OKRs

Overview

OKRs trace to Andy Grove, the legendary CEO of Intel, who developed the approach in the 1970s as a more rigorous version of Peter Drucker's Management by Objectives (introduced in The Practice of Management, 1954). Grove described his methodology in High Output Management (1983, Random House) — still one of the best management books written. Grove's version emphasized separating aspirational direction (the Objective) from measurable outcomes (the Key Results), and running the cycle quarterly rather than annually.

John Doerr, a venture capitalist at Kleiner Perkins who had worked under Grove at Intel, introduced OKRs to Google in 1999 when the company had fewer than 40 employees. Google's sustained use — and the companies that followed — made OKRs the dominant goal-setting framework in tech and beyond. Doerr's book Measure What Matters (2018, Portfolio/Penguin) brought the methodology to a broader business audience through Google's story and case studies from Intel, the Gates Foundation, and others.

The structure:

• Objective: qualitative, inspirational, directional — answers "where are we going?" A good Objective is memorable, motivating, and unambiguous in direction. Not a number.
• Key Results: measurable, time-bound, typically 3–5 per Objective — answers "how will we know we got there?" Key Results are outcomes, not activities. Not "launch the product" — "achieve 10,000 active users by end of quarter."

Key design principles that distinguish OKRs from ordinary goal-setting:

• Quarterly cadence: fast enough to stay relevant in a changing environment; slow enough to make real progress
• 0–1 scoring: KRs are graded at period close; 0.7 is considered success — a 1.0 suggests the target was set too conservatively
• Deliberate stretch: OKRs should be aspirational enough that 70% still represents meaningful progress; sandbagging defeats the purpose
• Transparency: OKRs are typically public within the organization, creating alignment and enabling people to see how their work connects to company direction
• Separation from compensation: unlike traditional MBOs, OKRs should not be directly tied to performance reviews — doing so incentivizes sandbagging and undermines the stretch principle

When to Use It

When a client needs to improve organizational alignment, prioritization, or accountability — particularly in organizations where strategy exists but execution is fragmented or unclear. OKRs are most powerful when the organization has a clear strategy and needs to translate it into quarterly execution focus. Less useful when the organization hasn't yet agreed on strategic direction — setting OKRs before that is prioritizing before choosing.

How It Works

1. Set the Objective — the most important thing the team needs to accomplish this quarter, written as a memorable, directional headline. Non-numeric.
2. Define 3–5 Key Results — for each KR, ask: "If we achieve this, would we be confident the Objective was met?" KRs must be measurable (have a number), outcomes-based (not activities), and collectively sufficient to validate the Objective.
3. Check for sandbagging — is each KR achievable at 70% effort? If yes, stretch it. Calibrated right means a genuine push to reach 0.7.
4. Cascade and connect — team OKRs connect to company OKRs; individual OKRs connect to team OKRs. The connection creates alignment without rigid top-down control.
5. Check in regularly — weekly or bi-weekly progress reviews; update confidence scores; surface blockers early. Cadence matters as much as the targets.
6. Score and learn — at quarter end, grade each KR, reflect on drivers of results, and use the learning to set better OKRs next quarter. The retrospective is where most of the value compounds.

Common Pitfalls

• KRs as activities — "launch the new feature" is an output; real KRs measure outcomes: "increase feature adoption to 40% of active users by quarter end"
• Too many OKRs — 10+ OKRs per quarter means nothing is prioritized; 3–5 Objectives with 3–5 KRs each is the upper limit; fewer is usually better
• Sandbagging — setting easily achievable targets to guarantee a good score destroys the framework's value and signals a risk-averse culture
• Tying directly to compensation — Grove and Doerr both explicitly caution against this; it creates powerful incentives to lowball targets and avoid ambitious goals
• Treating OKRs as a reporting tool — OKRs are for alignment and learning, not status updates; if the weekly check-in becomes a recitation of metrics, the system has been bureaucratized

Recommended Books

• Measure What Matters — John Doerr
• High Output Management — Andy Grove
• Radical Focus — Christina Wodtke