Top 3 Mistakes People Make When Implementing OKRs

Over the last few years, I’ve seen some recurring patterns crop up as I’ve helped people attempt to organize their goal setting around the powerful but deceptively simple approach of OKRs (Objectives and Key Results). To be clear, I’ve never encountered a group that didn’t struggle at least a little to implement an effective OKR system, and I love getting to help people do this better.

For anyone unfamiliar with OKRs, here is a very brief primer. The objectives in OKRs represent what it is that a person or group is ultimately trying to achieve. Great objectives should be crystal clear, action-oriented, and at least a little inspiring. Being inspiring and being crystal clear, by the way, are features of objectives that I find are often in tension with one another, insofar as people tend to get hazy and abstract when they are trying to express aspirational goals. Keeping each of these core features of great objectives balanced with the other is part of the art of writing effective objectives.

Key results represent the concrete actions or outcomes that will be necessary to achieve the objective they are tied to. Generally, well-crafted objectives should have no more than 3-5 associated key results. If you feel you need more than this, there is a good chance that your objective needs clarifying. Great key results should be specific, reliably evaluated, and timebound. Although it’s not a universal law, a good rule of thumb is that key results should typically have a number attached to them (e.g., 100 widgets will be rolled off the production line by the end of the week). Being reliably measured means that there is no vagueness in what it will mean to achieve this key result, so two independent people will consistently agree whether the key result has been achieved or not. In other words, there isn’t much subjective judgment involved with well written key results. This feature is where the ultimate accountability of the OKR system resides. Everyone involved will know (and agree) that a key result was or was not achieved when the time comes to evaluate progress.

There are more details and nuances that we could dive into regarding how to write good OKRs, but these are their most fundamental elements, in a nutshell. Here are three of the most common problems that I’ve seen when people try to implement OKRs.

Problem #1: Leaders being too heavy-handed in ‘cascading’ their organization’s OKRs

There are two extremes in creating goals for an organization. The first is the top-down approach, in which the top dog in the organization hands down goals like Moses coming down from the mountain with the Ten Commandments. The other is the bottom-up approach, in which the leader gives full rein to each person in the organization to come up with their own goals. Both extremes tend to be problematic.

Problem #1 comes from the first, top-down approach. To be clear, there are times when a leader needs to be fairly directive with a group’s objectives. Change management situations are often such times. Likewise, some employees need more directing than others do when it comes to effective goal setting. But when a leader routinely or completely hands down all of a team’s objectives and key results, it has the effect of reducing buy-in among team members and causing morale to implode. Even when a leader feels the need to be rather directive with respect to a team’s objectives, the team members ought to produce (or at least have input into) the key results that correspond to those objectives. If a leader does not feel confident in allowing her team to have a say in at least the key results, that’s a sign that the leader might not have the right people in the right roles with the knowledge and expertise required to do their jobs well—or that the leader needs to learn to delegate and trust the team (people who believe, deep down, that being a good leader means having all the answers tend to struggle with this).

Problem #2: Followers cramming everything they do into their OKRs

Problem #2 reflects the second extreme that I mentioned above, the bottom-up approach to implementing OKRs. I’ve seen leaders make this mistake when they are trying to enhance buy-in from followers, or when they are exhibiting a highly democratic style of leadership (and occasionally out of a somewhat lazy, hands-off approach to leading). This OKR implementation error has the tendency of causing followers to stuff everything they do in their jobs into their OKRs, which is not what OKRs are for. OKRs represent an organization’s (or person’s) priorities, not 100% of the tasks they do in a typical week. And yet this is what OKRs tend to look like when no input has been given from a leader. People attempt to justify their existence, showing how busy or important they are. “Look at all my many objectives and key results! Aren’t I valuable? What would you do without me?” they seem to say. 

OKRs are not a job description. They indicate what can wait, when push comes to shove, and what can’t. 

Problem #3: Objectives that aren’t particularly inspirational

The third common problem I’ve seen with people’s implementation of OKRs derives from other goal-related systems and a confusion between OKRs and KPIs (key performance indicators, which are usually tied to compensation). Consider, for instance, the well-known framework of SMART goals. SMART stands for specific, measurable, attainable, relevant (sometimes reframed as realistic), and time-bound. SMART goals have been touted for a long time as the best way to create goals, and there are certainly some praiseworthy aspects of this approach to goal setting. Effective goals should be specific, rather than vague, and they also ought to be measurable. OKRs include both of these qualities, as well as being time-bound. 

But the biggest distinction between SMART goals and OKRs is that where the former tend to be conservative (in order to remain attainable and/or realistic), the latter are inspirational. People know that they are supposed to attain all of their SMART goals. Not doing so is a failure. In contrast, the inspirational nature of OKRs means that they won’t always be achieved, at least not fully. Indeed, companies like Google, which has used OKRs since the late 1990s, set achievement benchmarks around 70% for employees. 70%! For a company like Google, if they hit much above 70%, they see that as a failure—specifically, a failure to be aspirational enough in their goal setting. With that sort of culture-shaping mentality, is it any wonder that Google has become one of the biggest, most successful tech companies in the world, a company known for its prediction-shattering innovations? Companies that insist on achieving 100% of their goals won’t tend to take the kinds of risks needed to innovate. Progress and growth in such company cultures is measured in inches, not miles.

The implications of this third problem with the implementation of OKRs can hardly be overstated. If all you want for yourself, your team, or your organization is to make small, incremental changes, growing in modest, predictable ways, then SMART goals are for you. If instead you want to make transformational changes and grow in profound ways, then OKRs represent a more effective system. OKRs are not KPIs, and they should not be tied to compensation (unless, perhaps, compensation is linked to whether objectives were appropriately aspirational, independent of whether those aspirations are achieved). The inspirational nature of strong OKRs provides the motivational impetus needed to achieve profound changes. That impetus can be neutralized completely by a focus on fully attainable, realistic goals that carry with them a risk of negative consequences if they aren’t achieved.