An Introduction to Objectives and Key Results (OKRs)

I love helping teams to get clear on what they’re doing, why they’re doing it, and how they can get better at doing it while enjoying the journey.

Since I discovered Objectives & Key Results (OKRs) in 2015, they’ve been a huge enabler of this work.

They were developed by Andy Grove at Intel in the early 80s, when the company faced serious competitive pressure. John Doerr later carried the practice into the organisations he invested in — most famously Google in 1999. Today they’re used across every industry. Used well, they drive focus and alignment, and make it obvious how day-to-day work connects to measurable outcomes.

This post is the front door: what an OKR is, how the cycle works, whether you’re ready, and what to try first. For a deeper “why”, see Why OKRs?.

What is an OKR?

In basic terms, an OKR is a big, inspiring objective plus two or three key results the team agrees to use as the measure of success. Key results are not the tactics for getting there. They are how you’ll know you’ve arrived (or how far short you came).

Here’s an illustrative example from a digital health delivery team:

Objective: Clinicians get insights they can act on sooner, with less rework.

Key results — we’ll know we’re getting there if we:

1. Reduce median time from sample received to report issued from 21 days to 10 days
2. Cut the share of reports returned for clarification from 18% to under 8%
3. Raise the percentage of requesting clinicians who rate the report “clear and actionable” from 55% to 80%

Clear direction, three numbers you can argue about in good faith, and nothing about “ship the new portal by Friday.” The portal might still be the right tactic. The OKR keeps you honest about whether it helped.

Marissa Mayer put it bluntly: every key result should contain a number. Stretch matters too. Key results should feel ambitious. Aim for positive, inspiring discomfort.

Many teams aim to start a cycle with roughly 50% confidence they’ll fully hit each KR. At Google, landing around 70% of a stretch key result is often treated as strong performance. Consistently hitting 100% usually means you weren’t stretching.

Those starting numbers are a compass, not a religion. In practice I care more about what happens in the weekly check-in: we talk about confidence, not just progress, because that’s where learning and alignment actually show up.

The cycle: four disciplines

In my work with thousands of people across dozens of organisations, the value of OKRs unlocks when we practise four disciplines together — not if we write a beautiful slide once a year and forget it the following week.

1. Setting — Within a team, draft a small number of real OKRs for a real cycle. Ambition in the objective; measurability in the key results; ownership in the room.
2. Aligning — Surface conflicts early (my Everest, your submarine) through open comparison, gallery walks, festivals, and frank negotiation across teams.
3. Checking in — Weekly for quarterly OKRs (or monthly for annual ones), matched to how fast you need to decide. Confidence, narrative, and next actions beat status theatre. Share widely.
4. Closing — Score the key results, capture lessons, celebrate the stretch, then choose what to keep, drop, or rewrite. The learning matters more than the score.

Expect several quarters of clumsy reps before this feels like a habit. That’s normal. More reps with lighter weight beats a perfect first cycle.

Low confidence and missed targets are data. Is the thing still valuable? Do we have the resources? Were we kidding ourselves? Adjust and go again.

Why not traditional goals?

Traditional long-range planning tends to be brittle. It usually assumes:

1. The problem space is well understood
2. Every step and dependency can be mapped in detail in advance
3. External conditions will stay mostly static
4. The plan will be mostly correct, needing only small tweaks

So teams commit to what they can control — activities — instead of outcomes that actually matter.

Those goals are often set in silos, annually, with thin data, glanced at in January, and barely challenged until December. Box-ticking rarely produces alignment or strong results.

Are you ready for OKRs?

It’s easy to get swept up in a new framework. Not every organisation is ready. Not every team needs OKRs.

Ask yourself:

Do you have a culture of upwards accountability and ownership?

If people are used to being told what to do and how to do it, they’ll struggle to write meaningful OKRs or use them to steer their actions. Most OKRs should be driven from the “bottom up” but aligned from the top down.

Do you have a culture of measurability?

OKRs need observable outcomes and timely data. If you can’t find numbers you trust, instrument first. Existing KPIs are often a good on-ramp.

Is it safe to fail?

Doerr’s line is to “stretch for amazing.” At HSBC, I heard and loved the phrase “the edge of impossibility”. Stretch means you’ll sometimes come up short — and learn more than if you’d played it safe. Leaders must model this culture, celebrate, and really understand the struggles. Without psychological safety, OKRs become a new way to punish people.

For more on outcomes, transparency, and alignment — including how we framed it at Genomics England — read Why OKRs?.

Where OKR adoption goes sideways

A few traps I see constantly:

• Murky shared vision at the top — If senior leaders aren’t clear and in agreement, teams will struggle to deliver via OKRs or any framework.
• Too much, too soon — Rolling OKRs out everywhere at once is a recipe for confusion, low-confidence, and frustration. Adopt in phases — inspect & adapt.
• OKRs as a task list — “Launch the website” is a tactic, not a key result. Ask what change you’ll observe if the launch worked.
• Private OKRs — If only your team can see them, you lose the alignment conversations that make the framework worth the overhead.
• Goal- theatre — Set once, ignore until December. Without check-ins, you have wallpaper. 😭
• 100% hit rate as success — If you always fully achieve every KR, you probably weren’t stretching.
• No closing — Skipping the score-and-learn step wastes the cycle. The retrospective is the point.

Try this week

Pick one team and one short cycle (even six weeks). Write one objective and two key results with numbers. Put them somewhere everyone can see. Run a single 30-minute confidence check-in. What does this conversation enable that you can’t get with a status-report?

Further reading

• Setting OKRs
• Aligning OKRs
• A guide to OKR check-ins
• Closing OKRs
• Free OKR tracking templates
• Why OKRs?

If you want help building the habit properly — workshops, coaching, or a programme with several teams — take a look at my OKR training and coaching, or get in touch.