The Dark Side of OKR

Why OKR are bad and you should not use them

All things are poisons, for there is nothing without poisonous qualities. It is only the dose which makes a thing poison.

— Paracelsus

OKR are like a hammer that makes everything look like a nail.

Credited for driving Google’s iconic success, OKR became table stakes for setting goals and driving execution. 

Like any tool, they are good for some things and bad for other things. To learn why OKR are good, read one of the thousands articles on internet.

This post is about why OKR are bad.

Memory Refresh: What is OKR

OKR stands for Objective and Key Results. Objective summarizes what you want to achieve, while Key Results describe EXACTLY how you plan to achieve the objective. Here is an example:

OKRs are typically set on a quarterly cadence: defined around beginning of the quarter and evaluated at the end. 

OKRs are typically cascaded: individual OKR align with team OKR which align with department-level OKR which align with company-level OKR.

The Problem

OKR are great when done right. They have net negative effect when used incorrectly. The rest of the article is about how to — or how not to — turn OKR into something much worse than a waste of time.

Here are a few recipes for disaster.

Recipe 1: Destroy Integrity by Measuring Performance with OKR

OKR intend to inspire achievement of greater things. They need to be aspirational. As soon as you start using OKR to measure someone’s performance, you force them to sandbag the aspirations and argue for less ambitious goals. You motivate them to lie about their achievements. Don’t put people in this position.

What if you want to set performance targets and at the same time use OKR framework? Create a special type of OKR, like Google did - they call it “commit OKR”. This is a special and explicitly labeled type of OKR that mean “this is not aspirational - this is something that is expected from you, and if you don’t achieve Key Results — it means you are doing a poor job”. 

Recipe 2: Lose Trust in Leadership by Planning Too Far Ahead

Startups and other dynamic companies often have no idea what will be important in 3 months from now. This is a fact of life. It’s ok. What is not ok is asking people in such companies to set goals for 3 months ahead and attempt to achieve them.

Best case — they will silently drop the OKR and work on what is actually important.

Worst case — they will focus on achieving the OKR even though it’s not what the business needs.

In either case, this will lead to demotivation and loss of trust in OKR — and often in goal-setting in general.

The solution? Either create a shorter OKR cycle — monthly or even weekly — or make it explicitly ok to replace OKR whenever they become irrelevant.

Recipe 3: Sow Chaos with Lack of Context

OKR are an alignment tool. When defining OKR, people and teams look at higher-level OKR and figure out how to contribute.

Asking individuals, teams, or departments to set OKR without providing higher-level OKR is both hypocritical and ineffective. Still, I’ve seen this happening way too often. 

It leads to people buying into and committing to goals that are not aligned with other teams. Best case — it leads to a waste of time. Worse case — it leads to inter-team conflicts and growing silos.

Solution: Define company-level OKR before defining any other OKR. Define team-level OKR before asking for individual OKR.

Recipe 4: Fake Goals with Poorly Defined Key Results

Good Key Results are the key for a good OKR.

Defining good key results is hard. This is why people often don’t do it, especially when doing it for the first time. If not corrected immediately, poorly defined Key Results become a norm and fixing it becomes exponentially more expensive. 

Poorly defined Key Results lead to misunderstanding and inability to assess achievements objectively. The leader ends up with the following options:

a) Accept the self-assessment without scrutiny, therefore saying “I don’t really care about your results”.

b) Attempt to define the expected Key Results post-factum with a high chance of a hard-to-resolve conflict emerging.

There is no win-win situation if you did not define the Key Results well upfront.

If you don’t know how good Key Results look or how to teach people to set them, check out whatmatters.com - a training website from one of the creators of OKR.

Recipe 5: Kill Strategy by Not Checking In

OKR are a focusing tool. After you set them, you should review them at least weekly to decide what to focus on. 

Lack of this habit leads to people forgetting the OKR and working on tactical things, only to remember “oh, yeah, we did have some goals” when the next OKR cycle comes.

This is ineffective and erodes trust in goal-setting. After a couple of quarters like this, focusing people on long-term goals will be 10x as hard.

Make sure you have recurring workflows or reminders to align your day-to-day work with the OKR.

Nothing is Better than Something

If you can’t do OKR right - better not do them at all. Poorly implemented OKR have net negative effect on the company performance and people morale. 

You can get pretty far with ad-hoc goal setting, self-direction, and managing with 1:1s. If you are not ready to put a lot of effort into rolling out a formal framework - don’t try.

---

Want a post like this in your inbox once per week?