---
name: 360-feedback-design
description: >
  Design and run a personal 360 — selecting raters, drafting questions, running the
  process, synthesizing the report, and integrating the findings. Reference when
  approaching a major transition, periodically calibrating self-perception against
  peer perception, or auditing whether current behaviors match intentions. A well-run
  360 every 18–24 months is the single most useful feedback instrument for senior
  professionals.
metadata:
  author: nirav
  version: "1.0"
compatibility: Designed for Claude Code
allowed-tools: Read
---

# 360 Feedback Design — Surfacing Blind Spots with Discipline

Self-assessment is unreliable. Senior professionals have known blind spots: they overestimate strengths, underestimate weaknesses, and miss specific behaviors that others see consistently. A well-run 360 surfaces what self-assessment hides — but only when the process is designed with care. Done badly, 360s produce vague platitudes; done well, they produce diagnoses sharp enough to drive real behavior change.

## Key Concepts

### When to Run a 360

| Trigger | Cadence |
|---|---|
| Before a major transition | Once, 6–12 months pre-transition |
| Annual / 18-month rhythm | Every 18–24 months |
| After a difficult period | Once, post-stabilization |
| New senior role | 6 months into the role |
| Pre-promotion | 3–6 months before promotion case |

Don't run 360s more often than every 12–18 months — the data takes time to change and the raters get fatigued.

### Selecting Raters

A useful 360 has rater diversity:

| Rater Type | Count | Why Include |
|---|---|---|
| **Manager (current)** | 1–2 | They see your work most often |
| **Manager (prior)** | 1–2 | Perspective across time |
| **Direct reports** | 3–5 | They see your leadership behaviors |
| **Peers (internal)** | 3–5 | They see your collaboration |
| **Cross-functional colleagues** | 2–3 | They see your influence-without-authority |
| **External (advisors, board, partners)** | 1–3 | They see your external-facing behavior |

Total: 12–20 raters is a good range. Fewer and the data is too noisy; more and analysis becomes unwieldy.

### Designing the Questions

The questions should be specific and behaviorally observable. Avoid:

- "Are you a good leader?" → vague
- "Do you communicate well?" → vague
- "What are your strengths and weaknesses?" → too broad

Use instead:

- "Describe a recent meeting [Name] ran. What did they do well? What could have been better?"
- "When [Name] disagrees with someone, how do they handle it?"
- "What's one thing [Name] could change that would make them significantly more effective?"

The shift: from trait-rating to behavior-description. The data is richer; the patterns are clearer.

### Anonymity vs Attribution

Most 360s are anonymous to encourage honesty. Tradeoffs:

| Anonymous | Attributed |
|---|---|
| More honest in critical areas | More accountable in claims |
| Raters often more candid | Allows clarifying conversations |
| Some "anonymous" feedback is still identifiable | Cleaner conversation flow |

Default: anonymous, with a synthesis written by the facilitator. The synthesis can name patterns ("3 of 4 direct reports mention X") without naming individuals.

### Using a Facilitator

For senior 360s, a facilitator (coach, HR, external consultant) significantly improves the process:

- Collects responses with appropriate confidentiality
- Synthesizes the data into themes (not raw quotes)
- Holds the debrief conversation
- Helps design follow-up actions

DIY 360s can work for less senior or smaller-scope reviews; for senior 360s, the facilitator is worth the cost.

### Synthesizing the Report

A good 360 report has:

1. **Executive summary** — Top 3 strengths, top 3 development areas
2. **Theme-level analysis** — Each strength and development area with supporting quotes (anonymized)
3. **Contradiction surfacing** — Where raters disagreed (often the most useful section)
4. **Comparison to self-assessment** — Where self-perception matches and where it diverges
5. **Suggested actions** — 2–3 specific behaviors to work on

### Integrating the Findings

The hard part isn't running the 360; it's integrating the findings:

1. **Sit with the report** for 24–48 hours before reacting. Emotional response is normal; let it pass.
2. **Triangulate with a trusted advisor.** What feels wrong? What feels right?
3. **Pick 2–3 themes to act on.** Not 10; the surface area is too broad.
4. **Design specific behavior experiments.** "I'll start every meeting by stating the objective" not "I'll communicate better."
5. **Re-test in 12 months.** A follow-up 360 (potentially smaller) tests whether the behaviors changed.

### Common Failure Modes

| Failure | Looks Like | Fix |
|---|---|---|
| Vague questions | "Rate communication on 1–5" | Behavior-specific questions |
| Too few raters | 4–5 raters; data too noisy | 12–20 raters |
| No rater diversity | All direct reports, or all peers | Include manager, peers, reports, externals |
| DIY at senior level | No facilitator | Use a coach or external for senior 360s |
| Skipping integration | Read the report, file it, move on | Pick 2–3 themes; design behavior experiments |
| Defensive reaction | "They don't understand me" | Sit with it; triangulate; act |

## Self-Coaching Track

**For your situation (MD → biotech VC/operator):**

1. **Decide if now is the right time.** Pre-transition is one strong time; mid-transition is too noisy; post-transition is another. Choose deliberately.

2. **Pick a facilitator.** Even a peer-coach or trusted advisor who can collect responses anonymously and synthesize. The third-party role matters.

3. **Build the rater list.** 12–20 raters across manager, prior manager, direct reports, peers, cross-functional colleagues, externals. Diversity matters more than count.

4. **Draft 6–10 behavior-specific questions.** Cross-load with the question types above. Test the questions by imagining how a rater would actually answer.

5. **Send invitations with context.** Each rater should know: this is for development, anonymous, 15–20 minutes, deadline. Generic invitations get generic responses.

6. **Receive the synthesis. Sit with it.** Don't react immediately. 24–48 hours of distance changes how you read it.

7. **Pick 2–3 themes to work on.** Design specific behavior experiments. Calendar a 12-month re-test.

8. **Cross-load with `idp-and-okrs`** — themes become quarterly OKR work.

## Teach / Mentor-Others Track

**When coaching a junior or peer through 360 design:**

1. **The cadence is important.** Mentees sometimes want 360s annually; the data doesn't change that fast. Coach 18–24 month cadence.

2. **Rater diversity is high-leverage.** Mentees often pick raters who'll be kind. Coach the diversity criteria explicitly.

3. **Behavior-specific questions transform the data.** Show one good question and one bad question side by side.

4. **DIY isn't the move for senior 360s.** Coach mentees toward a facilitator at senior levels. The investment pays back.

5. **Coach the integration discipline.** Many mentees run the 360 and file the report. The integration is where the value lives.

6. **Coach against the defensive reaction.** Some mentees, faced with critical feedback, dismiss it. Reframe: feedback is data, not verdict; the question is what to do with it.

## When This Applies

- Approaching a major transition
- 18–24 month rhythm
- Pre-promotion
- After a difficult period
- Onboarding to a new senior role (with 6-month delay)

## Cross-Domain Connections

- **feedback-loops/performance-review-craft** — Formal review is the more-structured cousin
- **feedback-loops/advisor-board** — Advisor board members may be 360 raters
- **feedback-loops/reflection-and-journaling** — Reflection practice integrates 360 findings
- **trajectory-design/skill-gap-analysis** — 360 themes often surface as gaps
- **trajectory-design/idp-and-okrs** — Themes become OKR work
- **executive-presence** — 360 commonly surfaces presence-specific feedback