Audit your prompt coverage across intent and funnel stages, then understand citation sources and quality mix.
Last updated
These views help you audit what you're monitoring and where AI platforms get their information — two complementary inputs to your visibility strategy.
The Prompt Coverage Matrix audits the composition of your own monitored prompt set so you can spot blind spots in your data-acquisition strategy. It maps every active prompt onto a 3 × 3 grid:
Each of the 9 cells shows the prompt count for that bucket, scaled by size (number of prompts) and fill intensity (share of your total). Empty cells render with a dashed outline so coverage gaps are obvious.
| Quadrant | Position | What it means |
|---|---|---|
| Discovery | Top-left | Top-of-funnel × Unbranded. Awareness-stage queries where users seek category answers without naming any brand. |
| Awareness | Top-right | Top-of-funnel × Branded. Users introducing themselves to your brand — "what is X?", "tell me about X". |
| Conversion Gap | Bottom-left | Bottom-of-funnel × Unbranded. High-intent buyers shopping the category without naming brands — a major opportunity surface. |
| Loyalty / Retention | Bottom-right | Bottom-of-funnel × Branded. Decision-stage queries from users who already know you — pricing, signup, support. |
The citation sources table shows the top domains that AI platforms reference when responding to your prompts.
Each source includes:
On the Ultra and Enterprise plans, the pages listed under a source, and every row of the Unique URLs list, also have Add to outreach. It puts the page's publisher on your Outreach board and looks for a contact the publisher has published. See Citation Outreach.
At the top of the table, chips show the percentage breakdown across source types. This tells you what kind of content AI platforms trust for your category.
| Source Type | Examples | What it means |
|---|---|---|
| Editorial | Forbes, TechCrunch, BBC, Reuters | AI trusts professional publications for your topic |
| UGC | Reddit, Quora, YouTube, Medium | Community content influences AI responses |
| Corporate | Company and brand websites, including your own domain | AI is citing vendor-owned pages for your topic |
| Competitor | Pages from your tracked competitor domains | AI prefers competitor content for your target queries |
| Other | .gov, .edu, Wikipedia, reference and utility sites | Institutional and reference sources carry weight |
| Other | Everything else | General web sources |
A pie chart showing the overall distribution of citation types across all your monitored prompts. Track this over time to see if your optimization efforts shift the mix toward more editorial or domain-specific citations.
The Cited passages panel sits under the citation sources table. It shows the part of a page an AI engine quoted when it answered your prompts, so you can see what won the citation and not only which page did.
It covers the engines that say which part of a page they used:
Other engines cite pages without saying which part, so they do not appear in this panel. Passages are collected going forward: dates before collection began show none.
Each row shows the passage, the page, the engine, how many of your prompts it was quoted for, how many answers quoted it in the selected dates, and when it was last quoted. A Gemini row also carries a format label read from the passage itself: Table, List, Contents or Text, plus Price when the passage states a price.
Viewers see the panel without the action buttons.
| Plan | Your pages | Pages cited instead of you |
|---|---|---|
| Free | Every passage | The page, the engine and the counts. The passage itself is locked: choose Reveal the passages that beat you to upgrade |
| Pro, Ultra, Enterprise | Every passage | Every passage |
Which engines appear depends on the engines your plan monitors. Gemini is on every plan. Google AI Overviews and Google AI Mode are on Ultra and Enterprise.
