The Objective - Key Result (OKR) framework is a powerful tool for tying short-term actions to long-term vision. Especially for teams, where you have multiple people to align and coordinate. The key to the model is the two-part framework which provides traceability to long-term goals while specifying short-term action items in a clear and easy-to-manage process.
As a coach, I use quarterly OKRs to help teams decide on improvements they are prioritizing and implementing. My goal is to create a simple and effective method for focusing and aligning the team's efforts. Of course like any framework, while the ideas can be straight forward the implementation can be difficult. For some teams this process can be fairly easy. For others its a challenge.
Regardless of your situation or history, below are several suggestions which will help make the process more successful.
Start with a sober assessment of where you currently stand
You need a clear vision of both your destination and your starting point. If you set overly-ambitious objectives, you risk frustration and disappointment. Avoid failure by starting with a honest assessment of your current skills and capabilities. Then set reasonable objectives that give you good reasonable chance for success. Build on that success by setting progressively more challenging and ambitious goals. You wouldn't go out and run a marathon if you’ve never completed a 5K.
Set OKR’s on quarterly basis
One calendar quarter is the ‘goldilocks’ time period for OKRs. Monthly is just too short. Yearly is too long. Three months is just right. 90 days gives enough time to set meaningful goals while still creating urgency. I find that with yearly OKRs, people will to wait until the last quarter to really focus on action, effectively wasting nine months of time. When you’re working in monthly cycles, a spike in regular project demands can through an entire cycle out the window. With a three-month cycle, when a project crisis hits and you need to put your OKR’s aside for a few weeks, you still have time to pick them back up and make good progress.
In the beginning, it’s better to under commit
Go easy the first round. Being conservative allows you can focus on learning the process and building a solid foundation. Once you have a good quarter and have the taste for success, then get more aggressive. I’ve seen teams who commit to far too much, complete only a fraction of their KR’s, getting deflated, and then spend several months rebuilding their confidence. I’d rather see a team get a few things done well and build momentum.
Objectives show progress towards a vision
When I write objectives, the first question I ask is, “how does this objective help advance me towards my vision?” An objective that doesn’t tie to a vision is at best a waste of time. At worst, it actively interferes with organization success. Good objectives have a clarity of purpose. Sometimes this takes a few rewrites. Everyone should be satisfied that achieving an objective will clearly advance the team in the right direction.
Here are a few examples where rewriting objectives can improved clarity:
- Get more feedback became Get more customer feedback early in the process so we can prioritize our work better
- Improve retrospectives became Go deeper in retrospectives so we can find and address root causes
- Do one-on-ones became Send time each week with direct reports to develop closer personal relationships
- Waste less time in meetings became Improve meeting productivity by having a clear goal and meeting agenda
- Work in iterations became Improve delivery consistency by planning and delivering our work in consistent iterations
- Meet our deadlines became Improve our ability to estimate complexity and capacity so that the business can prioritize and plan confidently
Target 3-5 objectives, each with 3-5 key results
I suggest teams have a total of between 15-20 KRs for the quarter. The size and complexity will depend on the team composition and size, but I find those numbers to work well. Using the 70-80% goal, that means a team is targeting to complete roughly 10-15 KR’s. That's about one a week, which sets a reasonable pace. Generally I recommend you invest about 10-15% of your total capacity in OKR work.
Assuming a team has 3-5 KR’s per objective, then you should have between 3-5 objectives to get to your 15-20 total KR count. Once you have a well-written objective, find three to five key results that, when completed, will show meaningful progress. The KR’s don’t need to “complete” the objective, just show substantial progress. An objective can go on for several quarters with different KR’s under it. Start with three to five KR’s per objective. Less than three and you leave too little room for choice. More than five and you’ll create too much focus on one objective or too many KR’s.
Select objectives based on passion and impact
It's easy to come up with dozens of objectives. Select those you feel both passionate about and will make a meaningful impact. This will motivate you. If you have an objective that the team is passionate about but doesn't have a clear impact on strategy, keep working on it and find an angle. I find that it’s easier channel passion into impact than to try and rally passion for an objective that lacks interest.
Key results need to be specific, definitive, and independent
Key results need three things to be effective. First, they need to be specific. This means that everyone knows the scope and what tasks are involved, and what tasks are not. Second they need to be definitive. You know when it’s done. And I mean done, done. Avoid subjective measures and vague definitions. Lastly, they need to be independent, which means that the team has the power to complete each KR without dependence on outside parties or resources. This includes independence from other KR’s. Your KR’s should not end up in a big Gantt chart with tasks strung together in interlocking threads.
Develop OKR’s as a group
OKR’s should not be a collection of individual OKR’s. Everyone should have their own personal OKR’s, but keep them separate. Teams OKR’s need to be about what the ‘team’ needs and should be team efforts to complete. One person should not go off by themselves and complete all of the KRs.
Don’t over-complicate tracking and scoring
I’ve seen teams come up with elaborate point values and scoring formulas for each KR. Over-complicating them in the beginning typically turns into a problem and focuses the team on process rather than progress. Better to start simple with a system that everyone can see and quickly calculate progress. If KR’s are of significantly different sizes, I typically have the team adjust them--splitting them apart or combining them or re-writing them--to get them to be in rough balance. For tracking, I suggest a simple chart where each KR can be checked off when completed. Progress is simply calculated by completed KR’s over the total number of KR’s to get a percentage.
Plan and prioritize your OKR work along side your other project work. I have teams keep separate task backlogs, one for OKR’s and one for project work; but when we plan iterations, it's one absolutely prioritized list. For teams working in iterations or sprints, I have them allocate 5-10% of their capacity for OKR stories. For teams working with kanban or continuous pull systems, I suggest pulling an OKR task every 10th task.
Burn-down and/or burn-up charts can help visualize work completed, work remaining, and current progress rate. This allows everyone to see the progress, or the lack there of, and set expectations accordingly. For OKR’s, since the target is 70-80% completion, I have the teams mark those rate lines on the chart so they can clearly see the target zone.
Assign one owner for each key results
While the entire team needs to be involved in the completion of OKR’s, each KR should have just one name next to it. The owner is responsible for managing tasks, delegating work, and removing roadblocks. Do not expect the owner to complete the KR by themselves, but they will likely be heavily involved. KR ownership should be distributed as evenly across the group as possible. I also encourage people to take KRs that are slightly out of their comfort zone. Again, this should be a group effort.
Target 70-80% completion of your KR's
You should have some stretch to your KR’s. If you complete all of their KR’s in a quarter, you haven’t stretched enough. Teams should target completing 70-80% of their KR's in a quarter. This gives you room to do more than planned while ensuring that the you are not over-scheduling yourselves and wasting time and energy. If the you are getting less than 60% completed for two quarters in a row, then cut back. If you are finishing more than 90% for two quarters in a row, add more. Again, do not over-complicate the measurement and tracking.
Don’t get stuck on a KR, if it’s not working move on to something else
If you realize that the KR just won’t deliver the value you expected, don’t be afraid to scrap it altogether. This is another reason to plan on 70-80% completion. It leaves room to drop a few if they turn out to be duds. Be careful of getting into this habit, however, as it will lead to sloppy OKR definition at the beginning of the quarter. And only do this with KR’s, not with objectives. In all cases, make sure to address these in a retrospective.
Sometimes you end up with an KR that turns out to be more difficult than you estimated. When this happens don't be afraid to move on to other KRs and put this one aside. If you have time at the end of the quarter you can pick it back up.
Retrospect after each quarter and apply what you’ve learn
Lastly, be sure to retrospect at the end of the quarter. This is where teams can learn about what’s working, what’s not, and where they can improve. I suggest the following retrospective structure: 1) create awareness by collecting and present information and data, 2) develop insights by asking questions and challenging assumptions, 3) develop a multitude of possible action items, and 4) choose what to implement based on impact and feasibility.
Bruce Eckfeldt is an entrepreneur and former Inc 500 CEO. He provides executive and team coaching and management training to startups and high-growth companies. For more information on Bruce and how he can help you and your company, visit http://www.eckfeldt.com or contact him at firstname.lastname@example.org.