Most organisations that run a 360 feedback round do it with the best intentions and the vaguest plan. Raters are nominated, questionnaires go out, a report lands on someone's desk, and then very little happens. The method itself is sound. The way it is usually executed is not.
This guide explains what 360 feedback is, how the process works step by step, what kinds of questions to ask, and, just as importantly, the conditions under which it helps development and the conditions under which it quietly damages trust.
What is 360 feedback?
360 feedback (also called 360-degree feedback or multi-rater feedback) is a structured process where an employee receives feedback from several perspectives at once: their manager, their peers, the people who report to them, and sometimes internal or external stakeholders. The employee also rates themselves, so the final report compares self-perception against how others experience them.
The name comes from the full-circle view: instead of one person's opinion (usually the manager's), you get a composite picture from the people who actually work with the person every day. That is the core promise. A manager sees perhaps a fraction of someone's working behaviour; colleagues and direct reports see the rest.
Two things distinguish a proper 360 from an informal round of opinions:
- Structure. Everyone answers the same questions on the same scale, usually with space for free-text comments.
- Purpose. A 360 is a development tool. It works best when the results feed a development conversation, not a salary review or a ranking exercise.
How the 360 feedback process works, step by step
A well-run 360 round follows a predictable arc. Skipping any of these steps is where most of the damage happens.
- Define the purpose and tell everyone. Decide whether this is for leadership development, team development or something else, and communicate it before a single question is sent. If people suspect the results will affect pay or promotion, they will answer politically, not honestly.
- Choose the competencies to assess. Five to eight behavioural areas is plenty: for example communication, collaboration, decision-making, delegation, and giving feedback. More than that and raters fatigue, which lowers the quality of every answer.
- Select raters carefully. A typical round includes the manager, three to five peers, three to five direct reports (where relevant) and a self-assessment. Raters need enough recent working contact to answer credibly. Letting the subject hand-pick only friendly raters undermines the whole exercise, so most organisations let the subject propose names and the manager or HR approve them.
- Guarantee anonymity for peers and direct reports. Responses should be aggregated so no individual answer can be traced, with a minimum group size before results are shown. Without that protection, upward feedback in particular becomes flattery. We have written more about why this matters in why anonymity is important in employee surveys; the same psychology applies with even more force when the subject is your own manager.
- Collect responses over a short, fixed window. One to two weeks is usually enough. A drawn-out collection period signals that the process is not a priority.
- Debrief the results in a conversation, never by email. The report should be walked through with a trained facilitator, HR partner or the manager, focusing on patterns rather than individual comments. The most useful signal in any 360 is the gap between self-rating and others' ratings, in both directions.
- Turn the report into two or three development actions. Not ten. A focused development plan with owners and dates is what separates a useful 360 from an expensive PDF. Our guide to development conversation questions covers how to structure that follow-up discussion.
- Follow up within three months. Check progress against the actions, and tell the raters, at least in general terms, that their input led somewhere. People who see their feedback acted on will give honest feedback again. People who do not, will not.
Sample 360 feedback questions
Good 360 questions describe observable behaviour, not personality. "Is a strategic thinker" invites guesswork; "explains how their team's priorities connect to company goals" invites an answer based on evidence. Most questions use an agreement scale, with free-text prompts at the end.
Communication
- Explains decisions and the reasoning behind them clearly.
- Listens to understand before responding.
- Keeps people informed about things that affect their work.
Collaboration
- Shares information and credit with colleagues.
- Is approachable when others need help or input.
- Handles disagreement constructively rather than avoiding it.
Leadership (for managers)
- Gives feedback that is specific and actionable.
- Delegates meaningful work, not just tasks.
- Creates an environment where people can raise concerns safely.
Free-text prompts
- What should this person keep doing?
- What is the one thing they could change that would have the biggest effect on how you work together?
If you are building a broader question set, the same design principles apply as for engagement surveys: ask about behaviour you can act on, and keep it short. Our breakdown of 42 employee engagement survey questions goes deeper into what makes a question predictive rather than merely interesting.
When 360 feedback helps
360 feedback earns its reputation under specific conditions:
- The purpose is development, and everyone believes it. When results stay with the individual and their development plan, raters answer honestly and subjects listen with curiosity instead of defensiveness.
- The subject has blind spots a single manager cannot see. This is the classic use case for leaders: a manager may perform well upwards while direct reports experience something quite different. Upward feedback is often the only structured channel for that signal.
- There is a real follow-up process. Feedback changes behaviour when it leads to conversations and commitments. The appetite for this is clearly there: 94% of employees prefer real-time feedback according to a Wakefield Research study, and 68% of employees who receive regular feedback say they are satisfied with their workplace.
- The organisation already has a baseline of trust. In teams with reasonable psychological safety, a 360 deepens honest dialogue. It cannot create that safety from nothing.
When 360 feedback backfires
The failure modes are just as predictable as the success conditions, and worth stating plainly.
- It is tied to pay or promotion. The moment ratings carry financial consequences, raters inflate scores for allies and the data stops describing reality. If you need an appraisal input, run that as a separate, transparent process.
- Anonymity is promised but not credible. In a team of three, "anonymous" upward feedback is nothing of the sort. Small rater groups either need to be merged into larger categories or excluded.
- The report is delivered without support. A dense report full of critical comments, read alone on a Friday afternoon, produces hurt and rationalisation rather than development. The debrief conversation is not optional.
- Nothing happens afterwards. This is the most common and most corrosive failure. Raters invested time and candour; silence tells them it was wasted. The patterns we describe in how to follow up on your team's feedback apply directly here: visible action is what keeps any feedback channel alive. The cost of ignoring this is real. In the research summarised in our development conversation guide, 24% of employees say they would leave an employer that did not give them sufficient feedback.
- It is run too often, or as a substitute for everyday feedback. A 360 is a periodic deep-dive, typically annual or tied to a development programme. Running it as the only feedback mechanism creates long silences punctuated by heavy reports.
A 360 report does not develop anyone. The conversation after it does.
360 rounds and continuous feedback: complements, not rivals
The honest critique of the classic 360 round is the same critique that applies to the annual engagement survey: it is a snapshot, delivered with a lag, at a moment chosen by the organisation rather than by events. Behaviour observed in March arrives as feedback in June.
Continuous listening closes that gap. Frequent pulse measurements track how teams experience leadership and collaboration week by week, and regular one-to-ones turn those signals into dialogue. The effect of rhythm is striking: a Forbes survey of 20,000 employees found that bi-weekly one-to-ones reduce burnout risk by 84%. Meanwhile only 51% of employees say their employer listens well, which suggests most organisations have more to gain from fixing everyday listening than from adding another annual ritual.
The practical division of labour looks like this:
- Continuous pulse and one-to-ones carry the everyday signal: engagement, workload, how leadership is landing, early warnings.
- A 360 round goes deep on one person's behaviour at a development milestone: a new leadership role, a leadership programme, a stalled career conversation.
If your continuous listening is working, the 360 will rarely surprise anyone. That is a sign of health, not redundancy.
What to do on Monday morning
- Write down the purpose in one sentence before designing anything. If you cannot say what decision or development the 360 feeds, do not run it yet.
- Decouple it from pay. Put it in writing and repeat it in every communication about the process.
- Limit the scope. Five to eight competencies, a handful of carefully chosen raters, a two-week collection window.
- Book the debrief and the 90-day follow-up before the survey opens. If those two meetings are not in the calendar, the round is not ready to launch.
- Check your everyday feedback rhythm first. If one-to-ones are sporadic and there is no continuous pulse on how teams are doing, strengthen that foundation. A 360 lands far better in an organisation where feedback is already normal.
Run this way, 360 feedback does what it was designed to do: it shows people the gap between how they think they show up and how they actually land, and gives them a structured, supported way to close it.