Debrief for {{Candidate name}} — {{Role title}}, held on {{Debrief date}}, facilitated by {{Facilitator name}}. The single rule that makes debriefs work: every interviewer submits scores and written evidence before reading anyone else's or hearing anyone speak. Discussion changes minds legitimately; hearing the room first changes scores illegitimately.

Part 1 — Independent scores (submitted before the meeting)

Part 2 — Facilitated discussion (30 minutes maximum)

The facilitator runs this order and enforces it:

1. Most junior speaks first. Seniority speaks last so the room does not anchor on the loudest title.
2. Evidence before opinions. Each interviewer gets two minutes: what they observed, then what they concluded. The facilitator interrupts conclusions offered without observations.
3. Disagreements get investigated, not averaged. When two interviewers scored the same competency differently, compare the underlying questions and answers — often one round simply got better data. Write down which.
4. Name the missing signal. List anything the loop failed to observe. Decide whether it is decision-blocking (add a targeted follow-up conversation) or acceptable risk (note it in onboarding).

Discussion notes

Part 3 — Decision

Rules of the room

• The hiring manager decides; the panel advises. Votes are input, not arithmetic — but a hiring manager overriding a unanimous panel must write down why, in this form.
• A strong yes with weak evidence loses to a moderate yes with strong evidence. Enthusiasm is not data.
• No new interviews by hallway: anyone who chatted with the candidate informally and formed a view either submits a written scorecard or stays silent.
• Feedback about protected characteristics, accent, appearance, or family situation stops the meeting for a reset — it never enters the record.
• The debrief ends with a decision or a named, dated blocking question. {{Let's sleep on it}} is a scheduling failure, not a decision.