Apples and Pears
Some background. KPI stands for Key Performance Indicator and I think they're the bees knees. OKR stands for Objectives and Key Results and from what I can gather in my various reading, are also the bees knees. How can both are true? Stick with me here.
We often tell a familiar story through a different lens (think of all the interpretations of Romeo and Juliet) and sometimes that new lens brings something more info focus. Six Sigma, for example, is a process improvement methodology famously championed by Jack Welch at General Electric in the 90s. Six Sigma has enjoyed consistent success and investment because it works, but it didn't start in the 90s. Japanese industrialists introduced quality management techniques (learned, at least in part, from American industrialists) in the 50s. The methodology in the US was called Quality Circles, then Total Quality Management, before the techniques became Six Sigma.
Objective and Key Results are originally attributed to Intel in the 80s and are further explored by John Doerr in his great book Measure What Matters. They continue to be popularized by Intel, Google, and other companies who are excited about them.
What are OKRs?
In short, OKRs are a stated objective (the O) and one or two key results (the KR) which is how the objective will be measured. The easy guide is "I will (objective) as measured by (key results)." Like "I will improve customer satisfaction on customer orders as measured by the post-order survey results." They are intended to be aggressive (but achievable) and relatively short duration (like quarterly). OKRs can be company-wide, department-specific, or individual which is super-handy. Some of what I read is that OKRs should be memorable and motivating, to inspire action. The thing I like about OKRs is the immediacy and the focus. Even for the sample one I wrote above, you're unlikely to need a multi-step project plan to make that happen, which is great. I have a pet peeve about project status updates that take time away from executing the actual project.
I think this is the point where I'm supposed to tell you that OKRs are better than KPIs, or why KPIs are better than OKRs. If I took a side, I get to contribute to the competition for organizational attention, and who doesn't love that? I mean, I DID just write a book about KPIs, so I have territory to defend here, right?
Nope. I'm going to instead tell you that they're the same. OKR folks for whom it's important to choose one over the other are citing qualitative (and unnecessary) differences like KPIs are stale and OKRs are fresh, or that KPIs are slow and clunky while OKRs are agile. There's a lot I'm learning from reading these blogs, and more that I'll include in my own thinking about how I think about measuring success. But one of the key points I took from Felipe Castro, who writes a lot about OKRs, is that it's bad when KPIs are constructed in a way that they don't drive improvement. Yes, I agree! I agree with all of this, but not in a one-or-the-other sort of way.
KPIs (or a measurement strategy by any acronym) needs to start at the top with a mission or a vision. An objective - as the motivational statement in OKR - should read like a subset of that mission. What makes it motivational is that the end should be in sight. With a mission toward "Best customer service", there needs to be shorter term objectives that a team can rally around. And at the end of a year or two, some of those short term objectives ideally have moved the needle on the overall mission (e.g. "best customer service").
"Key results" reads likes KPIs - at least my version of KPIs - with the emphasis on "key". At any given moment, particularly in the data-rich environment in which most organizations work, there are way more things that can be measured than should be measured. This is why KPIs or OKRs need to be clearly in service of some aspirational mission ("We will be...") so that the focus continues to be on action and improvement toward that mission. And also so that extraneous (albeit interesting) measures don't get the same air time as actual key indicators of performance.
In Measuring Success I only talk about KPIs, and I describe how to narrow a high level mission down to strategies, goals, and initiatives that can be measured with clear results. OKRs do a great job of focusing effort in the short term, but there isn't any need to reject one acronym for the other.