Only some leavers answer the exit question: how to read the reasons without being misled

A one-click question when someone leaves tells you why, but the people who answer are not the people who leave. Newcomers close the window without a word while veterans love to explain. This entry shows how to weigh the answers so they match who actually left.

Why this works

The people who answer are not a fair picture of the people who leave. Each group is known by how many left, so each answer should speak for as many as its group lost. Scaling answers this way corrects who happened to reply, and admits groups no one spoke for.

When not to use it

Do not use it when the reply rate inside a group depends on the very reason for leaving, such as only angry people answering. Weights cannot fix that. Then treat the answers as examples to read, not as numbers, and talk to a few people.

In the full entry

• What to build, step by step
• Starting values and the reason for each
• 3 common mistakes and the fix
• 29 lines of example code with 3 checks that pass
• How your AI shows you that it works

Say this to your AI

Use the Spielerando entry "Only some leavers answer the exit question: how to read the reasons without being misled" (read_entry exit-survey-answers-reweighted) in my game. Build it step by step, and show me how you checked that it works.

Or skip this and connect the MCP: your AI then finds all of this by itself.Connect the MCP