How to Write Good OKRs: Objectives and Key Results That Work

A good OKR pairs one inspiring objective with two to four key results that prove you got there. The objective says where you're going in plain words, and the key results are measurable outcomes, not a list of tasks.

That sounds simple. In practice most OKRs drift into to-do lists with a motivational title. This guide shows how to write ones that actually steer a team.

What OKRs Are

OKR stands for objectives and key results. The approach began at Intel under Andy Grove, who built it into how the company set goals. John Doerr learned it there and later brought it to Google and many other companies. His book, Measure What Matters, made it popular well beyond Silicon Valley.

An objective is a short, motivating statement of what you want to achieve. It should be qualitative and memorable. A key result is a measurable signal that the objective has been met. Each objective usually has two to four of them.

The magic is in the gap between the two. The objective gives people a reason to care. The key results stop anyone pretending they succeeded when they didn't.

When to Choose OKRs

Choose OKRs when you need goals that say what success looks like and leave teams free to decide how. They work best when you trust teams to find the route and you want them aligned on the destination.

They're a poor fit when the work is fixed and fully specified. If a team's job is simply to deliver a known list, OKRs add ceremony without adding focus. They also struggle in organisations that punish missed targets, because people quietly set safe goals.

How to Write OKRs, Step by Step

1. Start with one problem worth solving. Ask what would make the biggest difference to customers or the business this quarter.
2. Write the objective in plain, motivating words. Aim for something a team would happily read aloud. "Make first-time buyers feel confident" beats "Improve conversion funnel performance".
3. Add two to four key results. Each should be a measurable outcome with a starting point and a target. Use "from X to Y" wherever you can.
4. Test each key result. Ask whether you could hit it while the objective still failed. If you could, rewrite it.
5. Let teams decide how. The initiatives and features come later, from the people doing the work.
6. Check progress regularly. A short weekly or fortnightly look keeps OKRs alive. Quarterly-only reviews let them fade.
7. Score honestly at the end. Many teams use a scale from 0 to 1. On a stretch goal, around 0.7 is often a good result.

A Worked Example

Imagine an online furniture shop. Plenty of people browse, but first-time buyers often return items or call to check sizes before ordering. The team's first draft looked like this.

Objective: Launch the new product pages.
Key results: Redesign the page template. Add 3D views. Publish a new size guide.

Every line is a task. The team could tick all three and nothing would change for customers. Here's the rewrite.

Objective: Make first-time buyers feel confident.
Key results:

• First-month returns from first-time buyers fall from 18% to 10%.
• Calls about sizing drop by a third.
• Buyer confidence score in the post-purchase survey rises from 6 to 8 out of 10.

Now the team is free to try anything that moves those numbers. The 3D views might help. So might better photos, a fit promise or a short video. The key results tell everyone whether the work mattered.

Common Mistakes

Key Results That Are Really Tasks

"Launch the new app" is a task. Ask what the launch is meant to change, then measure that.

Too Many OKRs

Five objectives with five key results each is a strategy document, not a focus. One to three objectives per team is plenty.

Setting and Forgetting

OKRs written in January and opened again in April teach people they don't matter. Build a short regular check-in into the team's rhythm.

Tying OKRs to Pay

When bonuses depend on scores, people set goals they know they'll hit. That kills the stretch OKRs are meant to encourage.

No Baseline

A key result without a starting point can't show progress. If you don't know the current number, the first piece of work is to measure it.

How OKRs Compare with Related Tools

OKRs sit comfortably alongside a north star metric. The north star is the one long-term measure of value customers get. OKRs are the quarterly goals that move it.

They also pair well with team objectives in Marty Cagan's sense, where teams are given problems to solve rather than features to build. OKRs are a neat way to express those problems.

Finally, a now, next, later roadmap can show which problems you're tackling. OKRs then describe what success on each one looks like.

A Quick Checklist Before You Share Them

• Could a new joiner understand the objective without a briefing?
• Does every key result have a starting point and a target?
• Could you hit the key results while the objective still failed? If so, rewrite them.
• Is every key result an outcome rather than a piece of work?
• Does the team believe the goal is a stretch but possible?
• Do you know when and how you'll check progress?

If you can answer yes to each one, the OKR is ready. If not, it's worth another twenty minutes now. A weak OKR costs a whole quarter of focus later.

Bringing It Together

Good OKRs are short, honest and owned by the people doing the work. Write an objective people care about, prove it with a few outcome measures and check in often. If you'd like a structured way to draft and score them, the Product Field Kit includes a fillable OKRs worksheet with a built-in scoring calculator.

Questions People Ask

← More Field Notes