The 4 Most Frequent Struggles
Across all 4 companies I've worked at, some version of the OKR framework has been implemented. At zero of them, it worked successfully.
Based on my experience and talks with co-workers about their experience, here's what I think causes the framework to not achieve its full potential:
No buy-in - the executives allowed OKRs but never truly believed it was useful and thus, didn't invest the necessary time to craft great key results
No strategy - the company didn't have a proper longer-term strategy so all OKRs were blindly set without any commitment to the investments needed to achieve their outcomes
No data - key results were defined but without specific metrics rooted in proper data analysis because we didn't have the data available
No bandwidth - everyone was caught up with day-to-day work which meant very little time was left to even think about improvements and OKRs
What To Do Instead
When you have no leadership buy-in
If there's no buy-in from leadership to run the OKR program, and you still believe it's THE way to work, then I suggest the following:
Ask great questions to your manager about how your performance, or the success of the project you're working on, is measured by leadership
If it's binary and based on whether or not the "input" to the project is completed on time, then I suggest that you come up with your key results yourself
A different approach is to ask leadership "Why are we working on this?" enough times until you hear a problem, which's existence can be measured as you progress
When you don't know the strategy of the company
Related to buy-in, if there's no strategy in place, it's difficult to know exactly what's allowed and what's not, without always asking leadership for permission.
A strategy can take many forms, but one of the key things that I've always wanted from one was a definition of the playing field:
Who are we battling against?
What product assortment do we aim for?
What are 1 or 2 key metrics that frequently tell us if we're moving in the right direction?
In cases with no strategy, I suggest you write your own. Including this in any presentation will clearly state the assumptions you made.
The worst thing that could happen is that someone would disagree with your assumptions, allowing you to ask that person to explain their belief about the strategy, and voila! maybe there is a strategy anyway.
When you don't have the data
If you lack the baseline data to craft useful and measurable OKRs, then I suggest you use proxy metrics.
Proxy metrics are essentially metrics you believe correlate with the actual metric you wish you had. This is a great start and will allow you to move fast and learn while waiting for the correct data.
The alternative solution, which I don't hope you use, is that you default to binary output-based key results like "Finish task x". While it's a good idea to track progress, achieving key results like this has 0 guarantee it will lead to your desired outcome.
When you don't have the bandwidth
I've worked at multiple companies where a large percentage of staff was very focused on the operational side of things and had close to 0 extra bandwidth for "OKR projects". For cases like these, I suggest one of two ways to go:
Use OKRs to celebrate operational metric performance
Leave the team alone
Customer service teams are a great example of this. This team is usually grown as the need for capacity grows, meaning that no agents sit on their hands all day.
Therefore, teams like this should either have key results like "All customers are responded to within 48 hours" or be entirely left out of the OKR framework and use standard KPIs or SLA metrics.
A Better Alternative To OKRs?
Honestly, I haven't found one. What IS great about OKRs is that it forces you to say what you'll focus on. This is a great point of reference whenever someone suggests new work, and you can quickly pull up the OKRs and ask if it's more important than the current work.
You could also ask; how can we use OKRs better?
One of the things I've thought about is this:
You don't get to get an OKR
If your job is simple. If you're being given tasks by your boss. If you're sure what do to. Then don't use OKR. Save yourself the headache and use your brain and time to do your job instead.
But if you've been given a goal without a clear path to achieving it, and you have your manager's trust to try (and likely fail), OKRs is the perfect tool.
Have You Seen It Work Incredibly Well?
If you've seen OKRs work incredibly well at a company you've worked for, I'd love to know!
While I respect the focus OKRs provide, I'm still not convinced it's worth the time.
But then again, I also haven't found a better approach.
Cheers,
/Lasse