fbpx

OKRs: let’s ditch those terrible metaphors

Francisco Homem de Mello

teaching

After years of studying OKRs, I can say that 99% of the content available out there does more harm than good for those who want to learn how to use OKRs to drive organizational performance.

These people are clearly well-intentioned and did a great job popularizing Objectives and Key-Results. But the examples given leave more questions than answers, and at times actually, contribute to people’s improper implementation of OKRs.

In this post, we’ll look at 3 popular OKR-learning resources and go over what’s wrong with them.

Rick Klau and the Google Ventures video

It all “started” with Rick Klau’s Google Ventures [1] video, entitled Startup Lab Workshop | How Google sets goals: OKRs, where he introduces the methodology to the GV community. The presentation (which has been watched more than 600k times on Youtube) is based on the original deck presented by John Doerr at a 1999 Google board meeting. In it, Doerr uses the metaphor of a football team to illustrate how OKRs permeate an organization.

The first example, drawn directly from Doerr’s deck, is the following:

A general manager for the team has the following OKR:

  • Objective: Make $$ for owners:
  • Key-Results:
    • Win Super Bowl
    • Fill stands to 88%

The objective is too broad and not inspirational at all. Nobody would want to work for a company that aims solely at making money for shareholders. Filling stands is also not a way to measure if money was made for shareholders. Stands could be filled by giving away tickets, or selling them at a deep discount. The same could be argued about winning the Super Bowl. A rewrite would be:

  • Objective: Grow revenues with profitability
  • Key-Results:
    • Increase ticket sales to U$ 150 million
    • Achieve a net profit margin of 15%
    • End year with U$ 50 million in free cash generated”

Doerr than goes on to unfold those weird general manager OKRs to the head coach and the public relations VP:

  • “Head Coach Objective: Win Super Bowl
  • Key-Results:
    • 200 yd passing attack
    • No. 3 in Defensive Stats
    • 25 yd punt return avg.”

Jeez, that’s awful. First of all, these Key-Results don’t help measure if the Super Bowl was won. You can hit all these stats and still fail miserably at winning the tournament. You could also argue Winning the Super Bowl is not an Objective, but a Key-Result, that measures “excellence” or something. That’s why I hate sports metaphors. Learning to “implement” OKRs at a football team hardly helps someone implement them at a company.

The same goes for the Objective unfolded to the Public Relations VP: “Fill stands to 88%.” That’s not an Objective, which should be qualitative, but a Key-Result, which looks like a smart goal. The Objective could be “increase the popularity of the team,” and that Objective could, in turn, be measured by the following Key-Results:

  • Fill stands at home stadium to 88% on average for the season
  • Increase the number of social mentions (Twitter, Facebook, and Instagram) by 25%
  • Show up 15 times in the major sports press

That would be a bit better. They could all measure the guy’s ability to increase the team’s popularity. And the OKR as a whole clearly contributes to “making $$ for the owner.”

The second questionable example used is the following:

  • “Objective: To develop a workable model for planning as measured by
  • Key-Results:
    • Finishing the presentation on time
    • Completing a sample set of 3 months OKRs
    • Having management agree to institute a trial system for a 3 month period”

We can all agree that that’s not an Objective, but a Project. The Project’s goal is to implement OKRs, and its milestones are three efforts, and not results. As we’ve already talked about, OKRs and Projects should coexist in the company’s management system, but they should be clearly distinguished. OKRs are all about business results. Projects are concerted efforts that could drive business results. They must be tracked and managed, but their completion shouldn’t be mistaken for a business result.

A good thing about the example, on the other hand, is that Doerr, and Klau, frame Key-Results as ways to measure if the Objective was completed. We can infer that by reading the “… as measured by…” final portion of the Objective. But “finishing the presentation on time” is hardly a business result.

In a later tweetstorm, Klau said the original video’s content was dated by many learnings he accrued since that day and pointed followers to Gooogle’s re:Work page on OKRs – which to me is even worse – as a source for updated content.

Google’s re:Work

On writing good objectives, Google’s HR blog suggests that you “… use expressions that convey endpoints and states, e.g., ‘climb the mountain,’ and ‘eat 5 pies.’” What a terrible choice of examples. They do more disservice than service to those who want to learn about OKRs.

First of all, if anything, these look more like Key-Results. First, they aren’t qualitative, and second, they couldn’t be coupled with Key-Results that’d measure if those objectives were attained. What could we use as key-results to “Eat 5 pies”? “Feel 5 pies reach my stomach”?

Second, they just don’t make sense. It’s much better to use actual company OKRs as examples, because that’s, after all, where the methodology can be applied to with best results for the effort.

OKRs and 50 Cent

We’ve already covered football OKRs and pie-eating OKRs. Finally, let’s talk about rap music OKRs.

In a recent article, Tren Griffin, an investor/author currently working at Microsoft, cites 50 Cent’s competition against Kanye West as an opportunity to teach, and learn, about OKRs.=

He uses the following OKR to illustrate 50 Cent’s ambitions prior to launching his famed album, Curtis:

  • Objective: “Independence – never work for another person again”
  • Key-Results:
    • Sell more albums than Kanye in a week
    • Get Rich or Die Tryin

Well, another confusing example. The Objective, even tough ok from the “qualitative” standpoint, looks more like a vision statement, very lofty and long-term. Objectives should be about improving something, and not some guideline of how the company (or in this case, rapper) should operate.

The Key-Results don’t help much either. “Sell more albums than Kanye in a week” is, on the surface, a reasonable Key-Result: It is quite measurable. What’s wrong with it is that it doesn’t help measure of the Objective was achieved. Fiddy could have sold 2 albums and Kanye 1. The Key-Result would be achieved, but would hardly mean financial independence. But “Get rich or die tryin” is a mess. Not measurable in any way. How much is “rich”? If it were measurable, it could even serve as a Key-Result to financial independence.

I hate OKR metaphors, but if I had to rewrite Griffin’s 50 Cent OKRs, they would look more like:

  • Objective 1: Increase financial independence
    • Key-Result:
      • Increase net worth by U$ Y million
      • “Reach monthly income of U$ Z million
  • Objective 2: Beat Kanye’s Graduation
    • Key-Result: Beat Graduation’s first week sales with the launch of Curtis

What we did is to set proper Key-Results (more like smart goals) and break them down in two, so that the KRs actually measure the achievement of the Os.

We’ve reached the end of our article. I sincerely hope you’re a bit less confused about OKRs than when you started it. If you have any comments, shoot me an email to kiko@qulturerocks.com. I love helping people implement OKRs, for real.

———
[1] Since then, Google Ventures’ name was changed to simply GV, after the company was spun out of Google and into Alphabet, Inc.