OKR (objectives and key results): definition
OKRs, short for objectives and key results, are a way of steering work that combines an inspiring qualitative objective with a few numeric key results showing whether it has been reached. Andy Grove formalized the approach at Intel, and John Doerr spread it, most famously at Google. Teams usually set OKRs each quarter and review them regularly during the period.
Definition written by Alex Dionisio, Senior Product Manager. Updated November 13, 2026.
Why it matters for a PM
Well written, OKRs tie a product team's work to changes in behavior or business results rather than to a list of deliverables. They give a PM a frame for trade-offs: an item that moves no key result has to justify its place. They also make the team's priorities visible to the rest of the company.
Example
An onboarding team sets the objective “New accounts reach their first success quickly.” Its key results: raise the share of accounts creating a first project within a week from 35% to 50%, and halve support tickets during the first seven days. The features to build remain bets, chosen afterward.
What is the difference between OKRs, KPIs and a roadmap?
A KPI is a metric you watch continuously to know whether the business is healthy: conversion rate, retention, support response time. An OKR is a time-boxed push for change: it names what the team wants to move during the period. A KPI can become a key result when the team deliberately sets out to shift it, then go back to being a health metric once the target level holds.
A roadmap answers a different question: which bets the team plans to make, and in what order. In a healthy setup, OKRs come first and the roadmap follows from them. When the order flips, key results end up restating the planned releases, and the whole exercise loses its point.
How do you write a good key result?
A good key result reads like a statement anyone can check once the quarter closes: a metric, a starting value, a target and a deadline. "Improve onboarding" does not qualify; "raise the share of new sign-ups who invite a colleague within fourteen days from 20% to 30%" does. If nobody can measure the starting value today, the first job is to instrument the product, not to pick a target at random.
Watch for side effects too. A key result built on volume can tempt a team to trade away quality; pairing it with a guardrail metric such as churn or satisfaction keeps that in check. Finally, make sure the team can actually influence the number. A key result that mostly depends on another team or on the market demoralizes people more than it rallies them.
Do OKRs suit every team?
The method assumes a certain maturity: reliable data, a readable strategy and leaders willing to let teams choose their means. Without those, it becomes one more reporting ritual. A small company still searching for its market is sometimes better off with a single company objective, revisited more often than once a quarter.
A few questions worth asking before adopting it. Who sets the objectives, and how much comes from the teams themselves? How do team OKRs connect to company OKRs without turning into a rigid cascade? What happens when a key result is missed: does the team draw lessons, or does someone look for a culprit? The answer to that last question often decides whether the approach survives.
How do you track OKRs during the quarter?
A short check-in every week or two is enough: for each key result, the current value, how confident the team feels about hitting the target, and what stands in the way. That rhythm lets the team redirect effort mid-course instead of discovering a miss when it is too late to react. When the quarter closes, an honest score followed by a conversation about what worked and what did not feeds the next set. A middling score on a stretch target is nothing to be ashamed of; it becomes a problem when it repeats and nothing changes.
Key points
- The objective says where you want to go; key results say how you will know you got there.
- A key result measures an outcome, not a task or a delivery.
- Three to five key results per objective are usually enough.
- Ambitious targets are fine, as long as the team knows whether an OKR is a commitment or a stretch goal.
Common mistakes
- Writing key results that are really features to ship.
- Setting so many objectives that everything becomes a priority.
- Tying OKRs directly to compensation, which pushes people to aim low.
- Setting them at the start of the quarter and only looking again at the end.
Go further with Module
The courses and lessons that cover this concept: