What this tab does
Most tools start with a question and ask whether your page answers it. Prompt Discovery works the other way: it starts with your page and shows a single ranked list of prompts, each tagged with what the page is doing on that prompt — already strong on it, close to winning, needing content work, or a gap to close.
The tab analyses the page’s actual content and structure, then surfaces the prompts a real user would type into ChatGPT, Claude, or Perplexity — not based on keywords or metadata, but based on what the text itself is saying and how it is organised.
Why it matters
Teams often assume a page is well positioned for certain questions because the topic is there somewhere. But “the topic is mentioned” is different from “the page answers this specific question.” A page that spends 80% of its content on broad context and 20% on the specific answer a user is looking for may surface as needing content work even though the information is technically present.
Knowing what the page already answers is half the picture. The other half is what’s missing — and the recommended action on each card tells you what to add or change to move a prompt up a tier.
How to read the results
Already strong — your page strongly answers this prompt. Protect and amplify what already works.
Close to winning — your page already answers most of this. The recommendation on the card would make it stronger.
Needs content work — your page touches this topic but doesn’t yet resolve it. A focused content addition would close the gap.
Gap — this is a realistic prompt people ask, but the page doesn’t answer it well yet. The recommended action names the section or content piece to add.
Warnings — signals that something about the page structure or content is reducing its clarity. Examples: content is spread too thinly across topics, the primary intent of the page is ambiguous, or the most useful section is buried too deep.
Limitations — Prompt Discovery is not a live ranking tracker, not search volume, and offers no guaranteed citation. The label reflects what the page is doing today, not predicted AI behavior.
How to use it
Already strong cards = validation. Confirm the page is doing its job for those prompts and protect what works.
Close to winning cards = the closest wins. Follow the recommended action on each card to move them into Already strong.
Needs content work cards = focused additions. The page touches the topic but doesn’t resolve it — the recommended action says what to add.
Gap cards = the next prompts to plan content for. Follow the recommended action, then validate with Prompt Fit.
Use it before testing Prompt Fit. Prompt Discovery tells you which prompts the page is positioned for and what state it’s in on each. Prompt Fit tests whether the page can answer one specific question well enough to appear in an AI response. Pick the prompts you care about most — typically from the Close to winning and Gap tiers — and run them through Prompt Fit to see the validation details.
What this tab is not
Prompt Discovery is not a live ranking tracker. It does not tell you whether your page is currently appearing in ChatGPT, Perplexity, or any other AI tool’s responses. The tier label reflects the state of the page on this prompt today, not how often the prompt is asked or how a specific AI model will rank it.
It is also not a keyword analysis. The results are based on the semantic meaning and structure of the page, not on individual terms or phrases.
An Already strong card in this tab still depends on the rest of the report being healthy. If the page has crawlability problems, content visibility gaps, or very slow TTFB, those are higher priority than refining content positioning.