360 feedback: who answers, who sees the result and what you do with it
Echipa HR 365 · reviewed 2026-09-12 · 5 min read
A 360 works when it answers a development question and fails when it is used for a decision. The difference shows immediately in the quality of the answers: when people suspect their opinion affects someone’s salary or job, those answers become, predictably, gentle and useless.
What it is for, and what it is not
| Use it for | Do not use it for |
|---|---|
| Developing someone who wants to know how they come across | Promotion or pay decisions |
| Managers who receive feedback from nowhere else | Resolving a conflict between two people |
| Checking a specific blind spot | Building a file on someone |
| A new role, at 6 months, as a baseline | Performance appraisal, as a substitute |
Who answers and how many
Between six and eight people, chosen together with the person being reviewed, from three directions: whoever is above them, whoever works alongside them, and — if they have any — whoever reports to them. Below six, anonymity disappears; above eight, the effort grows without any gain in information.
- The person proposes the list; the manager adds one or two names if an important perspective is missing.
- Choose people they have actually worked with in the last six months, not people who like them.
- Avoid groups where the answer could be attributed: if they have a single direct report, that perspective is not reported separately.
- The same questions for everyone; otherwise the answers cannot be compared.
The first point looks like a weakness — the person picks who assesses them — and it is not: the purpose is their development, so they have every reason to ask the people who see them work. The manager’s addition covers the risk of a list that is too comfortable.
What you ask
Few questions, and open ones. A questionnaire with twenty scales produces averages nobody can use; three open questions produce concrete examples, which are the only thing anyone can learn from.
| The question | What it produces |
|---|---|
| What do they do well and should keep doing? Give an example. | Confirms strengths; it is also what makes the rest bearable |
| What would make working with them easier? Give an example. | “Easier” surfaces observations that “what do they do badly” never would |
| What does nobody tell them that they should know? | The blind spot; the highest-yield question of the three |
How the result is handed back
The order that matters
- HR or facilitator — aggregates the answers by theme, without attribution; removes phrasing that would identify the author
- HR and the person — discuss the result in private, before anyone else sees it
- The person — chooses what to share with their manager and the team — it is not automatic
- The person — formulates one or two actions; without them the exercise stays a collection of opinions
The third step is the rule that makes a 360 sustainable. From the moment the result goes automatically to the manager, the person can no longer receive difficult feedback without consequences — and will choose gentler reviewers next round.
Anonymity and its limits
In a small team, anonymity is a promise that is hard to keep: “someone said they interrupt in meetings” identifies the author if the team has four people. The honest thing is to say so upfront and let each person decide how much to write.
The alternative, in very small teams, is open, owned feedback delivered directly — harder, but less hypocritical than an anonymity everyone decodes in five minutes.
What if someone refuses to be a reviewer?
Accept it without discussion and replace them. A reviewer who fills it in out of obligation produces polite answers, which is exactly what you do not need.
How often?
Once a year at most. More often and nothing has time to change between rounds, while people tire of filling it in — which shows immediately in the quality of the answers.
For everyone or only for managers?
Start with managers: they receive the least feedback and have the greatest effect on everyone else. Extending it to everyone can come later, if the exercise proves useful.
Where to start
Decide first what it is for, and write it down. If the answer contains the word “appraisal”, stop — you need a different instrument, and a 360 built that way will cause damage.
Then run it for a single manager who wants it. One example that went well convinces more than any explanation of methodology.
The four administrative mistakes
| The mistake | The effect |
|---|---|
| Launching it for the whole company at once | A volume nobody can process, delayed results, the exercise dies in round one |
| Sending the result automatically to the manager | Reviewers turn cautious, and by round two the answers are useless |
| Using numeric averages on scales | Nobody can act on “3.4 for communication” |
| No action follows | The person is left with a list of opinions and no next step |
The first is the most common in companies starting out: launching once for everyone looks efficient, and the result is a volume of answers nobody processes in time. Debriefs slip by months, and by the second round nobody fills anything in.
360 feedback cycles with aggregated answers and an anonymity threshold
Answers are collected against the same questions and aggregated automatically, and the result reaches the person first — so the handback rule does not depend on who administers the process.
You can create an account in a few minutes and use every module for 7 days, no card required.