Peer evaluations are the one moment in a group project where honesty is formally requested and almost never supplied. Everyone gives everyone fives. The teammate who vanished for three weeks gets "great collaborator." Then people complain that peer evaluation is meaningless, which by then it is.
The incentives explain it. You'll take another class with these people, you may want them as references, and the social cost of a hard review is immediate while the benefit is diffuse. Understandable—but there's a way to be honest that doesn't require being brutal.
Why uniformly positive evaluations hurt everyone
Start with what inflation actually costs, because it isn't obvious.
It costs the person being reviewed most of all. Someone whose contributions consistently arrive late and half-finished, and who is told every term that they're a great teammate, gets to be surprised by that feedback for the first time from a manager who can fire them. Business school is close to the last environment where that information is cheap to receive.
It costs the team, because instructors calibrate future group assignments partly on these signals. And it costs the people who did the work, quietly, in a way that accumulates over a term.
Write about behaviour, not character
The whole difficulty collapses once you stop evaluating people and start describing events.
"Unreliable" is a character judgment. It's contestable, it's insulting, and there's nothing actionable in it. Compare:
Marcus's sections were strong when they arrived, but three of four came in after the internal deadline—in two cases the night before the final deliverable. That left no time to integrate his analysis with the model, so we presented with a gap between the two we'd have caught with a day's margin.
That's harder to argue with, because it's a description rather than a verdict. It's fair, because it names the cost to the team rather than implying a defect in the person. And it's genuinely useful to Marcus, who may not know that late-but-good reads very differently to a team than it does to him.
Three rules that keep you on the right side of this:
- Name the observable thing, not the inferred motive. You saw a missed deadline. You did not see "didn't care."
- Name the consequence. Feedback without cost attached is easy to file away as someone's preference.
- Say what good would have looked like. "Flagging on Sunday that Tuesday would slip" is a specific alternative.
MBA Lab's free peer evaluation drafter walks one teammate at a time through exactly those three rules, flags character verdicts and inferred motives as you type, and asks the proportionality question below before you paste anything into a form.
Say what actually worked, specifically
Positive feedback suffers from vagueness too. "Great teammate" tells nobody anything. What did they do?
When the client data came back inconsistent, Aisha rebuilt the reconciliation in a night and documented the assumptions so the rest of us could check them. That's the reason the model held up in Q&A.
People are frequently unaware of which of their habits are valuable—especially the quiet ones like documenting assumptions or absorbing an ambiguous task without being asked. Naming it specifically is one of the more useful things you can do for a classmate.
The proportionality question
One test before writing anything hard: was this raised during the project?
Feedback that appears for the first time in a final evaluation is an ambush, and it's the main reason peer review breeds resentment. If a problem mattered enough to write down at the end, it mattered enough to say in week four—when it could still have been fixed and when the person had a chance to respond. There's a set of scripts for those in-project conversations, and a short retrospective after each deliverable creates the opening without anyone having to be brave.
If you didn't raise it, you can still write it—but write it acknowledging that: "I should have flagged this in week four rather than here." That's honest about your own part, and it's the difference between feedback and a grievance.
The best version of this never needs the conversation to feel brave in the first place—agreeing at kickoff, in writing, that peer evaluations should never contain a surprise. That's what the accountability section of a team charter is for. MBA Lab's Leadership Lab has a longer piece on holding someone to a standard without it feeling like an attack, which is close to the same skill a fair peer evaluation depends on.
Receiving one that stings
Different skill entirely, and mostly a matter of timing.
Don't reply the same day. Nothing good has ever been written in the first hour after reading a hard evaluation.
Separate the claim from the framing. Someone may have written "checked out in the second half" in a way that feels unfair or overstated. Underneath there is usually a factual core—you did submit two things late, you did miss two meetings. Argue with the framing in your head if you like; take the core seriously.
Ask one clarifying question, not a rebuttal. "Was there a point where it started to feel that way? I'd rather know for the next project." That gets you information. A defence gets you a longer argument and a worse relationship.
Look for the second instance. One person's read of you is a data point. The same theme from a second person on a different team is a pattern, and patterns are worth acting on.
What most people get wrong
Two failure modes, roughly equal in frequency. The first is inflation—perfect scores for everyone, which makes the whole exercise theatre. The second is the anonymous vent, where months of accumulated irritation arrive in one paragraph at the moment the target can't respond.
Both come from the same root: not having said anything earlier. Teams that surface small frictions as they happen almost never face a hard evaluation at the end, because there's nothing left that hasn't already been discussed. The peer evaluation is a lagging indicator of how honest the team was willing to be in week four.
There's a career-shaped reason to get good at this, too. Reviewing colleagues fairly is a real skill you'll need with actual stakes soon enough, and the stories that come out of teams honest enough to have these conversations tend to be the ones that hold up in interviews.