If you’re approaching the end of on OKR cycle, it may be time to refresh on how to grade them. Grading OKRs is an opportunity to reflect on accomplishments and what might be done differently the next time around. Low scores force a reassessment, while high scores provide empirical proof of delivery.

If you’re using OKR software like BetterWorks or Lattice, OKR scores will be system generated for you. If you’re not, you’ll have to do some math. But, no matter how you calculate your score, it’s good to understand the meaning behind the numbers and where they come from.

The Andy Grove method of grading OKRs is a simple “yes” or “no” approach. To see it in action, let’s use a soccer analogy. Imagine you’re a recruiter for a team. Your OKR might look something like this for the quarter:

O: Recruit three new players

KR1: Attend 25 games to scout out potential recruits

KR2: Approach 30 players throughout these games

KR2: Contact the agents of 10 potential recruits

Here’s how you would come up with your score:

• Attend 25 games to scout out potential recruits? No
• Approach 30 players throughout these games? Yes
• Contact the agents of 10 potential recruits? Yes

This is the basic “yes/no” of OKRs.

There is also a more advanced way to score each key result on a scale. “0” equates to failure and “1.0” means the objective was completely achieved. Within these metrics, each individual key result is graded and averaged to score the objective.

The scale goes like this:

• 0.7 to 1.0 = green. (We delivered.)
• 0.4 to 0.6 = yellow. (We made progress, but fell short of completion.)
• 0.0 to 0.3 = red. (We failed to make real progress.)

To visualize it, let’s use the same soccer recruiter OKR—but with a little bit more information:

• You could only get to 20 games so that’s 0.8, an admirable score.
• You approached 30 players, so that’s a perfect 1.0.
• You were only able to get ahold of 6 agents for 0.6, a borderline green.

Altogether, the average is 80%—or a raw score of 0.8, a passing grade.

However, OKRs go beyond this and require self-assessment, too.

In analyzing OKR performance, objective data should be looked at subjectively, also, because, for any goal, there may be extenuating circumstances.

Let’s say you are again a soccer recruiter but that you were only able to get ahold of one agent—but that agent represented the star center midfielder that everyone else was after. Your raw score for that KR would be 0.1 but you might give yourself a 0.9 because you were able to get her signed, too. As John Doerr writes in Measure What Matters, “The point of objectives and key results, after all, is to get everyone working on the right things.”

One final thing worth noting is that you want to see variation in your key results. As stated earlier, 70% is a good score. If everything is 100% or 30%, that kind of homogeneity is suspicious. It may mean that you need to set stretch goals or completely rethink your OKRs.

The answer is in the numbers.

Have you graded your OKRs recently? We’d love to hear what they were and how they went so please let us know. And, if you’re looking for more OKR information, you can find other FAQs, Resources, and Stories right here on WhatMatters.com.