Rupert Picardo · Assessments · May 2026
A 360 degree feedback process collects structured feedback about a leader from the people who work directly with them — their manager, their peers, and their direct reports — to provide a multi-directional view of their working behaviour and leadership effectiveness. When designed and facilitated well, 360 feedback produces development insights that self-assessment alone cannot: blind spots are named, strengths are confirmed, and the gap between how someone believes they come across and how they actually come across becomes visible and workable. When designed poorly, it produces defensive responses, political gaming of the rater selection process, and relationships that are subtly more damaged after the process than before it.
The difference between these two outcomes is almost entirely a function of process design, not tool selection. Organisations that experience 360 feedback backfiring typically attribute it to the wrong cause — the platform, the timing, the participants' maturity, the organisation's readiness. In most cases, the cause is simpler: the design decisions that determine the outcome were not made explicitly, and so the defaults produced the wrong result.
The three ways 360 feedback fails — and why they're all design problems
Anonymity that isn't. 360 feedback is designed to be anonymous — raters should feel safe providing honest responses without worrying about the participant identifying them. But anonymity is a function of pool size, not of the anonymity declaration on the form. If a participant has two peers and three direct reports, a candid negative response from a direct report is identifiable by process of elimination. Any pool of fewer than four raters per category compromises anonymity. Most 360 processes set minimum pool sizes too low and then wonder why the data is uniformly positive.
Development versus evaluation — when the purposes are mixed. 360 feedback designed for development produces a report that the participant owns and uses to plan their growth. 360 feedback used to inform performance ratings or promotion decisions produces a report that the organisation owns and uses to make decisions. The rater behaviour in these two contexts is different: raters who believe their feedback will influence compensation produce more hedged, more positive responses. Mixing these purposes — using a "development-focused" 360 and then referencing it in the performance conversation — destroys the process. Raters learn quickly what happens with their responses and calibrate accordingly.
Feedback without facilitation. The most common failure mode in 360 feedback is also the simplest: the report lands, nobody explains how to read it, and the participant either becomes defensive or becomes overwhelmed. An average 360 report contains 40 to 60 data points across multiple competencies and rater groups. Interpreting it without facilitation support almost always produces the wrong conclusions. The pattern that matters is rarely the lowest score. It's usually the divergence between how different rater groups perceive the same behaviour — and that divergence requires a trained conversation to surface.
Before you launch: the design decisions that determine the outcome
The decision that matters most is one that seems obvious but is frequently skipped: development or evaluation? Decide this explicitly, document it, communicate it to participants and raters, and do not allow scope creep after the process begins. If the 360 is developmental, the participant owns the report and chooses who to share it with. Full stop.
The second decision is rater selection: does the participant choose their own raters, does the manager choose, or does HR? Each approach has different risks. Participant-selected raters tend toward people who will be positive. Manager-selected raters tend toward people the manager believes will produce useful data, which introduces its own bias. The most robust approach is collaborative: the participant proposes, the manager reviews, HR ensures adequate pool size across categories. This is slower. It is worth it.
Pool size minimums: four raters in each category (peers, direct reports) as a hard floor. Fewer than four in any category and the data in that category is unreliable and the anonymity claim is questionable. If the participant has only two direct reports, either expand the category definition or acknowledge that direct report data will be group-identified rather than anonymous.
Competency framework selection is the step most commonly done carelessly. Using a generic leadership competency framework for a first-time team leader produces a report that maps against behaviours they're not yet expected to demonstrate. The framework has to match the participant's level and context. A framework built for senior executives applied to a new manager is disorienting and largely unusable as a development tool.
Designing the questions — what most 360 tools get wrong
The quality of a 360 report is entirely dependent on the quality of the behavioural indicator statements. Vague indicators produce noise. Precise, observable indicators produce signal.
A vague indicator: "This leader communicates effectively with their team." A precise indicator: "When this leader makes a decision that affects the team, they explain the reasoning behind it before asking for input." Raters can give a meaningful response to the second. The first measures nothing useful.
The best single open-text question in any 360 instrument is: "What is one thing this leader should do more of, less of, or continue doing that would have the most impact on the team's performance?" This format, popularised by Marshall Goldsmith, consistently produces the most actionable qualitative data. It is specific in its ask and open in its answer. Most commercial 360 tools include this question or a variant. Make sure it's there.
Likert scale design: use a frequency scale ("never, rarely, sometimes, often, always") rather than an agreement scale ("strongly disagree to strongly agree"). Frequency scales are more observable and produce more honest responses. A rater who rates "often" on "this leader explains their reasoning" is making an observation. A rater who rates "agree" on "this leader communicates effectively" is making a judgement.
Facilitating the feedback conversation — the highest-leverage step
The facilitated feedback conversation — the session where the participant works through the report with a trained facilitator — is the step that determines whether the process produces development. It is also the step most frequently cut when budgets are tight. This is the wrong trade-off.
The facilitator's job in this conversation is not to interpret the data for the participant, and not to soften difficult feedback. It is to help the participant move from defensiveness to curiosity — to shift from "this isn't accurate" to "what might be producing this perception?" That shift, when it happens, is where the development begins.
The most common facilitator error is dwelling on negative scores. Low scores are attention-catching but are often less useful than divergence data. A score of 3.2 on stakeholder communication from peers but 4.6 from direct reports does not primarily mean "this leader communicates poorly with stakeholders." It means this leader's communication style produces very different experiences in different relationship contexts. That's a far more specific and actionable development insight — and it requires a facilitator to surface it.
Practically: budget ninety minutes for the initial feedback conversation. Sixty minutes is not enough. The first thirty minutes are almost always used managing the participant's initial response to the data. The actual development conversation happens in the second half.
The development plan that follows — and why most don't work
A 360 process that ends at the feedback conversation has produced insight but not development. Development requires a plan, a structure for follow-through, and a mechanism for accountability.
The most common problem with 360 development plans is that they are action-item lists rather than development commitments. An action item says "communicate reasoning before asking for input." A development commitment says "in every team meeting where I'm making a directional decision, I will share the business context before presenting the options — and I will ask [name] to give me feedback on whether I'm doing this by the end of Q2." The second is specific, observable, timebound, and accountability-supported.
The minimum viable follow-through structure is a thirty-day check-in between the participant and their manager. Not to review the report — the manager may or may not have seen it — but to create a regular touchpoint where the development commitment is in the room. Participants who have a scheduled check-in with their manager on their development goals are significantly more likely to maintain focus on them than participants who do not.
Rupert's Take
The test of a well-run 360 process is not the quality of the report. It is what the individual does with it six months later.
Most 360 processes fail this test not because the data was wrong, or the platform was poor, or the participant wasn't ready. They fail because nobody built the follow-through into the design. The facilitated conversation ends, the development plan is filed, and the next interaction the participant has with their 360 data is when the next cycle begins twelve months later.
The report is the beginning of the development work. The organisations that treat it as the end — that run the 360, produce the report, file it, and move on — have paid for the data collection but not for the development. The development is the facilitation, the plan, and the accountability structure. Without those, the 360 was an expensive survey.
Frequently Asked Questions
What is 360 degree feedback and how does it work?
360 degree feedback is a multi-rater assessment process in which structured feedback about a leader's behaviour is collected from their manager, peers, and direct reports — and sometimes from external stakeholders or clients. The responses are aggregated into a report that shows how the leader's behaviour is perceived across different rater groups. The report forms the basis of a facilitated development conversation and, ideally, a structured development plan. The "360" refers to the multi-directional view: feedback comes from all directions around the participant, not just from above.
Should 360 feedback be anonymous?
Yes, with a caveat: anonymity is only meaningful when the pool size is large enough that individual responses cannot be identified. The standard practice is to require a minimum of four raters per category — peers, direct reports — before that category's data is included in the report. If a participant has fewer than four direct reports, the data from that group should either be excluded or should be clearly marked as group-identified. Anonymity that cannot be maintained should not be promised — broken anonymity is one of the fastest ways to damage trust in the process.
Who should have access to 360 feedback results?
In a development-oriented 360 process, the participant owns the report. They decide who else, if anyone, sees it. The facilitator who runs the feedback conversation sees it. The HR sponsor sees aggregate data but not individual reports. The participant's manager does not automatically receive the report — if the participant chooses to share it with their manager, that is their decision. If the manager is to receive the report as part of an evaluation process, this must be communicated to raters in advance, and the process is no longer purely developmental.
How often should 360 degree feedback be run?
Most organisations run 360 feedback annually or bi-annually. Annual cycles are most common. Running more frequently than once per year rarely produces useful data — rater fatigue sets in, and behaviours observed over a short window are less reliable than patterns over a longer period. Running less frequently than every two years means the development insights age before they can be acted on at the next cycle. The exception is a targeted, single-competency pulse check (not a full 360) which can be run quarterly as a development tracking tool.
Can 360 feedback be used for performance appraisal?
Using 360 data in a performance appraisal creates a fundamental conflict of purpose. Raters who believe their responses will influence compensation produce more hedged responses. The development data degrades. The participant becomes defensive rather than curious. If the organisation wants to use multi-rater data for talent decisions, the instrument should be designed explicitly for that purpose, communicated as such, and separated completely from any development 360 process. Running a nominally developmental 360 that is then referenced in the appraisal conversation destroys both processes.
What is the difference between 360 feedback and a leadership competency assessment?
A 360 degree feedback process collects subjective perceptions of behaviour from multiple raters who know the participant. A leadership competency assessment is a self-administered psychometric or situational judgement instrument that maps the participant's own responses against a validated behavioural framework. The two tools produce different data and serve different purposes. A 360 tells you how your behaviour is perceived by others. A competency assessment tells you how your thinking and decision-making map against leadership competencies. Used together, they produce a richer development picture than either does alone.