content opportunities - read the actual responses to see why.
Each list shows Prompt, Visibility Score, Presence Rate, Sentiment, Runs, and Last Seen (Lost Prompts show Runs, Avg Competitor Mentions, and Last Run instead). Click any row to open the prompt’s detail view.
7. Citations Tab
Tracks which websites AI trusts most in your category, four ways.
Competitor Citation Comparison
A line chart of citation counts over time, for you and each tracked competitor.
Citation Mix
A pie chart of citations in three buckets - your own domain, competitor domains, and neutral third-party sources. If your own domain is rarely cited, AI is referencing you but not linking to you as a source - a structural content problem needing clearer positioning and schema.
Citation Source Types
A breakdown by the type of site cited - for example News, Blog, Forum, Review Site, Documentation, Company Website, or Official Docs - so you can see whether your category leans toward community platforms, media, or vendor content. See Source Type in the glossary for the full list.
Citation Trend
A stacked area chart of brand, competitor, and neutral citations over time.
8. Sentiment & Position Tab
Tracks the tone of how AI describes your brand, and where you land in the response.
Sentiment Drift
Your average sentiment (-100 to +100) over time, shown alongside your top competitors. Watch for:
- Declining trend - AI is describing you less favourably. A visibility drop usually follows 4-8 weeks later.
- Sudden negative spike - Usually traces to a specific source: a new negative review, a competitor’s comparison page, or outdated content.
Sentiment Distribution
A histogram of your brand’s mentions in 20-point sentiment bands, from -100 to +100. Hover a bar to see whether the band counts as negative (below 0), neutral (0 to +20), or positive (above +20). A healthy distribution for most B2B brands is 60%+ of mentions positive, under 10% negative. If negative exceeds 20%, investigate the source.
Position Distribution
A table splitting your mentions, and all competitor mentions combined, into three tiers by where in the response they land: Hero (first third), Middle, and Tail (last third). Each cell shows the share of mentions in that tier and the count. A brand with most of its mentions in the Hero tier reads as the default recommendation; one stuck in the Tail reads as an afterthought, independent of sentiment.
Common Reporting Cadences
| Cadence | What to Review |
|---|---|
| Weekly | Overview trend, weakest and lost prompts |
| Monthly | Full dashboard, Sentiment & Position, Citations, new/lost domains |
| Quarterly | All tabs, Competitors tab win rates, Topics tab |
For report templates and export guidance, see Reporting & Exports.