Survey & voice-of-customer analyzer
Reads hundreds of open-ended answers, builds the codebook, classifies topic and sentiment, cross-tabs by campus, pulls quotes, and delivers the report.

The problem
Open-ended survey questions are the most valuable and the least analyzed: reading and coding hundreds of comments by hand takes weeks, so they usually go unused.
What it delivers
- Codebook
- Quantified topics and sentiment
- Verbatim quotes
- Downloadable report
Agent tools
resumen_encuestaleer_respuestasbuscar_comentariosguardar_clasificaciontabla_temascitas_representativasexportar_informeEach tool is a real Python function the agent chooses to call. Every call is shown on screen, so the process can be audited.
How it works
Not a chat that only replies: it works in a loop, picks the right tool, checks what it gets, and delivers.
- 1Summarizes the survey: size, questions, and segments.
- 2Reads the open answers in batches.
- 3Builds a codebook from the emerging topics.
- 4Classifies topic and sentiment for each answer.
- 5Counts and cross-tabs topics by campus or segment.
- 6Pulls representative quotes and exports the report.
The agent in action
Real screenshots of the app. Click to enlarge.
Built with Anthropic Claude. Runs on your infrastructure, with a demo mode and a live reasoning mode.
More prototypes from this service
Ready to turn your data into smart decisions?
We build this kind of solution with your own data. Let’s start with a free 30-minute call.




