---
name: objectives-and-key-results
description: Use when asked to write OKRs (Objectives and Key Results) — an inspirational, team-actionable objective paired with measurable key results — grounded in joelparkerhenderson/objectives-and-key-results, including known real-world pros/cons, as distinct from KPIs alone (see key-performance-indicators).
---

# Objectives and Key Results

OKR (Objectives and Key Results) is a lightweight goal-setting method:
the **Objective** states what you want to achieve; the **Key Results**
state how you'll know you're making progress. A simple template: *"I will
(Objective) as measured by (this set of Key Results)."*

## History

Invented at Intel and championed by CEO **Andy Grove** (in his book
*High Output Management*), later popularized further by venture capitalist
**John Doerr** (*Measure What Matters*). Widely used at Intel, Google,
Microsoft, Twitter, and Uber, among others.

## What makes a good objective

- **Inspirational** — provides a real sense of meaning and progress; skip
  anything as unambitious as "a small percentage gain."
- **Measurable by the team independently** — no excuse like "that other
  team didn't measure it."
- **Actionable by the team independently** — no excuse like "marketing
  didn't market it"; the objective has to be genuinely the team's own.
- **Relatable** — plain language the team actually uses, not abstract
  jargon.
- **Timely** — a clear sprint toward the goal, doable within roughly a
  month or a quarter; if it takes a year, it's probably strategy or
  mission, not an OKR-scale objective.

## What makes a good key result

Ask: "how would we know if we met our objective?" Typically 1–3 key
results per objective, always quantified — a key result is not a general
or subjective planned action. Metrics commonly draw from growth,
engagement, revenue, performance, or quality. Andy Grove's own bar: "to
be useful a Key Result must contain very specific wording and dates, so
that when the deadline arrives, there is no room for ambiguity."

## Ground rules

- OKRs are **transparent by default** — visible to the whole organization
  (rare exceptions for maximum-security/secrecy work). Visibility helps
  teams understand what others are doing and supports cross-team
  alignment.
- OKRs **nest**: if a subordinate's objectives are met, the supervisor's
  objectives should be met as well — the hierarchy should be a genuine
  cause-and-effect chain, not independently-invented goals at each level.
- OKRs are meant to **pace a person** (Grove's own framing: "a stopwatch
  in their own hand"), not to serve as the legal basis for a performance
  review on their own — mechanically grading someone against a rigid OKR
  ignores legitimate emerging opportunities the OKR didn't anticipate.
- Aim for roughly **70% achievement** as a good outcome on a genuinely
  stretching OKR — 100% attainment routinely suggests the objective
  wasn't ambitious enough.

## How OKRs differ from strategy, KPIs, and task lists

- **Strategy vs. objectives** — a strategy explains *how* a target will
  be reached; an objective is only the target itself. An organization
  can state an objective with no real strategy behind it, which leaves a
  hundred possible paths and no actual direction — see
  [[strategic-balanced-scorecard]] for a framework that supplies
  the strategic layer OKRs alone don't.
- **[[key-performance-indicators]]** — KPIs are often used
  *alongside* OKRs as the next layer down: an objective's key result
  might be hard to hit directly, so KPIs track the constituent numbers
  that drive it, letting a team find the right combination empirically.
- **Task lists / work boards** — OKRs deliberately capture only the top
  few goals and progress points; task lists capture a full enumeration of
  work items. OKRs answer "what matters most"; task lists answer
  "what needs doing."

## Real-world tradeoffs (both sides, honestly)

OKRs have a large, genuinely mixed body of practitioner experience.
Reported benefits: forced alignment conversations that surface real
cross-team disconnects, visible priorities that help new hires orient
quickly, and a shared vocabulary for saying no to work outside the
current quarter's focus. Reported failure modes: OKRs imposed top-down
with no bottom-up input become disconnected from day-to-day reality;
tying compensation/performance review too tightly to OKR attainment
incentivizes sandbagging or gaming; and OKRs treated as quarterly
theater — written once, promptly ignored, backfilled with a plausible
narrative at quarter-end — provide none of the intended benefit while
still costing the time to write them. Whether OKRs work in a given
organization depends heavily on genuine leadership commitment to the
system, not on the OKR format itself being inherently good or bad.

## Common pitfalls

- **Non-measurable key results** — "improve X" without a number and
  timeframe fails Doerr's "as measured by" test and invites disputed,
  unfalsifiable claims of success later.
- **Objectives set with no strategic anchor** — see the "strategy vs.
  objectives" distinction above; without it, OKRs measure activity
  against an arbitrary target rather than genuine strategic progress.
- **Tying OKR attainment directly to compensation/performance ratings** —
  a well-documented way to turn OKRs into a gaming exercise rather than
  an honest stretch-goal tool.
- **Setting OKRs and never revisiting them until quarter-end** — OKRs are
  meant to be looked at regularly (many teams: weekly), not written once
  and rediscovered at review time.
- **Purely top-down or purely bottom-up OKR-setting** — top-down alone
  disconnects from operational reality; bottom-up alone isn't real
  strategic planning; a cycle combining both (context down, proposed key
  results up, then reconciliation) is what most mature practitioners
  report actually working.

## Learn more

- [joelparkerhenderson/objectives-and-key-results](https://github.com/joelparkerhenderson/objectives-and-key-results) — the source for this skill, including a large curated set of real practitioner pros/cons/cautions.
- Andy Grove, *High Output Management*; John Doerr, *Measure What Matters*
- [[key-performance-indicators]], [[smart-criteria]], [[strategic-balanced-scorecard]] for the adjacent goal-setting and measurement frameworks OKRs connect to.