A tape measure for performance

OKRs vs. performance management – Why you need both

While the performance management process appears to serve a similar purpose as Objectives and Key Results (OKRs), its role differs fundamentally.

OKRs (Objectives and Key Results) are a simple tool to create alignment and engagement around measurable goals and one of the keys to successful implementation of OKRs is to separate OKRs from compensation and promotions. Rick Klau, then an executive at Google, is cited in John Doerr’s book:

OKRs are not synonymous with employee evaluations. OKRs are about the company’s goals and how each employee contributes to those goals. Performance evaluations –  which are entirely about evaluating how an employee performed in a given period  –  should be independent from their OKRs.
Rick Klau (Google)1

Most larger organizations already have elaborate processes for goal setting in place, based upon which the annual performance review is conducted. This process goes by different names but features prominently in the human resources (HR) toolkit of most organizations.

The objective of this process is to support career development and encourage the continuous improvement of employees’ skills, competences and contributions to the organization. It is touted as

  • an opportunity to celebrate achievements,
  • reflect on where to improve,
  • get feedback on past performance and, finally,
  • to realign with the leader on expectations and career development.

As a consequence, the performance management process differs from OKRs both in its purpose and scope, as outlined below:

AspectPerformance managementObjectives and Key Results (OKR)
Scope
  • Individual performance
  • Organization
Purpose
  • Assess individual performance
  • Focus and align organization

For this reason, John Doerr suggests to use OKRs only as one input among many for performance reviews.

It is not a legal document upon which to base a performance review but should be just one input used to determine how well an individual is doing.
Andy Grove (Intel)1

In my own implementations of OKRs, I have always defined the OKRs independently from the individual. This means that if a new person had inherited the role of someone, the OKRs would not have changed. But their performance evaluation very well might have been quite different.

In a nutshell

OKRs and performance management serve different purposes. OKRs align the organization around shared goals. Performance management assesses individuals. Keeping them separate lets each do its job.

When OKRs are tied to compensation, people set conservative targets they know they can hit. The quarterly review becomes a negotiation about what counts as “achieved” rather than an honest assessment of where the organization stands. Separating OKRs from performance reviews removes this incentive. People set targets based on what the organization needs, not on what is safe to promise.

This also changes how teams handle missed key results. In a system where OKRs feed directly into bonuses, a missed key result is a career risk. In a system where OKRs are separate, a missed key result is information: the market shifted, the estimate was wrong, or the dependency didn’t hold. Teams can discuss what happened and adjust, rather than scrambling to redefine success after the fact.

Performance reviews typically run on annual cycles. OKRs are set quarterly. This difference matters because an organization that reviews its direction four times a year can correct course when conditions change — a quarter where a key result proves unreachable becomes a signal to reassess, not a failure to explain away twelve months later.

The quarterly cadence also makes OKRs lighter weight. Setting three to five objectives every quarter takes less overhead than the elaborate annual goal-setting process most organizations run through HR. Teams spend less time on the process and more time on the work.

Performance goals belong to individuals or, at most, to a single team. OKRs can span teams. When two departments share a key result — say, reducing customer onboarding time from fourteen days to three — they have a concrete reason to coordinate. Neither team can claim success alone, which forces the kind of collaboration that org charts and good intentions rarely produce on their own.

This shared ownership also surfaces dependencies. If the engineering team’s OKR depends on a data migration that the platform team owns, that dependency is visible in the OKRs themselves, not buried in a project plan that only one team reads.