What is the difference between OKRs and performance reviews? OKRs (objectives and key results) are a goal-setting method: a short list of ambitious outcomes for the company, a team or a person, each with a few measurable key results, set and graded every quarter so that everyone can see where effort is going. A performance review is a judgement about one person: how well they contributed over a period, both what they delivered and how they worked, recorded against an agreed standard and used for development, promotion and often pay. In short, OKRs steer the work and a review assesses the person. The two should be connected, but an OKR score should be one piece of evidence in a review, not the review itself.

Key takeaways

  • Different jobs. OKRs answer “what are we trying to move this quarter?”; a review answers “how well did this person contribute, and what happens next for them?”
  • Different scoring. Google’s re:Work guide grades key results from 0.0 to 1.0 and treats 0.6 to 0.7 as the healthy range for stretch goals. A review rating that expected people to miss a third of their targets would make little sense.
  • Google itself separates them. The same guide says OKRs are not synonymous with employee evaluation; at most they summarise what someone worked on and how it fed the bigger objectives.
  • Goals work, with side effects. Specific, hard goals raise performance (Locke and Latham, 2002), but Ordóñez and colleagues (2009) list narrowed focus, riskier choices, unethical shortcuts and less learning when goals are over-used.
  • Unclear expectations are the cost of getting it wrong. Gallup’s January 2026 analysis found clarity of expectations had the largest drop of any US engagement item since 2020, down nine points.

Follow us on Google

OKRs and performance reviews side by side

Most of the confusion comes from the fact that both involve goals, both happen on a cycle and both end with a score. The overlap stops there. The table sets out where they differ for a typical company of 100 to 1,500 people.

QuestionOKRsPerformance reviews
PurposeFocus and align effort on a few outcomes that matter nowJudge an individual’s contribution and decide development, promotion and pay inputs
Main unitCompany and team first; individual OKRs are optionalAlways the individual
What is measuredProgress on measurable key resultsResults and behaviours, skills and values against a written bar for the role
Typical cadenceSet and graded quarterly, with check-ins in betweenFormal review once or twice a year, often with quarterly check-ins
Scoring0.0 to 1.0 or traffic lights; missing some stretch results is expectedRating or grade against a standard; meant to mean the same thing across teams
VisibilityPublic inside the companyConfidential to the person, their manager and HR
ConsequencesPlans change; resources move; next quarter’s OKRs improveDevelopment plans, promotion decisions, pay inputs, and improvement plans
Who checks fairnessLittle need; a key result is mostly a fact about the workCalibration across managers before ratings are final

The last row is the one People teams most often overlook. A key result is either hit or not, so there is little to calibrate. A rating is a judgement made by many different managers, which is why it needs performance calibration before it can be trusted.

What the evidence says

Specific, hard goals work. Edwin Locke and Gary Latham’s review of 35 years of goal-setting research, Building a practically useful theory of goal setting and task motivation (American Psychologist, 2002), concludes that specific, difficult goals lead to higher performance than vague instructions to do one’s best, with feedback, commitment, ability and task complexity shaping how large the effect is. This is the research OKRs rest on.

Goals also have side effects. Lisa Ordóñez, Maurice Schweitzer, Adam Galinsky and Max Bazerman argue in Goals gone wild (Academy of Management Perspectives, 2009) that the harm caused by goal setting has been largely ignored. They list a narrow focus that neglects work outside the goal, distorted attitudes to risk, more unethical behaviour, less learning, damage to culture and lower intrinsic motivation, and recommend treating goals like a prescription medicine rather than an over-the-counter remedy.

Task type matters. The CIPD’s evidence review Performance management: could do better? (2016) found that specific, challenging goals help most on relatively straightforward tasks, but not on complex work that involves interrelated steps, unfamiliar cues or new skills, where learning or behaviour goals do better. It also found that feedback is essential for goals to work, and that ratings in developmental appraisals tend to be stricter and draw on more examples than ratings used for pay and promotion, so it recommends separating developmental conversations from administrative ones.

The people who popularised OKRs kept them apart from evaluation. Google’s re:Work guide Set goals with OKRs (2016) describes grading key results from 0.0 to 1.0 with a healthy average around 0.6 to 0.7, and states that OKRs are not synonymous with employee evaluation; they can show what a person worked on and how it contributed, but they are not a complete way to judge someone. The What Matters site, which draws on John Doerr’s Measure What Matters (2018), explains in What is an OKR? that Andy Grove’s OKRs differed from the older management-by-objectives approach in being quarterly rather than annual and in being separated from compensation.

The gap shows up as unclear expectations. In U.S. employee engagement declines from 2020 peak (January 2026), Gallup reports that “I know what is expected of me at work” had the largest decline of any engagement item among US employees since 2020, down nine points. Of employees who answered an open question on what would help, 35% pointed to better communication. These are US figures, but the mechanism is familiar to UK and Irish teams: goals in one place, reviews in another, and nobody joining them up in conversation.

Where the confusion costs a 100 to 1,500-person company

Companies of this size usually adopt OKRs during a growth phase and keep a review process inherited from an earlier stage. The two then collide in predictable ways:

  • Sandbagging. If the review rating is the average OKR grade, the sensible move for any employee is to set goals they know they can hit. Stretch quietly disappears.
  • Invisible work. Mentoring, fixing incidents, hiring and helping other teams rarely appear as key results. Ordóñez and colleagues call this the narrow-focus effect; in a review it shows up as good colleagues rated as average.
  • Team results credited to individuals. Most OKRs belong to a team. Turning a team grade into a personal rating rewards people for their team’s luck and penalises them for its dependencies.
  • Goals that stop at the deck. Company OKRs live in a slide, managers set individual objectives in a separate form, and at review time nobody can show the line between them.
  • Mid-quarter changes nobody records. Priorities shift, OKRs are dropped, but the review form still lists the original objectives, so managers rate people against work they were told to stop.

The fix is not to choose one over the other. It is to give each its own job and connect them deliberately.

How to connect them without breaking either

The working model I see in well-run companies of this size treats OKRs as evidence and the review as the judgement:

  • Keep OKR grades honest by keeping them low-stakes. Grade key results openly at the end of the quarter and treat a 0.6 on a real stretch as a good result. Do not average them into a rating.
  • Review contribution, not completion. In the review, the manager asks what the person did to move the team’s key results, how hard the goal was, what changed during the quarter and what they did that no OKR captured.
  • Score the “how” separately. A scorecard with deliverables, skills and values stops a good OKR quarter excusing poor behaviour, and stops a missed stretch goal sinking a strong contributor. The performance review template shows one way to lay this out.
  • Match the goal type to the work. Use outcome key results for predictable work and learning or behaviour goals for complex or new work, as the CIPD review suggests. The goal-setting guide covers the options.
  • Calibrate the rating, not the OKR. Managers compare review ratings against the same bar before they are final. OKR grades are inputs to that conversation, not the thing being calibrated.
  • Keep pay downstream of the calibrated rating. If you link reviews to pay, do it through the rating and written criteria, never directly through the OKR score. The reasoning is set out in Should performance reviews be linked to pay?

For engineering and product teams, where most key results are shared, the OKR and performance review alignment guide works through the same model with worked examples.

Our perspective

The People leaders I speak to rarely have an OKR problem or a review problem on its own. They have a seam problem. Objectives are set in one tool in January, reviewed in a form in June, and the only thing linking them is a manager’s memory. When the review arrives, managers either score the OKR spreadsheet, which punishes ambition, or ignore it, which tells everyone the goals did not matter.

My view is simple: let OKRs be public, ambitious and a little uncomfortable, and let reviews be private, calibrated and fair. Then build one bridge between them, a short note per person each quarter on what they contributed to the key results and how. That note is what the review should draw on. If your managers cannot write it in ten minutes, the problem is not the framework; it is that goals and day-to-day work are not visible in the same place.

— Ash Taherian, Product Sales Executive, Revolut People

How Revolut People handles goals and reviews

I work at Revolut People, so read this section with that in mind. The product keeps goals, 1:1s and scorecard reviews in one place: company goals can cascade to teams and individuals, goal progress can update from tools such as Jira, and the review scorecard pulls in deliverables alongside skills and values, with calibration before ratings are final. That is useful for the seam described above, because the evidence a manager needs at review time is already attached to the person. You can see how this looks in the product tour, and from the operating side in Revolut People for Chiefs of Staff.

The limits matter too. Software will not write good key results, decide how much stretch your culture can take, or stop a leadership team averaging OKR grades into ratings if that is what it chooses to do. If you only need a place to publish company OKRs and are happy with your review process, a dedicated goals tool or a well-kept spreadsheet may be enough. The case for one platform is strongest when the hand-off from goals to reviews is where your cycle breaks.

If you want a second view on how your goals and reviews connect today, book a 15-minute review. Bring one team’s OKRs and your review form, with names removed.

Step plan: one quarter

  1. Week 0: write down the two jobs. One paragraph each: what OKRs are for and what reviews are for, including a line saying OKR grades are not averaged into ratings. Share it with managers before the quarter starts.
  2. Week 1: set company and team OKRs. Three to five objectives, each with about three outcome-based key results. Mark each OKR as committed or stretch so a 0.6 is read correctly later. The goal cascade worked example shows how this flows from a leader to squads.
  3. Week 2: agree individual contribution, not individual OKRs. Each person writes two or three lines on how they will contribute to their team’s key results, plus one learning or behaviour goal where the work is complex.
  4. Weeks 4, 8 and 11: check in. Use 1:1s to update progress and record any OKR that is dropped or changed, with the reason. This is where the CIPD finding on feedback earns its keep.
  5. Week 13: grade OKRs in the open. Team grades are shared and discussed as data for next quarter, not as verdicts on people.
  6. Week 13: write the contribution note. Managers add a short note per person: what they did to move the key results, what they did outside them, and how they worked. These notes feed the next formal review.
  7. At review time: rate against the bar, then calibrate. Use the scorecard and the quarter’s notes, then run calibration across managers. The performance review cycle checklist covers the review steps end to end.

Not sure where your current set-up leaks? The two-minute performance health check scores goal alignment alongside standards, bias and admin load.

Bottom line

OKRs and performance reviews are different tools with different jobs. OKRs set direction for teams and should be ambitious, public and graded without fear. Reviews judge individual contribution against a consistent bar and should be calibrated before they drive any decision. Connect them through evidence of contribution, not by turning an OKR score into a rating, and keep pay downstream of the calibrated review. For the wider system these two sit inside, see the performance management framework.

FAQ

Should OKRs be used in performance reviews?

Yes, as evidence rather than as the score. A review can draw on what someone contributed to their team’s key results, how difficult those results were and what changed during the quarter. Google’s re:Work guidance treats OKRs as a summary of what a person worked on, not a full evaluation, so the rating should also cover behaviours, skills and work that no OKR captured.

Should OKRs be tied to bonuses or pay?

Not directly. If an OKR grade sets a bonus, people will set goals they know they can hit, and stretch disappears. The What Matters summary of Andy Grove’s approach describes OKRs as separate from compensation. Where pay is linked to performance, the safer route is through a calibrated review rating and written pay criteria, with OKR contribution as one input to that rating.

How often should OKRs be reviewed?

OKRs are usually set and graded quarterly, with progress checked during the quarter, often at the midpoint or in regular 1:1s. Google’s re:Work guide notes that some teams revisit them a few times a quarter to drop objectives that will clearly not happen and add effort where results are borderline. In many companies, formal performance reviews run less often, once or twice a year.

Can OKRs replace performance reviews?

No. OKRs say little about how someone worked, how they developed or how their contribution compares with the bar for their role, and they are mostly owned by teams rather than individuals. A company that drops reviews still needs some way to make fair promotion, pay and development decisions, and OKR grades alone are not designed for that.

Sources

About the author: Ash Taherian is a Product Sales Executive at Revolut People, based in Dublin, working with People and Finance leaders on goal setting, review cycles and calibration.

Disclosure: I work for Revolut People, which sells performance management software, so read the product section with that in mind. This post is general information, not HR or legal advice for your organisation.