360-degree feedback: how to design and run it well

360-degree feedback is a structured process in which an employee receives feedback on their behaviour and skills from the people around them: their manager, peers, direct reports, sometimes internal or external customers, and themselves. The responses are collected on a common questionnaire, kept anonymous by rater group, and compiled into a report that shows how others see the person compared with how they see themselves. It works best as a development tool for managers and leaders, not as the main input to a performance rating.

This guide covers how to design and run a 360 programme. For the annual review cycle it often sits inside, see our performance appraisal process guide, and for rating distributions and calibration, see bell curve performance appraisal.

What is 360-degree feedback, and how is it different from an appraisal?

An appraisal answers "what did this person achieve against their goals, and what rating and increment follow?" A 360 answers "how does this person behave, as experienced by the people who work with them?" The first is about results, the second about how results are achieved.

Aspect	Performance appraisal	360-degree feedback
Main question	Did the employee deliver?	How does the employee work with others?
Inputs	Goals, KPIs, manager judgement	Ratings and comments from several rater groups
Who decides	Manager, reviewer, calibration committee	Nobody decides; the report informs
Identity of raters	Known	Anonymous by group, except the manager
Output	Rating, increment, bonus	Development report and action plan
Frequency	Annual or half-yearly	Annual, or every 12 to 18 months per person

Who should give feedback in a 360?

Rater group	Typical number	What they see best	Anonymous?
Self	1	Intentions, effort, context	Not applicable
Manager	1 (or 2 in a matrix)	Delivery, judgement, alignment with goals	No; reported separately
Peers	3 to 6	Collaboration, reliability, influence without authority	Yes
Direct reports	3 or more	Leadership, delegation, coaching, fairness	Yes
Internal customers or cross-functional partners	2 to 4	Responsiveness, service, problem solving	Yes
Skip-level manager	1	Potential, strategic contribution	Usually not anonymous

A participant usually nominates peers and partners, and the manager approves the list. This prevents a list made only of friends while keeping the participant involved. Aim for 8 to 12 raters in total; more adds survey fatigue without much more insight.

Anonymity thresholds: why "minimum three" matters

Honest feedback depends on raters trusting that they cannot be identified. Set a minimum number of responses per rater group below which that group's scores are not shown separately. Three is the common minimum; some companies use four or five for direct reports in small teams.

• If only two direct reports respond, their scores are merged into an "others" category or suppressed.
• Comments are shown verbatim but unattributed, and HR screens them for identifying details.
• The manager's feedback is shown separately and is not anonymous, because there is only one manager.
• Nobody, including HR business partners, sees individual responses linked to names. If the tool allows administrators to see them, say who and why in the programme notice.

Response data is personal data under the DPDP Act, so tell participants and raters what is collected, who sees the report and how long it is kept. See our DPDP Act guide for HR teams.

Sample 360 question bank

Use 25 to 40 rating questions across 5 to 7 competencies, plus 2 or 3 open-text questions. Use a 5-point frequency or agreement scale and a "not observed" option, so raters do not guess.

Competency	Sample statements (rated 1 to 5)
Communication	Explains decisions and the reasons behind them clearly. Listens without interrupting. Keeps others informed of changes that affect them.
Collaboration	Shares information and credit with other teams. Resolves disagreements constructively. Follows through on commitments made to colleagues.
Leading people	Gives regular, specific feedback. Delegates work with clear expectations. Treats team members fairly and consistently. Develops people for bigger roles.
Decision making	Makes timely decisions with the information available. Considers the impact on other teams. Takes ownership when a decision does not work out.
Customer focus	Understands what internal or external customers need. Responds within agreed timelines.
Integrity and values	Acts consistently with the company's code of conduct. Raises concerns openly.
Driving results	Sets clear priorities. Removes obstacles for the team. Holds self accountable for outcomes.

Open-text questions that produce useful comments:

1. What should this person continue doing?
2. What should this person start doing to be more effective?
3. What should this person stop doing?

Write statements about observable behaviour, not personality. "Gives specific feedback after client meetings" can be rated; "is a good leader" cannot.

Developmental vs evaluative use

This is the most important design decision, and it should be announced before the first survey goes out.

Use	How results are used	Benefit	Risk
Developmental only	Report goes to the participant and a coach or manager; feeds a development plan	Candid feedback; low gaming	Some managers ignore it
Evaluative	Scores feed the performance rating, increment or promotion	Behaviour counts visibly	Raters inflate scores for friends, deflate for rivals; peers trade favourable ratings
Hybrid	Development report, plus manager may consider themes (not scores) in the review	Balances candour and accountability	Needs clear rules and trust

Most companies get better results by starting with developmental use for two cycles, then deciding whether to bring behavioural themes into the appraisal. If 360 scores feed pay, raters know it, and the data changes. For how ratings connect to increments, see the merit increase matrix.

How to roll out a 360 programme

1. Define the purpose and audience. Start with people managers and leaders, typically the top two or three layers.
2. Build the competency model and questionnaire. Reuse your leadership competencies if you have them.
3. Set rules: rater groups, minimums, anonymity threshold, timelines, who sees the report.
4. Communicate. Explain the purpose, anonymity and use of results to participants and raters.
5. Nominate and approve raters, with a cap on how many 360s any one person rates in a cycle, often 5 to 6.
6. Collect responses over two to three weeks, with reminders.
7. Generate reports and suppress groups below the threshold.
8. Debrief. Each participant reviews the report with a trained coach, HR partner or manager.
9. Create a development plan with 2 or 3 focus areas, linked to courses, projects or coaching.
10. Repeat after 12 to 18 months to measure change.

For a 300-manager programme, a rough HR effort budget is useful. If each debrief takes 1.5 hours of an HR partner's time and report preparation takes 30 minutes, that is 600 hours. At an internal cost of ₹1,500 an hour, the programme costs about ₹9,00,000 in HR time before any external coaching, which is why many companies phase it by level.

Sample 360 report structure

A clear report is short, visual and points to action.

1. Introduction: purpose, rater groups and response counts.
2. Summary by competency: self score vs average of others.
3. Scores by rater group: manager, peers, direct reports, others, with groups below threshold suppressed.
4. Highest and lowest rated statements.
5. Hidden strengths and blind spots: where others rate higher or lower than self.
6. Verbatim comments, grouped by continue, start and stop.
7. Reflection questions and development plan template.

Worked example: reading the gaps

Vikram, a regional sales manager, receives feedback from his manager, 5 peers and 6 direct reports.

Competency	Self	Manager	Peers	Direct reports	Gap (others avg − self)
Driving results	4.2	4.5	4.3	4.4	+0.2
Collaboration	4.0	3.8	3.1	3.9	−0.4
Leading people	4.4	3.9	3.7	3.2	−0.8
Communication	3.6	4.0	3.9	4.1	+0.4

"Others average" is the simple average of the three rater-group scores. Leading people is a blind spot: Vikram rates himself highest there, and his direct reports rate it lowest of all. Communication is a hidden strength. The development plan focuses on delegation and regular one-to-ones, and the next 360 checks whether direct report scores move.

Common mistakes

• Using 360 scores directly in pay decisions in the first cycle.
• Showing scores for rater groups of one or two, which destroys anonymity.
• Asking 80 questions, so raters click through without thinking.
• Handing reports to participants with no debrief or plan.
• Letting participants pick only friendly raters, with no manager approval.
• Rating personality traits instead of observable behaviour.
• Running a 360 during a restructuring or layoffs, when trust is low.

FAQs

How many raters are needed for 360-degree feedback? Typically 8 to 12 in total, with at least 3 in each anonymous group such as peers and direct reports.

Is 360-degree feedback anonymous? Peer, direct report and other groups should be anonymous, with results shown only when a group meets the minimum number of responses. The manager's feedback is usually identified.

Should 360 feedback be linked to appraisal ratings? Most organisations start with developmental use only. Linking scores to pay tends to make raters less candid.

How often should 360 feedback be done? Once every 12 to 18 months per person is common, which leaves enough time to act on the feedback and show change.

Can 360 feedback be used for all employees? It can, but it is most valuable for people managers and leaders, where behaviour affects many others. For individual contributors, a lighter peer feedback step in the appraisal often works.

What is a good 360 feedback response rate? Aim for most nominated raters to respond. Clear communication, a short questionnaire and limiting how many 360s each person rates help most.

How NeevHR helps

NeevHR performance runs appraisal cycles with 360 feedback built in: nominate peer, manager and report reviewers, then calibrate talent on a 9-box grid alongside weighted goals and rating distribution. Learning holds the courses that development plans point to, succession uses the same calibrated 9-box for high-potential identification, and surveys with an anonymity threshold cover wider employee listening. Read more on the nine-box grid and succession planning, or book a demo to see a review cycle.

This article provides general information for educational purposes. Last reviewed: 17 Oct 2026.