Focus groups with AI agents
Run AI-moderated focus group discussions with multiple synthetic personas.
Focus groups let you simulate multi-participant discussions using AI personas. Create a group of personas with different backgrounds and watch them discuss your research topic, generating diverse perspectives and debate.
Steps
- Open Focus GroupsClick Focus Groups in the main navigation sidebar, or access focus group features from within the survey editor.
- Step 1: Select participantsChoose 2-6 AI personas to participate. You can use existing personas from your Audiences library or create new ones. Each participant is color-coded for easy identification in the transcript.
- Step 2: Write a discussion guideAdd up to 5 discussion topics. For each topic, write a main question and optional probe questions (up to 5 probes per topic). Set a round limit per topic to control discussion depth.
- Step 3: Configure the sessionChoose the AI model, set a moderator style (neutral, challenging, or empathetic), pick a group dynamic (natural, debate, or consensus), and set the maximum number of rounds (5-50).
- Review the transcript and summaryWatch the discussion unfold in a color-coded, real-time transcript with topic dividers. When the session completes, an AI-generated summary highlights key themes, points of agreement and disagreement, notable quotes, and per-participant analysis.
Focus groups complement synthetic testing by generating qualitative discussion rather than individual survey responses.
Each persona brings a unique perspective shaped by their demographics, attitudes, and background. Participants speak in round-robin order, and the AI moderator steers the conversation based on your discussion guide.
After the session, an auto-generated summary includes 3-5 key themes, points of consensus and conflict, notable participant quotes, and a per-participant analysis showing key contributions and stance evolution.
Use focus groups early in research design to identify blind spots, test question wording, and discover topics you may have missed.