The use case of weighted key results is that some key results are more important than others – hence progress on the important key results should count for more than progress on the other key results. Here is a quick example that we hear a lot:
- Objective: Accelerate growth
- KR 1: Hit $40m in revenue
- KR 2: Close 50 new customers
- KR 3: Discounts less than 10%
- KR 4: Onboard 5 new account executives
The logic goes, that KR 1 – “Hit $40m in revenue” – is most important – and hence, the progress on this particular key result should contribute much more to the progress of objective than the other key results.
The problem with weights
Adding weights to key results causes unnecessary complexity . Not only we have to define objective and key results, but we also need to define the relative importance of the key results.
Secondly, weights introduce a level of subjectivity . It’s relatively easy to say that something is more important than something else, but by how much? By 20%? By 47%?
Thirdly, weights hinder transparency – as we cannot glimpse quickly over the attainments, but we also have to decode an arbitrary formula behind it.
Finally, if we think of key results as a definition of success – then weights are non-sensical. If our objective is to build a car and drive from San Francisco to Cupertino – we could argue that building an engine is more important than putting on tires, but we are not going anywhere without tires.
We can agree that there are many issues with weighted key results, but the question is what should we do in situations where there is obvious disbalance between key results.
There are three solutions to this problem.
Remove key result
First and most obvious solution is to ask yourself, if one of the key results is not that important – why have it in the first place? To go to our example, does our key result to “Onboard 5 new account executives” really defines the success of accelerating growth? Probably not.
Just as good writing is simple and avoids anything that is not essential, so are good OKRs. Keep it simple and to the point.
Break up OKRs
Sometimes, you really do need to keep a key result but, perhaps, it should be under a different objective with other peer key results that are of the same importance. Going back to our example, we could create a new objective which would be called “Build a sales machine” and move the “Onboard 5 new account executives” under that objective. That new objective would then support our original “Accelerate growth” OKR.
The last, sort of a compromise solution, is to adjust targets – or implicitly weight chances of succeeding. For example, if discounting less than 10% on average is not that important to us, then we could say that we aim to discount less than 20%. While this is a possible solution, we do not recommend it – as it starts smelling of system gaming, and OKRs are not about getting good grades, but achieving results.
At Gtmhub we have lost deals because we refuse to provide weighted key results and we will surely lose some more – and that is fine. Our mission is to make organizations successful by making OKRs easy, and revenue is a function of that.
Our interests are perfectly aligned with the interests of our customers, as our relationship lasts for years and we are clinical when it comes to data on what makes one organization successful with OKRs.
So, if it wasn’t clear – no weighted key results. 🏋️♀️