Developing OKRs you’ll actually use

I’ve talked previously about OKRs and how much i love to hate them. For me OKRs are more a philosophy and approach rather than a tool or framework to use, i find that if you try to implement OKRs ‘out of the box’ its a very frustrating experience, and can be counter productive since you end up investment massive amounts of time on developing and managing OKRs instead of getting things done.

The problem, however, is that even if you agree that OKRs are more of a mindset or philosophy, how do you then go from this conceptual model to something executable and able to deliver results quickly and effectively?

Here you see an example i was using to help one of the companies i’m working with (excuse the typos) you see it split into 3 sections the top section is the formal OKR structure, in the middle you have a variety of notes, and at the bottom we have some calculations.

as a TL;DR, basically whats going on is i’m filling out the formal structure at the top, by first taking notes and investigating aspects of the business model (in the middle list of notes), then using what i understand of the business model to fill out key metrics at the bottom, which are then re-worded for OKRs at the top.

Diving deeper – the first step of the OKR mindset is to define the key problems that the company faces and go into a deep discussion about the dependancies and factors that are affecting that problem. in this example the issue was the company need to reach a milestone in revenues to meet certain expectations, there was a clear causal-dependant relationship to Proof of Concepts and feature development – we actually spend a good hour exploring and trying to understand the relationship between the key parts of the machine, in this case, the value of the POC, the number of POCs needed, and how this would affect hiring and feature development.

Once we understood the relationships we could work back from the target number to understand the relational numbers, after putting that in, we further discussed if these numbers made sense, these became the outlines for the KRs

Here’s where we reach another key area in the OKR methodology worth spending time on. Very often the KRs will be written SMART, but they will also tend to fall into ‘lagging indicators’ – meaning the KR would be a number that shows a result of an effect, rather than measuring an effect. for example, you’d more often see a KR like “increase revenue by 20%” which is a measure of the result of an effort, its not a measure of the impact of the effort, so a more effective way to rework this “lagging KR’ is to reword it in a way thats more proactive for example “upsell X customers to 20% higher value packages” – The trick here is to really think about what is within your control to impact within the KR, we can’t necessarily get a customer to buy (so revenue from what they bought is a lagging indicator), but we can influence how often we engage with the customer, and what product we upsell to them (leading indications) which would have an impact on that revenue number.

So, once we’ve re-defined the KRs to be around leading indicators, we can then rewrite the OKR in the more succinct format and track progress to that OKR on a regular basis.

Leave a Reply

Your email address will not be published. Required fields are marked *