Q&A clustering
Q&A clustering groups the questions your agent asked and shows how users responded. It runs automatically once your study reaches 100 conversations. Each cluster shows a question theme, how many times it was asked, and how users answered — broken down into percentage buckets. Look for questions with highly varied answers; these often reveal user segments with different needs. If new conversations come in after your last analysis, a banner appears letting you regenerate the clustering with the latest data.Processing takes roughly 0.17 seconds per conversation. A study with 500 conversations takes about a minute and a half.
Research Chat
The Research Chat panel is always visible on the right side of your study. Type a question or pick a quick prompt to get answers grounded in your actual conversations, with source references you can click to jump to the original transcript.Quick prompts
- Summarize key findings — Overview of what your conversations reveal
- Executive summary — High-level narrative for stakeholders
- Most requested features — What users want added or improved
- Conversations we should read — The most valuable interviews your team should review
- Most frequent issues — Recurring problems across conversations
Adding context
You can focus Research Chat on specific data:- From a conversation: Click Add to chat in the conversation header to load that interview as context
- From Q&A clustering: Click Add to chat on a question cluster to dig deeper into a specific pattern
Managing conversations
Research Chat keeps your conversation history. Use the tabs at the top of the chat panel to switch between threads, click + to start a new one, or use the history dropdown to browse and reopen past conversations.A small-sample banner appears when your study has fewer than 10 conversations with replies, so you can interpret answers with the right context.
Next steps
Broadcast more
Get more conversations to strengthen your analysis.
Study analysis concept
Learn how Q&A clustering and Research Chat work.