360 Feedback Form Template
Structured peer review for development, not judgment — a behavior matrix, relationship context, and written strengths and growth areas on a second page.
You are giving development feedback about a colleague. Be specific and be fair — this goes into a growth conversation, not a permanent record.
A 360 review collects the view from every side of a person — manager, peers, reports, cross-functional partners — because each side sees behavior the others structurally cannot. The template's design problem is different from every other feedback form: the respondent is not a customer with an opinion but a colleague with an obligation, and the enemy is not low response rates but vague kindness. Every choice here pushes toward specific, usable observations.
Why these fields. The relationship question is the analytical key: a "rarely" on handling disagreement means one thing from a peer and another from a direct report, and synthesis without this context flattens the signal 360s exist to capture. The behavior matrix uses a frequency scale — rarely to consistently — rather than a quality scale, because "how often do they do X" is answerable from memory while "how good are they at X" invites bias and grade inflation. The five rows are behaviors visible from outside: communication, delivery, developing others, conflict, and initiative. The second page shifts from ticking to writing, and the prompts do the coaching — strengths must come with a witnessed moment as evidence, growth areas are capped at one or two and framed as opportunity. The final advice question consistently produces the line the subject actually remembers a year later.
What we left out. Numeric overall scores (they get averaged into rankings, and development instantly becomes judgment), anonymous free-fire comment boxes (anonymity plus no structure equals axe-grinding), and twenty-competency frameworks — five behaviors answered honestly beat twenty answered on autopilot.
Who uses this. People teams run it in review cycles for managers and senior ICs, founders use it on themselves with their leadership team as reviewers, and coaches gather it before an engagement begins.
Make it yours. Swap the matrix rows for your company's actual competency language so feedback lands in words your development plans already use. Clone the form per review subject to keep response pools cleanly separated, and set a close date to hold the review window. Export each pool as CSV for synthesis — the relationship column lets you split every rating by reviewer angle, which is where the interesting patterns live.
Composing the reviewer panel. The relationship question only pays off if the panel actually spans it, so pick reviewers before opening the form: the manager, two or three peers, at least two direct reports where they exist, and one cross-functional partner who feels the person's work without sharing their goals. Two per angle is the practical floor — a single voice per relationship is neither anonymous nor representative. Resist packing the panel with allies; subjects tend to nominate their fans, and a cycle run on fans produces warm noise. Balanced panels are also kinder in the end, because praise that survives skeptics is praise a person can build on.
Frequently asked questions
Should reviewers be anonymous to the subject?
Reviewers are identified by relationship, not name, by default. The synthesis step — combining responses into themes — is what protects individuals while keeping context.
How do we run this for ten people at once?
Duplicate the form per subject — one click each — so every person has their own response pool and close date. A shared naming convention keeps the exports tidy.
Can reviewers save and come back?
Partial responses are captured automatically as reviewers type, so a half-written review survives a closed laptop. Thoughtful 360s are often written across two sittings.
How is the matrix data analyzed?
The CSV export gives one column per behavior row. Average by reviewer relationship, and gaps between the manager view and the report view become immediately visible.
Why two pages instead of one?
The page break separates quick structured ratings from slower written reflection — reviewers finish the matrix, then commit to prose. Completion improves when the gear change is explicit.