The daily interview digest

Finds every session that completed today across your active studies and posts a short summary where the team already looks.

Workflow 1 of 12Research opsSeptember 30, 2026

Get the workflow

Enter your details and the prompt appears here, ready to copy.

Can AI summarise user interviews as they finish?

Yes, if it can reach your research platform. Connected through the Great Question MCP, Claude, ChatGPT, Cursor or Copilot can list the interview sessions that completed today, read each one, and write a one-line summary per participant for Slack. It only reads, so it is safe to run on a schedule, and every line points back to a session anyone can check.

Who runs it
Research ops, or whoever fields the “how's recruitment going” question.
Where you sign off · Read only
Nothing. Read only.
Schedule it · Daily
Yes, daily. It only reads, and it answers a question someone asks every day.

How to run it in Claude, ChatGPT, Cursor or Copilot

The prompt is the same in every tool. What changes is where you connect Great Question and where you paste it.

Run it in Claude

  1. 1

    Connect Great Question to Claude

    In Claude Desktop, open Settings → Connectors → Add custom connector. Name it Great Question, paste https://greatquestion.co/api/mcp/v1, then Save and Connect and sign in with your Great Question account. Restart Claude so the tools load.

  2. 2

    Connect the other end

    This workflow also needs Slack, if you want it to post the message itself connected in Claude, through that tool's own connector.

  3. 3

    Paste the prompt

    Start a new chat and paste the prompt. Claude asks before it uses each Great Question tool; allow the read tools, and read anything it proposes to write before you allow it.

Your admin has to switch the MCP on in Great Question first. The MCP guide covers every tool, the security model and a team rollout.

How to run it on a schedule

  1. 1

    Give the agent its own token

    A scheduled run has nobody there to sign in, so it uses an agent API token instead of OAuth. The token sees exactly what its user can see, and you can revoke it from your Great Question account.

  2. 2

    Run the prompt headless

    Claude Code runs a prompt without a chat window with claude -p "<the prompt>", and Codex with codex exec "<the prompt>". Either works once the Great Question MCP is added to it.

  3. 3

    Put it on a daily timer

    Schedule that command with whatever your team already uses: cron, a CI job, or your AI tool's own scheduled tasks. Check the first few runs by hand before you stop looking.

The tools it calls

Named so you can see exactly what the agent touches before you let it near anything.

  • list_interview_studies
  • list_interview_responses
  • get_interview_response

What we never automate

Four rules sit under every workflow here, and they matter more than the workflows.

  • Nothing reaches a participant without a person

    No workflow sends an invitation, an email or a reminder. The MCP can do those things. They are left out on purpose, because a scheduled job that contacts your customers is a different category of risk from one that reads a transcript.

  • Nothing launches a study without a person

    An agent can build the screener and draft the messages. Someone still presses go.

  • Nothing reaches a stakeholder unchecked

    The synthesis workflows hand you quotes with links attached so you can check them in one click. Use the click. A misattributed quote in a board deck costs more than the hour it saved.

  • Nothing sees more than the person who connected it

    Permissions follow the individual, so an agent connected by a PM reaches exactly what that PM reaches. Your admin decides whether it is available at all.

Questions

Can AI summarize user interview transcripts?

Yes. Once it can read the transcript, an AI tool can summarise it in a line or a page. The digest keeps it to one line per session on purpose: it tells the team what happened today, and anyone who needs more opens the session rather than trusting a longer summary.

How do I post the digest to Slack automatically?

Run the prompt on a timer from Claude Code or Codex with an agent token, and connect Slack in the same tool so it can post the message. Or keep it manual: the prompt returns one message you paste in yourself, which is how most teams start.

What if no sessions completed today?

The prompt tells the agent to say so in one line, so a quiet day still gets a message and nobody wonders whether the job broke.

Related reading

More workflows

See all 12