Writing OKRs in 1h or less
Objectives and Key Results. Sounds simple, no?
I mean, we're all familiar with the concept of objectives. It's part of our life. Every year on the 1st of January we usually set our objectives for the year: "This year I'm gonna finish university". Or "this year I'm saving money to buy the car I've always wanted".
How about key results? What are those?
Well, let's go back to the dream car. Let's say your balance in your bank account right now is zero. Your net income is, say, 2000 (your favourite currency here), you spend 1000 on bills and the car costs 3600. Of course you don't want to just spend money on bills and on the car, you want to have a life. Let's say you can afford to save 200 every month. That means that, if every month you look at your bank account at the end of the month and there's 200 more left than the previous month, in 18 months you'll be able to buy your dream car.
That's your key result: to look at your bank account every last day of the month and see you've got 200 more there.
Going back to the question in the beginning, yes, it is simple. And that's why it's so effective. You don't need fancy tools to keep track of goals and results. You don't need complicated words. Objective: buy my dream car. Key Result: every last day of the month my balance has grown by 200.
There are lots of amazing authors in the interwebs that can teach you all about OKRs in a much better way than me. If you're new to it, I'd recommend reading Felipe Castro's Beginner's Guide to OKRs or Perdoo's guide to writing great OKRs.
What I really want to talk about is how to structure a session for a team to uncover their OKRs. My team did an experiment with this at ThoughtWorks and I thought you could find this useful.
Important things first
Before talking about objectives for the team, it's important to know what's important to your team. What's their mission? What do they value? How can they bring value to the organization? Going back to the dream car, setting it as an objective wouldn't make much sense if you're not into cars, right?
So our first step was to discuss what matters to us, as a team. This is an example of some of the things we've pointed out.
Objectives: Where do we see ourselves in X months?
OKRs are all about cadence. They're constantly reset and re-evaluated every X months, to keep up with constant changes in the business and to ensure the team focuses on what really matters at that moment. Many authors recommend defining OKRs for cycles of 1–3 months.
There are many ways to write OKRs. You can phrase them as inspirational goals ("to buy my dream car"), as a ground-breaking-disruptive-letter-of-intentions ("to provide the most efficient battery in the history of mankind") or can even represent the visualization of a future state ("I'm standing at the top of the Everest").
We wanted to discuss objectives in a engaging and productive way. We didn't want to just write post-it's with long corporate-looking phrases, we wanted to have laughs while going through them. So we decided to talk about objectives in terms of instagram posts over a timeline.
(The initial idea was to put pictures and all, but we realized that would take too long, so we went or a hybrid approach: instagrams frames + post-it's! 😅)
It's important to note that time is a limited resource. No team will ever be able to tackle all of the goals they set for a quarter. OKRs is all about focus. About picking the things that matter the most and work had to deliver them.
You'll find lots of references in the literature suggesting keeping the count of objectives in between 3 to 5. So before going into Key Results what we did was prioritizing what mattered the most. We did a quick voting session and decided to focus on 3 goals, because time is limited, we're a small team and we're doing this for the first time, so most likely we won't get it right, not just yet.
Key Results: how do we know we're on the right track?
Ok, we know where we want to be and by when we want to be there. We've got 2 dimensions of OKR's, that's great.
But how do we know we're on the right track? How do we know the path we're following is the one that makes us get there on time, and not the one that makes us get there 24 hours late?
That's when Key Results get into the picture.
Personally I like to compare Key Results to Google Maps. Let's say I want to get to Sagrada Familia church. If 52m after my departure point I turn right onto Carrer de Joaquim Ruyra, it's a good start. I should be there on time. 55m later, I turn left on Carrer del Pare Lainez. Good, we're getting there. If after walking another 350m I can't find Carrer de Sardenya, I know I'm in trouble and maybe I won't be at Sagrada Familia in 15 minutes. Maybe it's time to change plans and take a cab to be there on time for my visit at Gaudi's master piece. And that's why KRs should be constantly reviewed and revisited: to decide as soon as possible when it's time to get a cab and avoid being late.
The important thing about KRs is for them to be measurable. If you can't put your KRs in terms of numbers, you're likely to have trouble understanding if you're in a good direction or not.
Here you can find an example of how we structured the Key Results for one of our goals:
And yet again, as much as we'd like having time to tackle every possible KR, it's not possible. So, once again, we apply prioritization: we quickly voted on the ones that we find more important and picked no more than 3 KRs per objective.
This is two-fold: it's about embracing things that we think can actually be delivered, but also about it keeping it small, since we're doing it for the first time and we want to see how well we go about it.
And all it took was 55 minutes…
And that was pretty much it. In less than 1 hour we managed to:
- review our goals as a team
- put together our objectives as projections in the future
- come up with metrics that, if measured them every two weeks, give us confidence that they tell us whether we're on a good path
Is that all?
Absolutely not :)
That was the easy part. Uncovering where we want to get and what numbers . "What's next?", you might ask.
Well, the next step would be discussing how to operationalize it. How to extract those numbers from the day to day work? Do you already have support from your tools? Do you need to customize something on your tooling? Is there manual work to be done? How often does it need to be done?
But that's a topic for another post!
