Key Results specify an Objective and make its success measurable. Defining them can, however, be really challenging. That’s why we talked to Christina Wodtke, an authority on this subject. Christina is a professor, author and successful businesswoman. She shared some interesting insights, based on her experience as an executive at companies like Zynga and LinkedIn.
Perdoo: What is the purpose of Key Results?
Christina: A Key Result is a metric that quantifies your Objective.
The Objective is aspirational and uses inspiring language, which is often vague. The Key Result makes that inspiring language specific. For example, you might have “successful product launch” in your Objective. Everyone has their own idea of what success is, so you need to spell it out: Is it DAU? Is it NPS? Is it conversion? Is it all three?
I’ve had CEOs ask me, “Can’t I just have Key Results and no Objectives?” But your company is made up of all kinds of people. There are those who are motivated by numbers, but there are also those who need a more inspiring goal.
Perdoo: You often read that Key Results need to be ambitious. How do I find the sweet spot between ambitious and impossible Key Results?
Christina: If you haven’t used OKRs yet, you probably don’t know what you are capable of. Most management frameworks are set up to be gamed. You have goals and targets with bonuses attached to them. But that kind of tayloristic thinking has been thoroughly debunked (Read Daniel Pink’s “Drive” for instance).
So you need to start setting somewhat ridiculous goals that aren’t tied to compensation, and see what your team can do when they are freed up to try the impossible.
Perdoo: What are the characteristics of a good Key Result? Is there a simple rule of thumb I can use to create them?
Christina: A Key Result is a result. Not a task, not a project, nor an action or todo. If a result is a result, you can change your tactics until the numbers move. If you made a result a project, you would find yourself locked into achieving it, even if you find out it’s the wrong approach.
A Key Result
- is measurable objectively.
- can be monitored over the OKR time period.
- is not a task you do or don’t do.
- specifies the Objectives. If you have the Objective “Launch an awesome alpha”, the result will quantify “awesome.”
- is hard to hit, but not impossible.
- answers the question, “How would we know if we succeeded?”
Some companies start with the Objective, some start with a Key Result; all companies will refine them as each informs the other. Start with the question, what MUST improve this quarter, but might be overlooked in the craziness. Depending on who or where you are, you might answer “We’ve got to get customer-focused” or “We’ve got to increase conversions.” The second answer is a Key Result, it just doesn’t have a number yet. If it’s a result, I ask myself: Why is that important? Why now? If it’s an Objective, but a sloppy one, I leave it be while we work on the KRs—it can be cleaned up once it’s better understood. Key Results almost always define an Objective, thus they should be created in tandem. Draft the Objective, come up with the KRs, refine the Objective, tweak the KRs.
As a rule of thumb, I recommend having a usage metric, a revenue metric and a satisfaction metric; but obviously that won’t always be the right choice for your Objective. The goal is to find different ways to measure success, in order to have sustained success across quarters. For example, having two revenue metrics means you might have an unbalanced approach to success. Focusing only on revenue can lead to employees gaming the system and developing short term approaches that can damage retention.
In the case of quarterly OKRs, you might theme each quarter, making Q1 all about retention, then Q2 all about conversion, and Q3 all about acquisition. We don’t pour water into a leaky funnel; we fix the leaks first.
Perdoo: Do I always need a Key Result for an Objective? What if I can’t come up with a Key Result?
Christina: Just like a Key Result needs an Objective to inspire, an Objective needs a Key Result to quantify and define. Just because the CEO knows what awesome is doesn’t mean anyone else does. That said, I’ve seen companies succeed with one Objective and one Key Result, or one Objective and five Key Results (more than that and your team can’t remember them all).
This is the magic sentence for creating Key Results: “How would we know if we achieved our Objective?”, i.e., “What would happen out in the world that we could observe?” Would organic visits go up? Would checkout conversion improve? Would NPS change? Then you can ask yourself “How much would things change?” For many companies, this means measuring things for the first time. But using OKRs means being a data-informed company. You can’t do that without data. I’ll be honest, the first time you set OKRs, you’re probably guessing. But that’s required if you want to know what kind of company you are: over-promisers or sandbaggers. It takes practice to learn what you can accomplish.
My background is in Design Thinking, so I take that approach to creating KRs. I have the executive team write down as many possible metrics for the quarter as possible. I give them five minutes, which is actually a long time when you are trying to think up metrics. And I give them permission to write down dumb ones—quantity leads to quality. Then the group sorts them, stack ranks them and picks the ones they feel are best to define the success of the Objective. It often leads to new ideas on how to measure the “unmeasurables”, like customer satisfaction or brand love.
Perdoo: Having three Key Results for each Objective is a widespread practice. Where is that number coming from?
Christina: It’s a way to avoid focussing on just one aspect of the Objective. But three isn’t a magic number, no matter what Schoolhouse Rock says. You can have fewer or more. Just take the time to think through what success really means. Don’t do fewer or more because you’re lazy or you find defining good Key Results too difficult. Be thoughtful and pick the right metrics to pay attention to.
Perdoo: How do results work together with actions?
Christina: Make sure you never mix up results and actions. Key Results are the metrics that have to move in order to achieve a goal. OKRs define what you want to do.
Actions, tasks, projects, etc. are necessary to get something done but they don’t tell you if you are doing the right thing. They are your hypothesis on how to make it occur.
Please note the word hypothesis. Even if you are right, the world keeps changing. You have to constantly try new tactics and change them if results don’t move. Lean has it right: do the smallest possible action first, measure the effect, then move to bigger efforts.
Christina Wodtke trains companies to move from insight to execution as principal of her firm, Wodtke Consulting, and teaches the next generation of entrepreneurs at California College of the Arts and Stanford Continuing Education.
Christina has led redesigns and initial product offerings for such companies as LinkedIn, Myspace, Zynga, Yahoo!, Hot Studio, and eGreetings. She has founded two consulting startups, a product startup, and Boxes and Arrows, an online magazine of design; and she co-founded the Information Architecture Institute.
She speaks everywhere from conferences to universities to boardrooms, and opines across the internet, but most often on eleganthack.com.