Rating scales corrupt 360 feedback (here's the alternative)
Read more 360 Reviews
Great communicator. Really collaborative. Could work on communication. If you have ever run a 360, you have read a report full of lines like these and felt your heart sink, because none of them tell the person a single thing they can act on. Vague 360 feedback is the default outcome, not the exception, and the usual advice blames the questions. Better questions do help. But the deeper reason feedback comes back vague is simpler and more uncomfortable: the vague answer is the safe answer. Being specific names a real moment and takes real work, so people reach for the generic version instead.
This is a diagnosis. Below is the full set of reasons your 360 feedback stays vague, and how to make people specific instead, which turns out to be mostly about lowering the cost of specificity, not rewording the prompt.
Key Takeaways
- Vague 360 feedback (“great communicator, really collaborative”) is the default, and it is mostly not a question-design problem. The specific answer feels socially risky and takes effort, so raters default to safe praise.
- Better questions help, but rewording alone will not fix vagueness. You also have to lower the social cost of being specific and force an example into the ask.
- Feedback given long after the moment is generic by default, because the specific example is already forgotten. Feedback close to the event stays concrete.
- The synthesis step quietly destroys the specifics you did collect, when the organiser summarises five real examples into “communication came up a lot.”
- The fix is to make the specific thing safe and easy to say, keep the actual examples attached, and refuse to accept the vague version.
Start with why a rater writes “great communicator” when they could write something useful. Usually it’s not laziness, but the fact that the specific version is exposing.
“Great communicator” commits to nothing. The specific version, “she walks the team through the tradeoffs before a decision so people actually buy in, but she goes quiet when a plan is falling apart and we find out late,” names real moments, and by naming them it points a finger. Even as praise, specificity feels a little like telling on someone. As criticism, it feels like an accusation you will have to own if it gets traced back to you.
So the rater does the safe thing. They write the version that could apply to anyone, cannot be argued with, and will never come back to bite them. Multiply that instinct across every rater and you get a report full of statements that are technically true and completely useless.
This is the part most articles skip, because the fix is uncomfortable: you cannot reword your way out of a safety problem. If people do not feel safe being specific, no question phrasing will make them. That is why safety is upstream of the whole process, and why anonymity, used well, often unlocks the specifics that attributed feedback buries.
Safety gets you willing raters. The second problem is that even willing raters need to be asked in a way that pulls the specific out of them, and most 360 questions do the opposite.
A question like “is this person a good communicator?” invites a one-word verdict. “Yes.” Done. The rater was ready to tell you more, but the question did not ask for more, so they matched their answer to the question and moved on. The fix is not a longer form. It is a prompt that makes the vague answer impossible: “describe the last time this person’s communication changed a decision or a plan.” Now they cannot answer with “good.” They have to reach for an actual moment.
This is the part of the vagueness problem that better question design genuinely solves, and I have written the full version separately, so I will not repeat it here. The pillar on 360 feedback questions covers how to build prompts that ask for observable behaviour instead of a trait rating. The one rule worth carrying into the rest of this piece: every prompt should demand an example, not an adjective.
Worth naming the worst offender briefly. If open questions produce vague words, rating scales produce vague numbers, and a number is the least specific answer a 360 can return. “4 out of 5 on leadership” tells the person where they allegedly stand and nothing about what to change. It’s a verdict with the evidence stripped out. I have made the full case that a rating scale on its own is not feedback; the short version here is that a score is vagueness with a decimal point, and it only earns its place when a written example sits next to it.
Here is a cause that has nothing to do with questions or scores: when you ask matters as much as how you ask.
If your 360 runs once a year, you’re asking people to recall a year of someone’s behaviour in one sitting. They cannot. The specific moment, the exact meeting where it went well or badly, is long gone, so what comes back is a general impression: “reliable, good with clients, could be more proactive.” That’s not evasiveness. It’s memory. Even a rater who wants to be specific has nothing specific left to reach for.
Feedback given close to the event is specific because the example is still fresh. This is one more reason to run feedback as a rhythm rather than a once-a-year reckoning, not just to lower the stakes, but to catch the details before they fade. Specifics have a short shelf life.
Say you get past all of that: safe raters, sharp prompts, timely feedback. You collected genuine specifics. There is still one place they go to die, and it is the summary.
The person running the 360 reads a dozen responses and, trying to be helpful, condenses them into themes. “Several people mentioned communication.” “Collaboration is a strength.” The intent is good, and the reviewee is left with the same vague nouns the raw responses were supposed to escape. The concrete examples that made “communication” mean something got averaged back into the word “communication.”
Imagine an organiser at a 50-person company summarising a 360 for a team lead. Three reviewers described one specific pattern: the lead rewrites people’s work without telling them, which is quietly demoralising the team. In the summary, that becomes “some feedback around delegation and trust.” That version is technically accurate and completely defused. The lead reads it, nods, and changes nothing, because there is nothing left to change.
The fix is to keep the examples attached. A summary should point to the specific quotes, not replace them. This is the one place a tool genuinely helps: it can hold the written responses in full next to the aggregated view, so the summary is a way in rather than a substitute. It’s how we built written feedback and criteria ratings together in Lynxify.me, with the words kept attached to the numbers and the summary sitting beside them, not on top of them. You can enforce the same rule in a shared document: no theme without the quote that earned it.
Pull it together. Getting specific 360 feedback is less about clever wording and more about removing the reasons people stay vague. Five moves, roughly in order of impact:
None of this requires software, and the mechanics of running a round this way are in how to run your first 360 review. What it requires is a decision to stop accepting the vague version, from the way you ask to the way you hand it back.
Vague feedback is not a mystery, and it is not your team being unhelpful. It is the predictable result of asking people to say something exposing, in a way that invites a shrug, about something they have half-forgotten, and then averaging whatever specifics survive back into a word. Fix those one at a time and the vagueness goes away on its own.
The reframe I would leave you with: specificity is not something you extract with a cleverer question. It’s what you get when the specific thing is safe to say, easy to give, and impossible to hide behind a theme. Build the process for that, and “great communicator” turns back into the real, useful, slightly uncomfortable sentence it was always standing in for.
If you would rather run it without a spreadsheet that quietly flattens the specifics, you can start your first 360 in Lynxify.me. But the rule matters more than the tool: never accept a theme without the example that earned it.
Why is 360 feedback so vague?
Because the specific answer is the risky answer. Naming a real moment, even a positive one, exposes the person giving it, so raters default to generic statements like “great communicator” that commit to nothing and cannot be traced back. Vagueness is also driven by questions that ask for a trait rather than an example, by feedback collected long after the moment when the specifics are forgotten, and by summaries that flatten real examples into themes. Better wording helps, but the root cause is usually safety and effort, not the question.
How do you give specific feedback in a 360 review?
Anchor it to one observable moment. Instead of “good communicator,” write what you saw and what it caused: “in the last planning session, she laid out the tradeoffs so clearly that the team stopped relitigating the decision.” The test for a specific piece of feedback is whether the recipient could picture the exact situation and know what to do more or less of. If it could apply to anyone, it is not specific yet.
How do you get more specific answers from raters?
Change what you ask for and lower what it costs. Replace trait questions (“is she reliable?”) with example prompts (“describe the last time you relied on her and it mattered”), keep the feedback developmental and anonymous where naming names is too costly, and run it close to the event so the examples are still fresh. Then protect the specifics in synthesis: never summarise an example into a theme without keeping the example attached.
Dmytro Shtapauk
Practical tips and experience from top-notch experts
