OKRs (Objectives and Key Results)

OKRs (Objectives and Key Results) is a quarterly or annual goal-setting framework with two parts:

• Objective: a qualitative, ambitious statement of intent — "Make Google synonymous with search"
• Key Results: 3–5 quantitative measures that verify whether the Objective was achieved — "Hit 100M queries per day", "Build a credible ad model", etc.

OKRs were developed by Andy Grove at Intel in the 1970s and brought to Google by John Doerr in 1999. They have since become the default goal framework at fast-moving tech companies (Google, LinkedIn, Twitter, Adobe, Atlassian, many more).

The core OKR rules

• Objectives are aspirational: hitting 70% of an OKR is treated as a success; 100% means the target was set too low
• Key Results are quantitative: each KR must be measurable and verifiable by a third party
• Compensation is decoupled from OKR achievement: tying pay to OKRs immediately incentivizes sandbagging targets, which destroys the ambition mechanism
• OKRs are transparent: published company-wide, not gated by hierarchy

OKRs vs KPIs

OKRs set ambitious direction over a quarter or year. KPIs (Key Performance Indicators) track ongoing operational health. They are complementary — OKRs answer "what are we trying to change?", KPIs answer "is the business operating correctly?". Confusing them produces meaningless goals and ignored health signals.

When OKRs fail

OKRs fail when targets get cascaded mechanically down the org chart (each layer adds compromises until KRs are meaningless), when Objectives proliferate (more than 3 per team diffuses focus), or when KRs become activity-based instead of outcome-based ("ship X features" is not a KR; "drive X metric to Y value" is).

Related

• OKR framework — full framework page with examples
• OKRs vs KPIs — when to use each
• Google's early OKRs — worked example