Good vs Bad OKRs
A beginner’s guide
last night, I attended a product anonymous meetup. It was the first time I’ve been to something like this in a very long time. My typical feeling towards meetups has been that my time was always better spent either working on our product at OHNO, or spending time with my family. But it was great, and I’m glad I went along.
The talks centred around the topic “F*#k Roadmaps”, a topic I’ve written about ad nauseam on Medium before.
But one topic that appeared both on stage and later, was about OKRs. Given OHNO is a tool that helps teams manage objectives like OKR’s, I thought I’d provide some observations about teams with good and bad OKRs that we see with our users. We can actually tell which teams in which industries are reaching their goals more than others, so we do have real data on this stuff.
But first, what is an OKR?
OKR’s stand for Objectives, and key results and were invented by Andy Grove at Intel. They found a champion in John Doer, who joined Intel in 73, and then as an investor encouraged two founders of a small startup in Silicon Valley to take them on board as a way to manage goals, Sergey Brin and Larry Page. To this day, Google’s proliferation of OKRs is in part responsible for their popularity but their adoption has gone well beyond the tech sector.
The process of setting OKR’s usually goes something like this. To set the objective, ask your self, in 3 months time, where do you want to be? To use the analogy from Googles online training program about OKR’s, let’s say our objective is,
Objective #1: Win the Superbowl
(forgive the details, I do not understand the rules of football very well)
The next part of the process is coming up with some key results. Think of key results as something that should have a number in it. You should be able to, in 3 months time, know if you achieved it with very little subjectivity. For instance, you either made a million dollars, or you didn’t. You score OKRs from 0–1. So if you only made $300,000 you might score yourself .3. Or you might just say you missed it and score it 0. The Key result and the context will help you decide which way to go in scoring. But make this clear when you set them. Teams will bring this up mid-way as they start to see where the numbers will land.
To work out what key results to set, you can ask yourself 2 questions.
Question 1.
If we had won the Superbowl, and we were to look around on that day at our team mates, the stadium and everything else, what would be true?
In order to win the Superbowl, the following things might be true.
- All players can run a 7 minute mile.
- We run an average of 10 yards per play.
- We have a field goal kicking success ratio of 85% or above.
Great, let’s put a pin in those for now. If we had completed 2 out of 3 of those key results, the over all score for that objective would likely be .66. We got 66% of them. Not a bad start.
Google is noted as saying their ideal key result score should be between 6 and 7. Anything above that and the objective is probably too easy to hit, anything below that and you might have some kind of problem. This perspective is very in line with Google’s stretch goals philosophy. But it might not be for everyone.
Personally, I don’t like this approach these days as much. I think perhaps it’s best to spell out what you want to hit and remove the bulls**t of stretch goals.
I think intentionally setting objectives people are unlikely to meet is a little misleading, and it encourages a kind of constant mental calculation. I’m not aiming for 100%, i’m aiming for 60%. So if I hit 60%, that’s really like hitting 100%. So 30% is really 50% etc.
My enthusiasm for the 6–7 ideal goal is fading, but like I said in the tips section, there are no hard and fast rules here.
Question 2.
The second question to ask is, what could we do today that gets us closer to winning the Superbowl?
Answers might include
- Start a fitness training program for our quarterbacks.
- Implement a new video recording software application that observers our players.
From talking to more and more people about OKR’s and watching how objectives are set with our customers, I think this is where the confusion comes into it and where I have seen many leaders argue about OKR’s.
Question 2 is where we start getting prescriptive. I think this might be the wrong way to go.
I noted even Marty Cagan while reading John Doers book, Measure What Matters about OKRs, noted that some of the key results in that book aren’t in fact key results, and are more tasks to be completed. So let me offer some council here.
If you use Question 1 only, you are encouraging more autonomy in the team. Kudos for you. This is the best situation. These key results are outcomes. For instance, being able to run a 7 minute mile is very different from saying, Get all players to practice running on Wednesday and Friday for 2 hours. If you phrase the key result as a solution, you are first, assuming that solution is the best way to get you what you want, and secondly, you might not be open to other solutions that get you to a 7 minute mile. This is where I made my first few mistakes when setting OKRs.
To articulate the point, here are a bunch of different ways of getting to a 7 minute mile.
- Run every single day
- Get a running coach who can teach you how it’s done.
- Perform general cardiovascular activities twice a day, and do one running session on Fridays.
- Genetically modify your body.
- Fire all players who can’t run a 7 minute mile and hire ones who can.
All those key results are different approaches. Some might have a really great time to value. Some might clash with your principals. Some might be harder for your team than others. The point is, any of those might work. Who knows! It’s best to let those closer to the action work it out.
By only using question 2 to get your key results, you might close the team off to adjusting things during the reporting period that get you to the outcome you want.
Imagine if your objective is to cook a steak for dinner. And your key result is buy a steak from Tony’s Butcher in the city. If you arrive at the butcher and it’s closed, you can’t really blame the failure. But if the key result is ‘Cook an amazing steak to medium temperature’ when they notice the butcher is closed, they are left to their own creative discretion about what to do next. Most likely, you’ll go to the supermarket instead. They’ll buy the steak, and you’ll get a great meal.
The best situation is to define some objective, win the Superbowl, and then say to the team, you work out how to get there.
The problem is, as my friend David Bignell pointed out during his talk last night, some people need plans.
Sadly, some people feel they need things spelled out for them because they’ve spent so long in rigid management structures, they’ve never had to exercise true creativity to achieve an objective in an autonomous way. So when setting OKRs, if you’re not sure which question to ask yourself, try this.
- If I give my team autonomy and they thrive in that, question 1 is best.
- If my team need things spelled out for them, include some question 2 key results. But try to convince them they don’t need this.
From all my conversations with OHNO customers, here’s some quick observations about those who’ve implemented OKRs well. I think these tips might help some people get the most out of the process.
Simple tips
- Try have 4–5 key results per objective. 3 is too little to make the scores meaningful.
- If you’ve never done it before, just start with 1–2 objectives. Keep it simple at first until you get the hang of it.
- What’s important is that the setting process doesn’t take too long. Keep it less than a week. Ideally, 1 day. Use it to get the team together and really flesh it out.
- Do at least 3 cycles to get the hang of it, or before you abandon the process.
- OKRs are typically not related to performance and compensation. This is the most asked question as traditional KPI’s often do relate to things like bonuses.
- The OKR’s and their scores should be transparent to all in the company. It’s important to be able to see what other teams priorities are.
- You usually have 3 month and 12 month OKRs, but the time periods can be whatever you think feels natural. Fullstory, an incredible user analytics tool sets brand new OKRs every 6 weeks.
- OKRs are best set at the team level, not the individual level.
- Key results are best set in a collaborative manner. I have found personally the best way (if you have a traditional hierarchy) is to have the leadership teams set the objectives, and let the teams come up with the time frames and the key results with some guidance.
- OKRs help communicate to other teams and yourself what not to work on. This last point is very important. OKRs help you say no.
- If you work in startup-land, I really think you should have at least one key result for every one of the 5 Startup Pirate Metrics. Acquisition, Activation, Retention, Revenue and Referral.
In the end, remember OKRs are nothing particularly special. It’s just setting an objective really. Any kind of goal setting is good. If you’re moving from no goal setting to any kind of goal setting, you’re in a much better place than you used to be.
About OHNO.ai
Our mission at OHNO is to reboot the workplace and liberate teams from mediocrity. We think there are too many ceremonies and practices we all know in our hearts to be useless & wasteful, and lots of them don’t serve our customers or help us enjoy what we do. At OHNO we value outcomes over traditions.
We want to liberate teams from those burdens by helping them get to the heart of what’s slowing them down using a really clever product we built called OHNO. If you liked this article or care about our mission, I’d love to hear from you. You can reach me on Twitter here.
And you can always try OHNO for free and discover what’s holding you back from your goals at http://ohno.ai
