Consultation Response Analysis Agent · Use case
How can a government body use an AI agent to analyse responses to a public consultation on its own infrastructure?
A consultation response analysis agent reads every response to a public consultation, groups them into themes, counts each theme, and quotes the responses behind it so the grouping can be checked. Policy staff check the themes against the responses and write the conclusions. The agent recommends no policy.
Ten thousand responses.
Six weeks to report.
- ✗ 01Reading and coding thousands of free-text responses takes a team weeks.
- ✗ 02A summary nobody can trace back to the responses will not survive a challenge.
Where it stops for a person
Before any conclusion. Policy staff check the themes and write the report.Before any conclusion is drawn. The themes are a proposal for policy staff to check.
It groups and quotes.
Policy staff conclude.
Step · 01
The policy officer uploads the responses or points the agent at the table that holds them.
Step · 02
The agent reads every response to each question.
Step · 03
It proposes themes for each question and assigns responses to them.
Step · 04
It counts each theme and quotes a sample of the responses behind it, with their reference numbers.
Step · 05
Policy staff merge, split or rename themes, check the samples, and write the conclusions.
Reads from · connectors
The consultation responses. Personal details in a response are redacted in what the agent returns where you set that.
Acts through · MCP servers
Nothing. It publishes nothing.
What you set up
- Load the responses as files or connect the table read only.
- Build the agent in Katonic Studio and give it the consultation's questions.
- Check its themes on a sample your team has coded by hand.
- Keep the run record with the consultation file.
Where it runs
The agent, its knowledge store and the model run on infrastructure you own or control. You choose the model and serve it yourself, so questions, documents and answers are not sent to a public AI service. Responses contain personal data and stay inside the body.
How you get it today
This page is labelled Use case: the agent is built in Katonic Studio. You describe the job, connect what is named above, add test questions and release the agent once it passes them. When the Catalog holds a tested template for this job, this page will carry the Template label and show how to take it.
Can we see why a response was put under a theme?
Yes. Every theme comes with the responses behind it, by reference number, so staff can check and move them.
Does it weigh responses or spot campaigns?
It counts and it can list responses with identical wording. What weight to give them is for policy staff.
Does it write the government's response?
No. It prepares the analysis. The conclusions and the response are written by people.