A 360-degree feedback report brings together ratings from the participant and from people who see their behaviour at work. These may include a manager, peers, direct reports and other colleagues or customers. The value comes from comparing perspectives rather than searching for one definitive score.

Differences are normal. People observe the participant in different situations, have different expectations and are affected by different aspects of the participant's behaviour. The task is to understand what a difference might mean and what evidence would help explain it.

Start with the pattern, not a single gap

A large self-observer difference may attract attention, but it should not be interpreted on its own. First look across the whole report:

• Is the difference repeated across several related competencies?
• Do most respondent groups see the behaviour similarly?
• Is one group noticeably more positive or critical than the others?
• Are the ratings tightly grouped, or is there substantial variation within a group?
• Does written feedback support or complicate the numerical pattern?

A consistent pattern across related areas usually deserves more attention than one isolated difference. Comments and workplace examples can then help test possible explanations.

Why self-ratings and observer ratings differ

Participants rate themselves using their intentions, effort and knowledge of situations that other people may never see. Observers rate what is visible to them and the effect it has. A manager may know that they are trying to consult colleagues more often, while colleagues may not yet experience a noticeable change.

The reverse can also happen. A participant may underestimate a capability that others value because it feels ordinary or effortless to them. A lower self-rating alongside stronger observer ratings may indicate an under-recognised strength rather than a problem.

Why respondent groups may see things differently

Managers, peers and direct reports often interact with the participant in different ways. A manager may focus on delivery and strategic judgement. Peers may be more affected by collaboration and information sharing. Direct reports may see delegation, availability and day-to-day leadership.

Different group scores can therefore reveal that behaviour changes with the audience or situation. They can also point to competing expectations. Being decisive may be valued by one group while another wants greater consultation. Neither perspective should be dismissed; the development question is how to balance the demands more effectively.

Variation within a group is useful information

An average can hide sharply different experiences. If respondents within the same group vary widely, consider whether they work with the participant in different contexts, have had different levels of contact or are responding to inconsistent behaviour.

Confidentiality must be protected. The purpose is not to identify individual respondents or challenge their ratings. The report should be used to explore patterns and seek examples through normal feedback and development conversations.

Turn differences into constructive questions

Useful feedback discussions remain curious and specific. Questions might include:

• When is this behaviour most and least visible?
• What impact might colleagues experience even if the intention is positive?
• Which examples support the more favourable and less favourable views?
• Is the participant applying the capability consistently across groups?
• What small change would make the desired behaviour easier to notice?

This approach reduces defensiveness because it moves the conversation away from proving a score and towards understanding behaviour and impact.

Choose one or two development priorities

A 360-degree report often contains more information than can be acted on at once. Priorities should reflect the importance of the capability to the role, the consistency of the evidence and the participant's readiness to work on it.

A useful action is observable. Rather than “communicate better”, the participant might summarise decisions and responsibilities at the end of meetings, invite questions before closing a discussion, or schedule regular individual check-ins. Progress can then be reviewed through examples and follow-up feedback.

Using AI to explore 360-degree feedback

ChatGPT can help compare groups, organise recurring themes and generate questions for a feedback discussion. Protect respondent confidentiality, remove identifying comments and instruct the tool not to infer who gave particular ratings.

The original report and professional judgement remain central. AI can support structured reflection, but it cannot know the workplace context, verify an explanation or decide which perspective is correct.