What Are OKRs?
In the late 1970s, Andy Grove — then president and later CEO of Intel — developed a management system he called "iMBOs" (Intel Management by Objectives). It was a refinement of Peter Drucker's Management by Objectives, but with a critical addition: measurable key results attached to every objective. Grove believed that objectives without concrete metrics were just wishes. The system he built at Intel was simple in structure but rigorous in practice: define what you want to achieve, then define exactly how you will know you achieved it.
The framework might have stayed inside Intel if not for John Doerr. A former Intel engineer, Doerr joined the venture capital firm Kleiner Perkins and in 1999 introduced OKRs to a 40-person startup called Google. Larry Page and Sergey Brin adopted the framework, and Google has used OKRs every quarter since. Doerr later published "Measure What Matters" in 2018, which brought OKRs into the mainstream. Today, organizations from Spotify to the Gates Foundation use the framework.
The core idea is two parts working together. An Objective is what you want to achieve — qualitative, ambitious, and ideally motivating. Key Results are how you measure whether you are getting there — quantitative, specific, and time-bound. Each Objective typically has two to five Key Results. The discipline is in the pairing: the Objective provides direction and inspiration; the Key Results provide accountability and clarity.
The Two Components
Objectives
What you want to achieve. Qualitative, ambitious, action-oriented. Should be motivating when spoken aloud. One sentence. No numbers.
Key Results
How you measure progress. Quantitative, specific, time-bound. Outcomes, not activities. 2-5 per Objective. Scored 0.0 to 1.0 at end of cycle.
What Good OKRs Look Like
Example: Product Team
Objective: Become the most trusted option for mid-market buyers in our category.
KR1: Increase NPS from 32 to 50 among mid-market accounts.
KR2: Reduce average onboarding time from 14 days to 5 days.
KR3: Achieve 3 unsolicited customer case studies published this quarter.
Example: Engineering Team
Objective: Ship a platform that engineers are proud to build on.
KR1: Reduce P1 incidents from 8 per month to 2 or fewer.
KR2: Increase deployment frequency from weekly to daily.
KR3: Achieve 90% code coverage on the three most critical services.
Notice what these examples share: the Objectives are directional and aspirational. The Key Results are specific and measurable. There is no ambiguity about whether you hit them. And the Key Results are outcomes (fewer incidents, higher NPS) rather than activities (write more tests, send more surveys).
The Alignment Problem OKRs Solve
The real power of OKRs is not in goal-setting — any framework can set goals. It is in alignment. In most organizations, teams operate in silos. Marketing is optimizing for leads. Sales is optimizing for closed revenue. Product is optimizing for features shipped. Engineering is optimizing for uptime. Everyone is busy. No one is pulling in the same direction.
OKRs create a visible hierarchy of priorities. Company-level OKRs define the three to five things that matter most this quarter. Department OKRs connect to those company priorities. Team OKRs connect to department priorities. When every team can trace their objectives back to a company-level goal, you get alignment without micromanagement. When OKRs are public across the organization, you get transparency without endless status meetings.
This is why OKRs are fundamentally a leadership tool, not a project management tool. They answer the question every team member has but rarely asks aloud: "Does what I am working on actually matter to this organization right now?"
When to Use OKRs
- Quarterly planning. OKRs work best on a quarterly cadence for team-level goals. Annual OKRs at the company level provide direction; quarterly OKRs at the team level provide focus and accountability.
- After rapid growth or reorganization. When teams have been added, restructured, or reoriented, OKRs re-establish shared priorities and make it explicit who is responsible for what.
- When output is high but impact is low. If your team is shipping a lot but the business metrics are not moving, OKRs force the conversation: are we measuring the right outcomes? Are we working on the right things?
Common Mistakes
Writing Key Results as Tasks
"Launch the new onboarding flow" is a task. "Reduce new-user drop-off from 40% to 20%" is a Key Result. The difference is crucial. Tasks describe activity; Key Results describe impact. If you ship the new onboarding flow and drop-off does not improve, the task is done but the Key Result is missed — which tells you something important about your assumptions. Leaders who let task lists masquerade as Key Results lose the diagnostic power of the framework.
Setting Too Many OKRs
If everything is a priority, nothing is. Grove recommended three to five Objectives per team per quarter, with two to five Key Results each. More than that and the framework collapses under its own weight. Teams lose focus, reviews become perfunctory, and OKRs turn into a bureaucratic exercise rather than a focusing tool. The discipline of OKRs is as much about what you choose not to pursue as what you commit to.
Tying OKRs Directly to Compensation
This is the mistake that kills OKR culture fastest. When Key Results are directly tied to bonuses or performance reviews, teams sandbag. They set conservative targets they know they can hit. The entire point of stretch goals — aiming for transformative results with a 70% achievement expectation — collapses. Grove and Doerr both explicitly warned against this. OKRs should inform performance conversations but should not be the performance conversation.
The OKR Rhythm
A Quarterly Cadence That Works
- Week 1: Set OKRs. Leadership finalizes company OKRs. Teams draft their OKRs and verify alignment with company priorities. Cross-functional dependencies are identified and negotiated.
- Weeks 2-6: Execute with weekly check-ins. Each team reviews Key Result progress for 10 minutes during their weekly standup. Red/yellow/green status on each KR. No lengthy presentations — just honest assessment of progress and blockers.
- Week 7: Mid-quarter review. Formal review of all OKRs. Are Key Results on track? If not, what needs to change — the approach, or the target? This is the moment to adjust, not wait until the end.
- Weeks 8-12: Push to close. Final execution sprint. Teams know exactly which Key Results are within reach and which require extraordinary effort.
- Week 13: Score and reflect. Each Key Result scored 0.0 to 1.0. Average of 0.6-0.7 across the set indicates appropriately ambitious targets. Below 0.4 suggests either poor execution or unrealistic goals. Above 0.9 suggests sandbagging. The retrospective informs next quarter's goal-setting.
Putting It Into Practice
Consider a head of customer success at a growing SaaS company. Churn is rising. The CEO has set a company OKR: "Build a customer base that grows through retention, not just acquisition." The CS leader translates this into her team's OKR.
Objective: Make our existing customers measurably more successful this quarter. KR1: Reduce monthly churn from 4.2% to 2.5%. KR2: Increase product adoption score (features used per account) from 35% to 55%. KR3: Achieve a 4.5+ average rating on quarterly business reviews from the top 20 accounts.
Each Key Result is measurable, time-bound, and connected to the company objective. The team can check progress weekly. If KR1 is lagging but KR2 is ahead, it raises an important question: are customers using more features but still churning? That diagnostic insight only emerges because the Key Results are specific enough to disagree with each other.
Cabinet coaches leaders through building and maintaining OKR systems like this — from writing your first set of Objectives to running the quarterly cadence that makes OKRs a living management practice rather than a one-time planning exercise.