← The library

Goal setting / focus & alignment

OKRs (Objectives & Key Results)

OKRs pair a qualitative Objective — where we are going and why it matters — with three-to-five measurable Key Results that prove arrival.

Scored transparently and set on a quarterly rhythm, they exist to focus effort and expose progress; used as a performance-rating input, they collapse into sandbagging.

Problem
Goal setting / focus & alignment
Altitude
Team to enterprise
Effort to run
Moderate
Evidence base
Established

Theory & origin

Andy Grove built the method at Intel from Drucker’s management-by-objectives, stripping MBO’s annual, top-down, pay-linked machinery down to a fast, transparent cadence. John Doerr carried it to Google in 1999, and Measure What Matters made it orthodoxy. The design principles carry the value: fewer objectives than feels comfortable, key results as evidence not tasks, ambition calibrated so ~0.7 is success, and — the rule most organizations break first — decoupled from compensation, because the moment OKRs price bonuses, everyone negotiates for targets they have already hit.

Explore the model

How a consultant runs it

  1. 01 Cut the list first: three objectives per team maximum — an OKR set with nine objectives is a task list wearing a costume.
  2. 02 Rewrite key results as outcomes with numbers ("activation 22% → 35%"), not activities ("launch the campaign").
  3. 03 Calibrate ambition openly: 0.7 is success, 1.0 across the board means targets were sandbagged.
  4. 04 Install the rhythm — weekly check-ins on confidence, quarterly scoring — because OKRs die of neglect, not of design.
  5. 05 Keep them out of the bonus formula, and say so in writing; it is the first thing a new CFO will try to change.

When to use

  1. 01 Focusing an organization that is busy everywhere and progressing nowhere
  2. 02 Making cross-team dependencies visible through transparent, public objectives
  3. 03 Replacing an annual MBO ritual that measures activity instead of outcomes

When not to use

  1. 01 Tied to compensation — the single most common implementation failure, and fatal
  2. 02 For run-the-business work: keeping the lights on belongs in health metrics, not aspirational objectives
  3. 03 In cultures unwilling to score honestly in public; secret or inflated scoring is worse than no OKRs

Worked example

A 60-person product company runs quarterly planning that produces 40 "priorities" and ships late on all of them. OKR reset: each of four teams gets a maximum of two objectives, key results rewritten from activities to outcomes, scores published in the all-hands. First quarter scores average 0.5 and two teams discover their key results were dependencies on each other — visibility the old system never produced. By the third quarter the average sits at 0.68, cycle time on top priorities has halved, and the exec team kills eleven zombie projects the OKR lens exposed as belonging to nobody’s objective.

Common pitfalls

  1. 01 Linking OKRs to bonuses and converting ambition into negotiation
  2. 02 Key results that are re-labelled tasks, scoring effort instead of outcomes
  3. 03 Set-and-forget: written in planning week, rediscovered in scoring week
  4. 04 Cascading mechanically down the org chart until frontline OKRs are fragments nobody chose

Sample deliverable

One real engagement, end to end — watch the numbers travel from raw input, onto the chart, into the artifact.

Quarter scorecard — Digital Banking growth team

Input — raw data

  • KR1 — activation 22% → 35%hit 33%
  • KR2 — trial-to-paid 8% → 12%hit 10.5%
  • KR3 — churn 3.1% → 2.4%hit 3.0%
  • KR4 — NPS 31 → 40hit 38
↓

Process — mapped

Key results are scored 0-1.0 in the open; the pattern calibrates ambition and exposes blocks

↓
OutputDeliverable

Quarter scorecard — Digital Banking growth team

  • Average0.60, healthy ambition
  • KR3 blockedchurn owned by no one
  • Next quarterchurn gets its own objective

Sources

Next in the library RACI Matrix →