Prompt research | Recited AI Docs

What it is

Answers > Prompt research is an explorer. Type a topic and Recited builds a research report: the questions people ask AI about it, how much demand each has, what the asker intends, and a live sample of how platforms answer today.

Use it before spending tracked-prompt slots. It answers "is this topic worth tracking?" without committing anything.

What a report contains

Related topics with AI demand and trend, so you can see which neighboring themes are growing.
Topic and prompt counts for the space you searched.
An intent breakdown: how many questions are research, comparison, or action oriented.
Brands mentioned: who AI already names when answering these questions.
Source domains: where the answers pull from.
An expandable Topics and Prompts table: the questions themselves, grouped, with per-prompt numbers.

Reports are saved. Each one keeps its own URL, so you can revisit it later or share it with a teammate without rerunning anything.

Research does not touch your tracked data

This is deliberate and worth being clear about: a research report writes only into itself.

It does not create scan runs.
It does not appear in Overview, Citations, or Market pages.
It does not move your visibility score or any other metric.

You can explore as freely as you like. Your dashboards describe only what you actually track.

From research to tracking

When a prompt in a report looks worth owning, press Track. The prompt joins your tracked set and is included from the next daily scan onward. It counts against your plan's prompt limit like any other prompt, and you assign it a topic like any other prompt.

The moment it is tracked, it behaves like every prompt on the Prompts page: daily answers, visibility, sentiment, position, and its own detail page.

When to reach for it

Before a content push. See what buyers actually ask about a theme before writing toward it.
When prompts underperform. If a tracked prompt has no demand, research the topic and swap in questions people really ask.
When entering a new market. Get the lay of the land: who AI names, what it cites, and how crowded the answers already are.