Debrief — Post-Interview Analysis & Outcome Learning
ES Reference: Phase 2 (INTOS-ES-P2-001) Kit 2.0 · Phase 3 (INTOS-ES-P3-001) Interview-Day & Debrief, Sections 9-10
Overview
Debrief mode captures your real interview experiences — the questions you were actually asked, how you answered, what signals you noticed, and the outcome. Phase 3 expands debrief into a full outcome learning loop: debrief + practice data → updated question priorities and evidence gaps → adjusted readiness and Next Best Action → targeted preparation tasks.
Debrief data is private by default and excluded from aggregate learning without explicit consent. The system never claims a specific answer or resume caused an outcome — it identifies patterns without asserting causality.
Key Principles
- Private by default — debrief data is not visible to coaches, institutions, or aggregate analytics without consent
- Correlation, not causation — patterns are identified; the system never says "your STAR answer caused the rejection"
- Source and confidence recorded — every outcome and feedback item has a source and confidence level
- User-correctable and deletable — you can edit debriefs and trigger recalculation; deletion propagates
- No employer scouting — debrief data never flows back to employers or job platforms
Screens in This Feature Area
1. Debriefs List (/dashboard/debrief)
Chronological log of all your interview debriefs with outcome badges and linked jobs.
Buttons & Actions:
| Button | Behavior |
|---|---|
| New Debrief | Record a new interview debrief |
| Debrief Card (click) | Open debrief detail with analysis |
| Filter by Job | Filter debriefs by job target |
| Filter by Outcome | Filter by next round, rejected, offer, hold, unknown |
2. Record Debrief (/dashboard/debrief/new)
Capture a new interview experience immediately after it happens.
Key Fields:
| Field | Description |
|---|---|
| Job/Role | Associated job target (required) |
| Interview Round | Recruiter screen, technical, hiring manager, panel, final, etc. |
| Interview Date | When the interview occurred |
| Interviewer(s) | Names, titles, functions of interviewers (optional) |
| Format | Video call, phone, in-person, panel, assessment center |
Buttons & Actions:
| Button | Behavior |
|---|---|
| Select Job/Round | Link debrief to a job and round type |
| Record Questions | Document the actual questions asked |
| Link to Kit Questions | Connect actual questions to prepared kit questions |
| Link to Competency | Map questions to competencies |
| Rate Your Answer | Self-assess per answer (1-5 scale) |
| Add Signals | Record interviewer reactions, follow-up depth, body language cues |
| Add Notes | Free-form reflection and observations |
| Save Debrief | Persist the debrief record |
3. Record Actual Questions
Document the questions you were actually asked, linking them to your preparation.
For each question:
| Field | Description |
|---|---|
| Question Text | What was actually asked |
| Linked Kit Question | The closest prepared question (if any) |
| Competency | Which competency this tested |
| Your Answer Summary | Brief summary of how you responded |
| Self-Rating | How well you think you answered (1-5) |
| Was It Prepared? | Did you have a prepared answer for this? |
| Surprise Factor | Was this question unexpected? |
| Notes | Specific observations about this Q&A |
4. Record Outcome
Update the interview outcome as you learn it.
Outcome Options:
| Outcome | Definition |
|---|---|
| Advanced to Next Round | Moving forward in the process |
| Rejected | Not moving forward |
| Offer Received | Received a job offer |
| On Hold | Process paused (hiring freeze, restructure, etc.) |
| Withdrew | You chose to exit the process |
| Unknown | Haven't heard back yet |
Additional Fields:
- Outcome date
- Source (email, phone, portal, recruiter)
- Feedback received (verbatim or summarized)
- Feedback source confidence (direct/indirect/hearsay)
5. Debrief Detail (/dashboard/debrief/[id])
Complete debrief with AI analysis, pattern recognition, and learning actions.
Analysis Provided:
| Insight | Description |
|---|---|
| Question Pattern Analysis | Types of questions recurring across interviews |
| Competency Weak Spots | Competencies where self-ratings are consistently lower |
| Surprise Question Themes | Topics you were under-prepared for |
| Linked Answer Effectiveness | How well prepared answers mapped to actual questions |
| Outcome Pattern | Stage progression trends across jobs |
Buttons & Actions:
| Button | Behavior |
|---|---|
| Edit Debrief | Modify debrief content (audit trail preserved) |
| Update Outcome | Change outcome as new information arrives |
| View Analysis | See AI-generated insights and patterns |
| Create Practice from Gaps | Generate a focused practice session targeting weak areas |
| Create Task | Add a preparation task based on debrief insight |
| Export | Download debrief summary |
| Delete Debrief | Remove debrief (triggers outcome learning recalculation) |
6. Outcome Learning Loop (Background)
The system processes debrief data to personalize future preparation:
Practice + Debrief Data
→ Normalize by question / competency / round
→ Update user-specific question priorities and evidence gaps
→ Adjust Readiness Index components
→ Update Next Best Action recommendations
→ Create targeted preparation tasks / recommended practice
→ Preserve explanation of what changed and why
Privacy Controls:
- User-specific data is used first — aggregate patterns require explicit consent, anonymity thresholds, and separate governance
- Sparse-data recommendations include a "limited confidence" disclosure
- Users can correct or delete debriefs — recalculation propagates
- Deletion removes debrief data from all learning computations
Debrief Data Model
| Object | Content |
|---|---|
| InterviewDebrief | Job, round, date, interviewer info, format, notes |
| ActualQuestion | Question text, linked kit question, linked competency |
| AnswerSelfRating | Per-question rating, summary, prepared/unprepared flag |
| InterviewOutcome | Outcome type, date, source, confidence, feedback text |
| OutcomeLearningRun | Version, input snapshot (practice + debriefs), updated priorities, explanation |
UX Rules
- Debrief data is clearly labeled as private by default
- Aggregate/learning consent is separate and explicit — never bundled
- Outcome learning explanations show "what changed and why" — not "you failed because..."
- Sparse-data states are disclosed: "Limited data — confidence in this recommendation is low"
- Deletion confirmation clearly states what will be recalculated
Related ES Requirements
| ID | Requirement |
|---|---|
| P3 Debrief | Create: select Job/round/date, record actual questions, answer used, confidence, signals, notes |
| P3 Question Capture | Create/link actual questions to existing questions/competencies |
| P3 Outcome Update | Record next round, rejection, hold, offer or unknown with date/source |
| P3 Feedback | Store user-reported/recruiter feedback with source and confidence |
| P3 Privacy | Debrief is private and excluded from aggregate learning without consent |
| P3-E07 | Learning loop: update priorities/readiness from practice/outcomes |
| P3 Outcome Learning | Use user-specific data first; aggregate patterns require consent, anonymity, governance |
| P3 Outcome Learning | Do not state a specific answer or resume caused an outcome |
| AC-07 | Interview-day brief and actual-interview debrief are operational/mobile usable |
| AC-08 | Outcome learning updates user-specific priorities without causal/offer claims |
Related Docs
- [[jobs]] — Debriefs are linked to job targets and update application stage
- [[practice]] — Practice performance + debrief data combine in outcome learning loop
- [[readiness]] — Outcome learning updates readiness components and recommendations
- [[nba]] — Outcome learning triggers updated Next Best Action recommendations
- [[brief]] — Compare what you planned (brief) vs. what actually happened (debrief)