OKRs explained: how objectives and key results work

An OKR pairs one objective, a qualitative statement of where you want to go, with two to five key results, measurable outcomes that show you got there. The objective inspires; the key results keep it honest. Teams usually set them for a quarter and check progress every week or two. The OKR generator drafts the format and scores progress, and this guide explains how to write OKRs that actually change what people work on.

What an OKR is

An OKR has three layers, and keeping them separate is most of the skill:

  • Objective โ€” short, memorable and qualitative. No numbers: "Make onboarding effortless", not "Increase activation 20%".
  • Key results โ€” measurable outcomes, ideally written "from X to Y" so the starting point is visible. If you cannot score it at the end of the quarter, it is not a key result.
  • Initiatives โ€” the projects and tasks you think will move the key results. They are bets, and you can change them mid-quarter without changing the OKR.

The generator nudges you towards this structure: it flags an objective that contains a number, asks for two to five key results, and checks that each key result has a number in it.

Where OKRs came from

OKRs grew out of Peter Drucker's management by objectives. Andy Grove adapted the idea at Intel in the 1970s and described it in his 1983 book High Output Management. John Doerr learned the method at Intel, later introduced it to Google in 1999 while an investor at Kleiner Perkins, and wrote about it in Measure What Matters (2018). Google's adoption is the main reason the method spread through technology companies and beyond.

A worked example

Imagine a six-person customer support team at a small accounting-software company, planning their next quarter:

  • Objective: Customers get answers so fast they barely notice they had a problem.
  • KR1: Cut median first-response time from 9 hours to 2 hours.
  • KR2: Raise customer satisfaction on closed tickets from 82% to 90%.
  • KR3: Reduce invoicing-related tickets from 300 to 150 a month.
  • Initiatives: a shared rota covering the morning peak; rewrite the five most-viewed help articles on invoicing; a weekly review of the ten slowest tickets.

Notice that KR3 forces collaboration with the product team, because the best way to halve invoicing tickets is to fix what confuses people. Good key results often cut across teams like that, and if responsibilities get blurry, a RACI matrix sorts out who owns what.

How scoring works

Each key result is scored from 0.0 to 1.0 by how far you moved from the starting value towards the target. The generator works it out as (current โˆ’ start) รท (target โˆ’ start), rounded to one decimal place. Halfway through the quarter, if median first response is 4 hours, KR1 scores (4 โˆ’ 9) รท (2 โˆ’ 9) = 0.7. Doerr describes two kinds of OKR. Aspirational ones are deliberately ambitious, and about 0.7 counts as success; hitting 1.0 every time means the targets were too safe. Committed ones, such as shipping a promised feature, are expected to reach 1.0. Say which kind each OKR is, or people will argue about whether 0.7 is good news.

Common mistakes

  • Key results that are tasks. "Launch the new help centre" is an initiative. The key result is what the launch should change.
  • Too many OKRs. Three objectives with three key results each is already a lot for one team in a quarter.
  • Tying OKRs to bonuses. Doerr advises keeping OKRs separate from pay, because people stop setting ambitious targets once a bonus depends on hitting them.
  • Never looking at them. OKRs set in a planning workshop and opened again at quarter end are decoration. A short fortnightly check-in is enough.

When not to use OKRs

OKRs add overhead: drafting, aligning, scoring and reviewing. For one person or a very small team with a single clear goal, a well-written SMART goal does the job with less ceremony; SMART goals explained: what each letter means and how to write one covers that, and the SMART goal generator drafts one in a minute. OKRs also suit improvement work better than steady operational work: if a team's job is to keep a service running, health metrics with agreed thresholds are clearer than an inspiring objective. Once OKRs are set, the day-to-day work still needs somewhere to live, and a kanban board is a simple place for the initiatives.

Sources and further reading

The claims in this guide rest on these references, which were checked when the guide was last updated. Spotted an error? The contact page says how to report it.

  1. Objectives and key results โ€” Wikipedia (Grove, Doerr, scoring)
  2. What is an OKR? โ€” What Matters
  3. Set goals with OKRs โ€” Google re:Work

Try the tool

Frequently asked questions

What does OKR stand for?

Objectives and Key Results: a qualitative objective paired with two to five measurable key results.

Who created OKRs?

Andy Grove developed the approach at Intel in the 1970s, and John Doerr introduced it to Google in 1999.

What is a good OKR score?

For ambitious, aspirational OKRs about 0.7 is usually treated as success; committed OKRs are expected to reach 1.0.

How many key results should an objective have?

Usually two to five. Fewer may not define success; more dilutes focus.