360 feedback questions for leaders and managers – with examples for every rater group

What makes a good 360 question

A 360° questionnaire collects views of one leader from several directions: the leader’s own, and those of supervisors, peers and direct reports. Bracken, Rose and Church define 360° feedback as a process for collecting, quantifying and reporting co-workers’ observations (Bracken, Rose & Church 2016). At its core, it records how far specific behaviours are shown and compares the ratings between groups and over time. Every question decides what the report can say. Four tests separate good questions from weak ones (the weak examples are invented):

• Observable. Raters should report what they see, not what they think goes on in the leader’s mind; Bracken and colleagues warn against wording such as “understands”, “believes” or “knows”. “Believes in the company’s strategy” fails this test.
• One behaviour per question. “Is punctual and well prepared” mixes two things, and a middling answer cannot tell you which one is the problem.
• Answerable by that rater group. Direct reports see how work is handed over; supervisors see how risks are escalated. A question that nobody in a group can observe produces guesses.
• Development-oriented. Ask about behaviour a leader can change. A trait label such as “Is charismatic” gives no clue what to do differently.

Five types of question – and who can answer each

Most 360° tools answer one question: what leadership behaviour do others observe? That is the core, but a questionnaire can combine five types of question, as long as each rater group only receives what it can judge:

• Behaviour frequency (Behavior). How often does the leader do something specific? Answered by every rater group and by the leader.
• Outcome. What follows from the behaviour, for example whether priorities hold or agreements stick? Also answered by everyone.
• Capability. What does the leader trust themselves to do, and what do others trust them to do? In the self-assessment this is self-efficacy in Bandura’s sense, not proven skill.
• Belief. What does the leader believe about leading, for instance about what a team needs? Self-assessment only.
• Reactions under pressure (Trigger). Which situations throw the leader off, and what impulse follows? Only the leader can report that from the inside; what shows on the outside is covered by behaviour questions.

The self-only types gain their value in comparison with what others observe. Self–other agreement on questions both sides answer is a research field of its own, and Fleenor and colleagues note that results depend partly on how agreement is calculated (Fleenor et al. 2010). For questionnaire design, the lesson is that leader and raters should answer the shared types (behaviour, outcome and capability) in comparable wording on the same scale. The methodology shows how LEADBeyond 360° assigns the types, and self-awareness in leadership explains why the comparison matters.

How many questions – and how many competencies

There is no ideal number of questions. Plan a time budget instead, because one rater may fill in several questionnaires per round. LEADBeyond 360° measures twelve dimensions in three pillars, plus limiting tendencies, reactions under pressure, Well-being and Outcome, with 99 items. The leader needs about 18 minutes; raters answer a version for their group in 10–15 minutes (how to select 360° raters lists the count per group). For competencies, Bracken and colleagues recommend rating behaviours rather than clusters. A practical rule: measure no more competencies than you can describe through observable behaviours and discuss in one debrief.

Example questions by rater group: supervisors, peers, direct reports

Example questions across Self-Leadership, Leading Teams and Leading Change

Rating scales and the “cannot assess” option

The five-point agree–disagree scale is the most common: Bracken, Rose and Church cite 3D Group benchmark research from 2013 in which 87% of companies used it. They also note that agree–disagree gives raters no frame of reference and that 360° data are sensitive to the choice of scale. Three decisions follow:

• Match the anchors to the question. Ask how often for behaviour and outcomes; keep agree–disagree for beliefs and capability.
• Decide on the midpoint. Respondents read a midpoint in different ways (Nadler et al. 2015); a six-point scale without one avoids that and asks raters to lean one way. The price: mixed impressions must still pick a side, so do not over-read small differences.
• Add “cannot assess” and keep the scale. The option must not count towards scores, and the scale should stay the same across rounds.

Open questions: start, stop, continue – with examples

Open questions supply what numbers cannot: examples, reasons and suggestions. A simple structure is start, stop, continue: start asks what is missing, stop what costs more than it gives, and continue which strength should stay. In a UK study, this format was associated with more constructive, in-depth comments than a free-text box, although it looked at course feedback rather than leadership (Hoon et al. 2015). Invented examples of vague and useful comments, including a strength and a weakness phrased as a development area:

• Start. Instead of “Could communicate better”, write: “Start sending the agenda the day before team meetings, so we can bring the figures you usually ask for.”
• Stop. Instead of “Sometimes difficult”, write: “Stop answering emails during our one-to-ones; the conversation keeps breaking off and we rarely finish.”
• Continue. Instead of “Great job!”, write: “Keep the Friday check-out where everyone names one thing that blocked them; we solve most of those before Monday.”
• Strength. Instead of “Good with people”, write: “Explains the reasoning behind staffing decisions, so nobody has to guess.”
• Development area. Instead of “Disorganised”, write: “Agreements from our one-to-ones are sometimes forgotten; writing down two or three actions at the end would help us both follow up.”

Write about behaviour, not personality: according to Kluger and DeNisi’s feedback intervention theory, feedback loses effect the more it shifts attention from the task to the self (Kluger & DeNisi 1996). Is 360° feedback anonymous? explains how to write comments that cannot be traced back to you.

Giving 360 feedback to your boss: examples that help

Feedback to your own manager can feel risky. In one study, a few specific criticisms went with more improvement than a long list. Among 176 managers, those who received a small number of unfavourable comments about behaviour or tasks improved more than others, while those who received many such comments declined more (Smither & Walker 2004). Choose one or two points and write each in three steps, shown here with an invented example:

1. Describe the situation. Name a recurring situation rather than a single incident: “In our weekly planning meeting, priorities often change after we have agreed them.”
2. Name the effect. Say what it does to you or the team: “We then redo work that was already finished.”
3. Suggest one change. Keep it concrete and doable: “Writing the three agreed priorities into the meeting notes would let us plan once.”

Templates, forms and a real instrument

Free 360 feedback templates and forms are a reasonable start for a one-off team check-in, provided each question passes the four tests above. For a leadership 360°, a template is rarely enough: it measures only what it asks and offers no hypothesis about why the self-view and others’ view differ. Follow-up matters too: in Smither, London and Reilly’s meta-analysis, improvement after multisource feedback was generally small; their review suggests it is more likely when recipients set goals and take action (Smither, London & Reilly 2005).