Give peer feedback that's honest and useful.
The structure that separates 'this feedback made me want to be better' from 'thanks for the encouragement, back to work'.
It's peer review season. You want your feedback to actually help — not the vague corporate 'strong contributor, would benefit from developing X'. This prompt writes the version that lands.
THE PROMPT
Draft peer feedback for [PERSON]. I'll paste our working relationship, what they did well, and where I think they could grow.
Structure the feedback exactly:
**What they do best (specific)**
The one thing this person does better than most peers in their role. Named with a specific example — a project, a moment, a call I saw them handle. Not "great communicator"; the specific communication moment that showed the skill.
**The one thing that would most compound their impact**
The single growth edge — not a laundry list. Named with the specific pattern I've observed (2-3 example moments). Framed as "what would happen if you did more of X" rather than "you don't do X well".
**What I'm not seeing but might be there**
Blind spot check — the thing about their work that I might be missing because of my vantage point. Naming this shows the feedback is honest, not omniscient.
**One specific ask for the next 6 months**
The one behavior change I'd love to see. Concrete enough that we'd both know if it happened.
No corporate softening ("would benefit from"). No triple-decker sandwich. No "just my two cents". Warm, direct, specific.
Context:
[PASTE RELATIONSHIP + OBSERVATIONS]**What Priya does best**
Priya asks the question that punctures the fake-consensus. In the June roadmap review, everyone was nodding at the proposed Q3 plan; she asked "what are we not committing to that we secretly know we should?" — the answer reshaped the whole quarter. This is a rare and load-bearing skill. Most senior people I've worked with never learn it.
**The growth edge**
The follow-through after the puncture. When Priya asks the sharp question, she often steps back and lets others resolve it. She could own the space more — propose the answer she thinks is right, not just the question that reveals the gap. In the same June meeting, when the question landed, she waited a beat and then handed the discussion to me. I think if she'd said "and I think the answer is X", we'd have landed the decision 40 minutes faster.
**What I'm not seeing**
I mostly see Priya in cross-functional meetings. I don't see her 1:1s with her own team, so if she's doing this follow-through there and I just don't see it, that's my miss.
**Ask for the next 6 months**
When you ask the puncturing question, follow with the answer you'd bet on — even if it's "and I don't know". Own the direction, not just the diagnostic. I'll notice if you do.