Skip to content
Merkhet Advisory
Menu
All articles

Articles

When to use OKRs, and when not to

Many companies start OKRs for one reason: other companies use them. Google uses them. A competitor started last year. The board read a book.

That is a weak reason. And it is expensive. When OKRs don't fit, the company spends months on templates and meetings. Teams learn that goals are a reporting exercise. And the next time leaders try to change how goals work, people are already tired of it.

Most of what I know about OKRs comes from the mistakes I made implementing them. The biggest one was simple: I started OKRs before asking if they fit. The templates were fine. The fit was wrong.

So this article does not ask "how do we write OKRs?" It asks an earlier question: should we use them at all, right now? If you already run OKRs and they have stalled, read Before you fix your OKRs, check they fit.

What OKRs are built to do

An OKR has one job: to change something important in a fixed time, usually one quarter. When the change is done, the OKR ends.

That job only works when four conditions are true. If one is missing, the OKR still looks fine on paper. But nothing changes.

  1. 1

    A real change

    Something important must move in the next 3 months. Not only “keep things running”.

  2. 2

    Choice

    Leaders agree on 3 to 5 priorities and say “no” to the rest.

  3. 3

    Freedom on tactics

    Leaders agree the goal. Teams choose how to reach it.

  4. 4

    A rhythm, not a reward

    Progress is checked every week or two, and OKRs stay separate from pay.

Figure 1. The four conditions OKRs need. Each question in the test below checks one of them.

Take the five-question test

Answer for your company as it is today, not as you want it to be. Nobody else sees your answers.

Are OKRs right for you now?

Tap yes or no for each question.

  1. 1Do you need to change something important in the next 3 months, not only keep things running?
  2. 2Can your leaders agree on 3 to 5 priorities, and say “no” to the rest?
  3. 3Can your teams choose how they reach a goal, not only follow orders?
  4. 4Can you check progress every week or two, and remove blockers?
  5. 5Can you keep OKRs separate from pay and bonuses?

Answer the five questions to see your result.

You will get one of three results: OKRs can work, fix the gaps first, or use a simpler tool.

Mostly yes? OKRs can work for you. Mostly no? Fix those points first, or use a simpler tool. A "no" is not a failure. It tells you where to start.

Two questions decide most cases

The test has five questions, but most decisions come down to two. Does something important need to change? And is the organisation ready to work in the OKR way? Put them together and you get a simple map.

A two by two map. Horizontal axis: how much must change this quarter. Vertical axis: how ready the organisation is. Top right: use OKRs. Top left: use KPIs. Bottom right: fix readiness first, with an action plan if urgent. Bottom left: KPIs or SMART objectives.STABLE AND READYUse KPIsThe business is healthy.Track it and protect it.CHANGE AND READYUse OKRsPick 3 to 5 objectives.Check in every week or two.STABLE, NOT READYKPIs or SMARTKeep it simple. Clear targetswith owners and dates.CHANGE, NOT READYFix readiness firstIf it is urgent, run a shortaction plan meanwhile.How much must change this quarterLittleA lotHow ready the organisation isNot readyReady
Figure 2. Only one corner of the map says “use OKRs”. That corner is where OKRs pay back the effort they take.

Four cases where a simpler tool is better

Here are the four situations I see most often where OKRs are the wrong answer, at least for now.

  • Case 1

    The work is stable and routine

    A warehouse ships 98% of orders on time, every month. Nothing needs to change. Writing an OKR here only adds meetings.

    Use instead: KPIs. Track them, and act when one drops.

  • Case 2

    The company is in a crisis

    A key customer leaves and cash is tight. You don’t need an inspiring objective for the quarter. You need ten actions, with owners and dates, checked every few days.

    Use instead: A short action plan with clear owners.

  • Case 3

    Leaders want OKRs for bonuses

    If OKR scores decide pay, people set only safe targets. Reporting a red status becomes risky, so problems stay hidden until it is too late.

    Use instead: Don’t start yet. Separate OKRs from performance pay first.

  • Case 4

    The team is small and the goal is clear

    A team of twelve has one big goal: launch the new product by March. Everyone already knows it. A full OKR process is more than they need.

    Use instead: One or two SMART objectives with a date.

The third case deserves its own article: Why OKRs should not decide bonuses.

Where your OKRs should come from

KPIs and OKRs are not rivals. They work together. A KPI tells you if the business is healthy. An OKR fixes or changes something important, and then it ends.

The best source for your next OKR is often a KPI that has dropped below its normal level. Here is a simple example from a support team.

Line chart over 27 weeks. Requests answered within 24 hours stay near 90 percent for 8 weeks, drop to 70 percent by week 13, then rise to 92 percent during a one-quarter OKR, and the measure goes back to being a KPI.OKR QUARTER100%90%80%70%60%Drops to 70%: write an OKRBack to 92%Normal level: track as a KPIWk 1Wk 13Wk 27Example KR: requests answered within 24 hours go from 70% to 92%.
Figure 3. Share of support requests answered within 24 hours (illustrative example). The number is a KPI while it is normal, becomes an OKR when it drops, and goes back to being a KPI when it recovers.

"Not ready" does not mean "never"

If your test showed some "no" answers, that is useful. Most readiness gaps can be fixed. Leaders can learn to agree on fewer priorities. A team can build a check-in habit in a few weeks. A company can move bonuses to a separate review.

The order matters. Fix the gaps first, then start OKRs. Starting first and fixing later is how most OKR programmes fail.

A good first step for any leadership team is to learn how OKRs are meant to work before deciding. The free OKR Foundation course covers this in about 75 minutes, in 9 short lessons you can take at your own pace.

  1. Step 1 · Measure

    Check the fit

    Are OKRs right for this company, its industry and its culture? Or would KPIs or SMART objectives do the job better?

  2. Step 2 · Align

    Check readiness

    Are leaders ready to choose, to give teams freedom on tactics, and to keep OKRs away from pay? Close the gaps.

  3. Step 3 · Build

    Then start

    Train leaders on the foundations, start at the top with a few objectives, and check in every week or two.

Figure 4. Measure. Align. Then build. The right framework starts with the right diagnosis.
OKRs are a tool. Choose them because they fit, not because they are popular.

Not sure where your company stands? The free 10-minute assessment shows whether OKRs, a hybrid, or KPIs with quarterly priorities suit you, and your three biggest gaps to close first. No sign-up needed.