What is upward feedback? Upward feedback is input from direct reports about their manager: how that person sets direction, coaches, unblocks work and treats the team. It can be collected on its own, or as the direct-report group inside a 360. It is not the same as the questions a manager asks an employee in a review, and it is not, on its own, a full 360.

What should you ask, and how do you use the answers? Ask a short set of questions about observable management behaviours, plus two or three open questions, and decide anonymity before you send the form. The Playbook starting point is to keep upward feedback anonymous and aggregated, and not to issue a separate report when there are fewer than three to five raters. Use the patterns as evidence for the skip-level manager and for the manager’s development plan. Do not turn the average score into the manager’s performance rating, and do not try to identify who wrote an anonymous comment.

This page is for People leaders and HR teams at employers of about 200 to 2,000 people who already run reviews — and often a 360 — and need a safe, usable way to ask direct reports about their manager. It is about managers, not the downward questions a manager asks an employee (those are on manager performance review questions), and it is not about an “upward review of salary” or pay progression. If you have a draft question set, or last cycle’s upward themes with names removed, you can book a 15-minute review with Ash Taherian, Product Sales Executive, to walk through how the answers should inform the manager’s review.

What upward feedback is

Upward feedback is input from direct reports about their manager — how that person sets direction, coaches, unblocks work and treats the team — collected so the organisation can develop managers and, carefully, inform the manager’s own review. It is also called an upward review or, in a 360, the direct-report rater group.

It is not the questions a manager asks an employee in a performance review. It is not, on its own, a full 360, which also includes peers, the line manager and a self-review. And it is not a popularity contest or a hidden formula that becomes the manager’s grade.

Used well, it gives the skip-level manager and People a view they cannot get from delivery numbers alone. Used badly, a small team’s comments are easy to unmask, or an average score is treated as the rating. The rest of this page is about avoiding both.

Many People teams already collect some of this input inside a 360, but only as a short block, and then are unsure what to do with it when the person under review is a manager. Others have never asked reports about managers at all, and only have downward reviews and delivery numbers. In both cases the need is the same: a short, safe question set, a clear anonymity rule, and a written rule for how the answers inform development and the manager’s review — without becoming a hidden grade.

Questions to ask about a manager

The question set below is an example to adapt, not a validated psychometric instrument and not a published “manager effectiveness index”. Write questions about observable management behaviours the rater has actually seen. Keep the list tight: the scored items here are ten, plus three open questions. Do not send a 40-item vendor list in the same cycle as a review.

Use a simple frequency scale, and always offer a way not to guess: Rarely or never · Sometimes · Usually · Almost always · Not able to observe. That last option matters. A report who has been on the team for three weeks should not have to invent a view on coaching. This scale is a starting point for a do-it-yourself form. It is not a claim that any software product uses it, and it is not the formal performance grade.

Example questions to adapt. They are original to this page and are not copied from the 360 template’s short upward block.

Direction and priorities

  • In the last period, how clearly has this manager set the team’s priorities so you know what to work on first?
  • When priorities change, how well does this manager explain what has changed and what can wait?

Coaching and feedback

  • How regularly does this manager give you useful feedback on your work, not only at formal review time?
  • When you need help improving, how practical is the coaching or guidance you receive?

Unblocking work

  • When you are blocked, how promptly does this manager help you get a decision, access or a stakeholder unstuck?
  • How well does this manager protect the team’s time from avoidable noise or conflicting requests?

Fairness and psychological safety

  • How consistently does this manager treat people in the team fairly, including how work and recognition are shared?
  • How safe is it to tell this manager about a problem, a mistake or a disagreement without it being held against you?

Development

  • How well does this manager support your development — stretch work, feedback, or time to build a skill?
  • How clearly can you see what “good” looks like in your role, based on what this manager has explained?

Open questions

  • What should this manager keep doing, because it helps the team?
  • What is one change this manager could make that would most help you do your job well?
  • Is there anything else People or the skip-level manager should know about how this team is managed? (Optional. Remind people not to include details that would identify them if the report is aggregated.)

Under each scored item, ask for one short example if the rater can give one: what happened, roughly when, and what the effect was. Do not ask people to diagnose personality.

How to collect it safely

Decide anonymity before you send the form, tell people the rule, and keep the same rule for the whole rater group. Do not change it after the first comments arrive.

This page does not rewrite the Playbook’s anonymity essay. The starting point already used on the 360 feedback template — and set out in should 360 feedback be anonymous — is to keep upward feedback anonymous and aggregated, with a three-to-five rater floor, or to combine groups if the team is smaller. Follow that starting point here. If a manager has two reports, a promised “anonymous upward report” is not anonymous. Two reports and a distinctive phrase will identify the writer. Combine groups, or give the skip-level manager a qualitative summary that cannot be traced, rather than a two-person score pack.

Never promise anonymity you cannot keep. Employees’ fear of retaliation is the reason the rule exists, not a reason to skip upward feedback altogether. If you cannot meet the rater floor, say so plainly: the input may still be collected, but it will be summarised for the skip-level manager rather than issued as a separate anonymous report.

Decide who will see comments — typically People and the skip-level manager, and the manager only in a form that cannot be traced — and write that on the form. Named, ongoing conversation still belongs in one-to-ones. Do not use a 1:1 to hunt for who wrote an anonymous comment, and do not ask a report to “stand behind” a theme in front of their manager.

How to use the answers in the manager’s review

Upward feedback is evidence the skip-level manager and People can consider. It is not a hidden formula that becomes the manager’s rating, and it should not be given a made-up weighting such as “20% of the grade”.

Look at patterns across reports, not at one comment and not at a single average score. Where you have them, read the upward themes next to other inputs: delivery against the manager’s goals, the manager’s own review evidence, and calibration against your published rating scale. One glowing report and two concerned reports is a pattern to explore, not a number to average away.

Use the output in three places:

  1. A development plan for the manager. Turn two or three themes into actions, support and a follow-up date. The development plan and career conversations guide is the place for that write-up; manager training for performance management is the wider skill-building context.
  2. A structured conversation between the skip-level manager and the manager. Focus on behaviours and examples, not on “who said this”. If a comment would identify the writer, keep it out of the pack or summarise it so it cannot be traced.
  3. One input at calibration when the person being calibrated is a manager. It sits alongside other evidence. It should not override the agreed rating scale. See also how to run a calibration meeting.

If the pattern is serious and sustained — for example, a consistent fairness or safety concern that does not change after a development plan — that is a different process, not a quieter use of the same survey. A formal route, where it applies, is the UK capability procedure or, for a performance-improvement plan, the PIP guide. Do not jump there from one cycle of mixed comments.

Close the loop. People stop answering honestly if they never see any change, or if they suspect the comments were used to settle a score. Tell the team, in general terms, what will be worked on. Do not read comments back in a way that points at an author. The write-up of the manager’s review itself can follow the performance review template.

How this sits next to a 360 and a manager-effectiveness survey

If you already run a 360, reuse the direct-report group rather than sending a second long form in the same cycle. The 360 feedback template is the multi-rater template, including a short upward block. This page is the upward-only how-to: a fuller question set, how to collect it safely, and how the answers are used in the manager’s review and development.

A short “manager effectiveness survey” can be the same upward question set, used mid-cycle or outside the review window, so the manager has time to work on themes before ratings are written. It is not a separate commercial instrument, and this page does not invent one.

Engagement surveys are a different product and a different purpose. They can tell you how people feel about work, leadership or the company in general. They are not a substitute for asking direct reports about a named manager’s behaviours, and they should not be treated as upward reviews.

Common mistakes

  • Sending downward “manager feedback questions for employees” by accident. Those questions face the other way. Use this page for reports about a manager, and the manager review questions when a manager is reviewing a team member.
  • Copying a 40-question vendor list. A short, observable set is more likely to be finished honestly.
  • Promising anonymity in a two-person team. If you cannot keep the promise, do not make it.
  • Showing raw comments to the manager in a way that identifies the writer. Edit or summarise before anything is shared.
  • Treating the average score as the manager’s performance rating. Patterns are evidence; they are not the grade.
  • Never closing the loop. If nothing is acknowledged, the next cycle will be less honest.
  • Using upward feedback as a surprise at year-end, with no development conversation first. Themes should reach the manager in time to act, not only as a rating-season shock.

Where Revolut People fits

If you use Revolut People, the documentation describes Upward Reviews — the docs also use the spelling “Upwards Review” — as an employee’s evaluation of their line and/or functional manager. The points below are from the live documentation page Completing Upward Reviews, checked on 15 November 2026. They are not a claim that the product ships this page’s question set, and they are not a software tutorial.

The docs say this feedback, “while not directly affecting performance assessments, serves as valuable data for higher-level managers.” That matches the rule above: upward input can inform the skip-level view and the manager’s development; it should not, on its own, decide the manager’s rating, and the product should not be described as applying a hidden upward weighting to the grade.

The same pages say upward reviews “provide anonymous feedback to the reviewed manager’s Line Manager.” In the steps, the reviewer is told: “Your review is anonymous. You can choose whether it’s shared with the reviewee and their managers, or just their managers.” The visibility options listed are: with just their managers (anonymous), or with the manager being reviewed and their managers (anonymous). Both options are described here as the docs state them. There is no documented admin “upward-review anonymity settings” screen, and this page does not invent one.

The upward scorecard covers management behaviours only. Those behaviours are defined by admins in the organisation. Raters are not rating deliverables, goals or skills in that scorecard. The rating choices stated on the docs are Don’t know, Developing, Performing, or Exceeding. Raters can optionally add written context. An optional Values section may be enabled; the docs say those expectations are consistent across all review types. The company may also include additional questions before submission. Do not treat Don’t know / Developing / Performing / Exceeding as the Playbook’s recommended scale for a do-it-yourself form — the frequency scale earlier on this page is the starting point for that use.

If you want a second look at a draft question set, or last cycle’s upward themes with names removed, book a 15-minute review with Ash Taherian, Product Sales Executive. Soft fit-check only.

FAQ

What is upward feedback?

Upward feedback is input from direct reports about their manager: how that person sets direction, coaches, unblocks work and treats the team. It can be collected on its own, or as the direct-report group inside a 360. It is not the questions a manager asks an employee, and it is not, on its own, a full 360.

Is upward feedback the same as a 360?

No. Upward feedback is one direction: reports about their manager. A 360 is multi-rater. You can run upward feedback on its own, or as the direct-report group inside a 360. If you already run a 360, reuse that group rather than sending a second long form in the same cycle. See the 360 feedback template.

What should you ask about a manager?

Ask a short set of questions about observable management behaviours — direction and priorities, coaching, unblocking work, fairness and psychological safety, and development — plus two or three open questions. The question set on this page is an example to adapt, not a validated instrument. Always include a “not able to observe” option.

How do you collect upward feedback safely?

Decide anonymity before you send the form, tell people, and keep the same rule for the group. The Playbook starting point, set out on the 360 template and the anonymity note, is to keep upward feedback anonymous and aggregated, with a three-to-five rater floor, or to combine groups if the team is smaller. Never promise anonymity you cannot keep.

How do you use the answers in the manager’s review?

Use patterns as evidence for the skip-level manager and for the manager’s development plan, and as one input at calibration when the person being calibrated is a manager. Do not turn the average score into the manager’s performance rating, and do not try to identify who wrote an anonymous comment.

Does upward feedback decide the manager’s rating?

It should not, on its own. It is evidence to consider alongside delivery, the manager’s own goals and calibration against your published scale. Do not invent a weighting that turns the upward average into the grade.

Related guides

360 feedback template · should 360 feedback be anonymous · manager performance review questions · one-to-one meeting template · manager training for performance management · manager engagement and coaching · development plans and career conversations · performance calibration · how to run a calibration meeting · employee performance rating scales · performance management framework · performance review template · free Playbook tools