OKRs, Alignment and Autonomy
I can only remember one occasion where a conference speaker got standing ovations after their talk. That one time was Stephen Bungay’s keynote “The Art of Action” at the LKCE conference back in 2011. The talk was entertaining, flawlessly delivered, and, most importantly, revealed so many interesting insights that I won’t even try to list all of them here. One of the many things that stood out was this diagram that Stephen showed:
Many people assume that alignment and autonomy are opposing poles, meaning that the more autonomy people have, the less alignment you will get and vice versa. But what this chart suggests¹ is that we not only can have both, but we need to have both, in order to be successful when operating in dynamic environments like business. We want high alignment on our goals (“what do we want to achieve, and why does it matter?”), while at the same time allowing for a great deal of autonomy in terms of how we will achieve these goals. In other words: We want to be stubborn on our goals and flexible on the means. In many environments, we can’t know in advance what’s the best way to achieve a goal. Instead, we need to try different things and iterate our way to success. This model also allows for an elegant combination of top-town and bottoms-up approaches: Strategic goals can be decided top-down (after gathering input from different people with different perspectives), while decisions about how to achieve these goals can be delegated to the people who are actually doing this work (within certain constraints of course).
Objectives and Key Results
Objectives and Key Results (OKRs) have become increasingly popular in the tech industry over the last decade. So popular actually, that there are now a gazillion different interpretations of what constitutes good OKRs and how companies should use them. I will not even try to summarize the different approaches here. Instead I just want to highlight one aspect that I personally have found extremely valuable when working with OKRs: Separating objectives from actions. And when we do this right, it fits perfectly into Stephen’s diagram from above.
Blended OKRs
Let’s first look at an example where we have not separated objectives from actions and some of the problems that this approach might cause.
Objective: Create a world class checkout experience
KR1: Checkout conversion rate > 65%
KR2: Avg. Customer Satisfaction Score > 8
KR3: Build a tool to remind people of their shopping cart (so customers don’t abandon their shopping cart without knowing)
KR4: Integrate sign-in via Amazon (so customers don’t have to create a new account)
While all of the things outlined in these OKRs might make sense, KR1 and KR2 are conceptually different from KR3 and KR4. The first two describe how we measure success of the objective, while the latter two outline what we need to build in order to achieve this success.
The first issue with this blending is tracking progress. Let’s assume we have a group of people who meet on a regular basis to monitor progress and discuss potential action steps. After a couple of weeks we might see that we are almost done with KR3 and KR4 (both things are close to being shipped), while KR1 and KR2 are still at 0% progress, because neither of the metrics have changed (how could they, if we haven’t shipped anything, yet?) So what’s the overall progress of the Objective? We can’t really tell. Or worse, we rate the overall progress at about 45%, which is at best misleading.
The second big issue with this approach is that it might be limiting and frustrating for teams, if they get handed such a list, which tells them what to build (KR3 and KR4). In my experience, development teams are motivated by solving problems (“how might we increase the conversion rate by x?”) rather than implementing solutions.
A third problem with this list of OKRs is that it constrains flexibility of the development teams more than necessary. Once we have committed to this list of OKRs for a certain amount of time (maybe a quarter), we will work hard on implementing them. But what if we find out that there’s actually a better way to increase conversion rate than building a reminder tool?
Separate Initiatives
Although there’s not one right way of writing OKRs, I would prefer limiting the KRs to measurable outcomes and keeping the actions (we can call them initiatives) separate. Then out list might look more like this:
Objective: Create a world class checkout experience
KR1: Checkout conversion rate > 65%
KR2: Avg. Customer Satisfaction Score > 8
Initiative 1: Build a tool to remind people of their shopping cart (so customers don’t abandon their shopping cart without knowing)
Initiative 2: Integrate sign-in via Amazon (so customers don’t have to create a new account)
The initiatives are owned (and hopefully also (co-)created) by the teams that are actually implementing them. That allows us to create high autonomy for the teams while also having high organizational alignment on the OKRs.
As an organization, we commit to the OKRs, but not necessarily to the initiatives, because now it’s easier to see that they are just a means to an end. If we learn something new, we should be prepared to change, swap or kill initiatives, while still trying to achieve the OKRs.
This separation also allows for a way of tracking progress on our OKRs that makes more sense to me. But that’s a topic for a future blog post.
Takeaways
Autonomy and alignment need not be at odds with each other. We can create high alignment on the goals while having high autonomy on how to achieve these goals. OKRs can be used in exactly this way, which can make them a highly valuable tool. In order to do so, we should separate initiatives from OKRs.
Recommended Reading
Stephen’s book The Art of Action is a masterpiece, which I cannot recommend highly enough (Amazon link).
I have interviewed Stephen on questions of alignment, autonomy and leadership. You can read the interview here.
In this video you can watch a short conversation between Mattias Skarin and Stephen Bungay on The Art of Action and how they relate to OKRs.
P.S. In December my brother Stefan and I will host a training class in Hamburg, in which we will also talk about OKRs as a means to increase organizational effectiveness. Read more here (German)
__________
¹ The chart is actually just a representation of one core element of “Auftragstaktik”, a method that made the Prussian army extremely successful in the 19th century. Stephen has studied this approach carefully and transferred core ideas into the business world, where he calls it The Art of Action.
Möchten Sie mehr über OKRs lernen?
Zusammen mit meinem Bruder Stefan (ja, das sind wir auf dem Bild) gebe ich am 18./19.12.2024 eine Schulung in Hamburg. Dort werden wir mit vielen Praxisbeispielen erörtern, wie man Unternehmen auf Effektivität, Flow und Innovation ausrichten kann.