Use case
Audience research a strategist can defend
Clustered verbatims plus GWI panel data on one surface, so the audience portrait in your brief is built from evidence instead of instinct.
The short answer
Waldo gives audience researchers two layers on one surface: clustered verbatims that show how an audience actually talks, and GWI panel data that shows who they are and what they believe. Both return through the same API and MCP server, so the portrait you build for a brief is queryable, sourced, and repeatable.
The job to be done
A strategist building an audience portrait works from fragments: panel data in one tool, social chatter in another, a subscription that expired, and a folder of screenshots from the last person who held the account. Stitching those into slides takes days, and the result is frozen. Ask a follow-up question in the readout and the answer is another week of digging.
The job is a defensible portrait: who the audience is, what they believe, how they actually talk, and where the conversation happens. Defensible means every claim traces to a source. Useful means the portrait can be re-run next quarter as a script instead of rebuilt from scratch as a project.
How Waldo does it
Start with the panel view
GWI audience panels supply the compositional read: demographics, attitudes, media habits. That is the skeleton of the portrait, available alongside everything else instead of behind a separate login.
Read the conversation itself
GET /audiences/{id}/conversations returns what the audience says unprompted, clustered into themes with verbatims attached. This is where the portrait picks up the language real people use, not the language surveys hand them.
Chase the specifics live
Discover endpoints query the platforms at request time, so the subreddit or hashtag community the brief hinges on gets a fresh pull even if nobody thought to track it in advance.
Assemble with receipts
Every output links back to its sources. Quotes drop into the brief with links, the skeptic in the readout can click through, and the whole portrait re-runs next quarter with one command.
The endpoints
/audiences/{id}/conversationsHow the audience talks, clustered with quotes attachedMCP: audience_insightsPanel and conversation reads inside a working chat sessionMCP: category_trendsWhat is rising in the spaces this audience lives inDiscover endpointsLive platform queries for communities not yet trackedQuestions teams ask
Where does the panel data come from?
From GWI, Waldo’s audience panel partner. Panel composition and attitudes sit alongside live conversation data on the same API, so the who and the what come back from one place.
Do I get raw quotes or analysis?
Your choice per call. Raw source material runs about one credit per twenty results; scored analysis runs about five credits. Most researchers pull analysis first to find the themes, then raw verbatims to quote them.
Can I research an audience Waldo does not track yet?
Yes. Discover endpoints hit the platforms live at request time. If you want the audience refreshed daily from then on, activate it as a tracked entity; the annual activation fee applies only to untracked entities.
Keep reading
Put Waldo behind your agents
Brand, category, and audience intelligence over 200+ API and MCP endpoints. Sign up, mint a key, and run it against the brands you actually track.