What this is: A listening layer. You tell the platform which places to watch, it collects what gets published there, and it summarises the interesting parts for you. Open it from Intelligence in the sidebar.
The page has three tabs: Sources, Signals and Digests. They are three steps of one pipeline: you configure sources, crawling turns them into signals, and a digest summarises the signals.
A source is one place to watch. Five kinds:
| Source type | What you enter |
|---|---|
| Reddit Subreddit | A subreddit, for example r/marketing. |
| LinkedIn Influencer | A LinkedIn profile or company handle. |
| PAA Seed Keyword | A search phrase whose "People also ask" questions you want tracked. |
| Competitor Blog | A competitor blog URL. |
| Brand Website | A website URL. |
Click Add source, pick the type, enter the identifier, and choose a Crawl schedule of Manual, Daily or Weekly. Each card then shows when it last ran, how many signals it has collected, a switch to pause it, and a Crawl now button for an immediate run.
Two faster ways to fill the list:
Gotcha: The CSV type column needs the raw values (reddit_subreddit,linkedin_influencer,paa_seed,competitor_blog,brand_website), not the friendly names shown elsewhere. The dialog also does not close itself after a successful import.
Gotcha: You can change a source's identifier, display name and schedule later, but not its type. To change the type, remove it and add it again.
Gotcha: Identifiers are tidied up when saved, so a full Reddit URL is stored as r/name. Adding the same place in a different spelling counts as a duplicate.
Every item a crawl collects lands here: the post, question or article, with its source, its date, a link to the original, and, where the AI could work it out, the topic, persona, funnel stage and intent.
Filter by signal type or by source using the two dropdowns. Signals are read-only, and there is nothing to approve here.
Gotcha: The two dropdowns search everything, but the Filter visible box only filters the 25 rows currently on screen. Narrow with the dropdowns first, then search.
A digest is a once-a-day summary of what the signals added up to: a headline, then items grouped as Idea, New gap, Trending, Competitor and PAA, each with a priority.
Each item offers:
Run digest now produces one immediately rather than waiting for the daily run.
Gotcha: You get one digest per day. If today's has already finished, Run digest now returns the existing one rather than building a fresh one.
Gotcha: Dismiss is final. There is no undo and no way to view dismissed items.
Prerequisite: Digests are built from signals collected in the last 24 hours. With no active sources, or none that have been crawled recently, a digest will arrive close to empty.
Start a 14-day free trial, or book a walkthrough.