The implementation of OKRs can fail for several reasons. Here, we will detail the main ones so that you can get ahead.
Not doing it gradually
OKRs should be implemented gradually:
-
From fewer OKRs to more OKRs
-
From company goals to individual goals
-
From shorter cycles (and faster corrections) to quarters
If you try to do too many things at once, such as cascading five company OKRs down to the individual level in an annual cycle, your project will almost certainly fail. People won't know what's going on, they won't remember their OKRs, teams won't monitor them, etc., etc. It's very important that the company starts at the right goal-setting cadence, and this means warming up the engines gradually, following the right rituals, and creating healthy habits. In the end, tracking results, and subsequently setting OKRs, is a surprising example of a managerial habit, much like exercising regularly.
Baby steps: start with one company-wide goal and three or four key results in a one-month cycle. See how it goes, learn from the process, and repeat. After three months, loosely cascade OKRs to teams. Another three monthly cycles. And so on. If the company has incorporated the cadence—the habit—of tracking its key KPIs, you will have taken a huge qualitative leap in your management and the strategic alignment of your company.
Setting too many OKRs
One of the great advantages of OKRs, especially for companies that have no practice of strategic planning or goal management, is that they lead to a significant gain in focus. Since OKR cycles are shorter, you can afford to have a narrower focus per cycle, aligning the entire company around important initiatives.
The total number of Os + KRs for any single entity (be it the company, a team, or an individual) should not exceed the fingers on one hand. More than that, and they won't be remembered.
Let's look at a practical example of OKRs for a product manager for the 1st quarter of 2016:
Objective 1
-
Improve engagement through the recurring billing product
Key Results
-
Increase the number of unique visitors by 15%
-
Increase the average time spent on each page by 5%
Objective 2
-
Achieve a high level of customer satisfaction
Key Results
-
NPS > 8 with > 90% of the customer base responding
Objective 3
-
Implementation of new features
Key Results
-
Launch the credit card reconciliation feature with at least 4 customers per day
As you can see, Fred has three things to focus on this quarter:
-
Engagement
-
Customer satisfaction
-
New reconciliation feature
These priorities are supported by four key results that underpin them. This is highly manageable and very easy to remember. As Marcel Telles, former CEO of AmBev (now SAB Miller + Anheuser Busch InBev), says: "Goals must fit on the fingers of one hand." We've expanded the number of fingers just a bit to fit OKRs
Not tracking OKRs
Goals lose their purpose if they are not monitored. Accountability is essential: employees should own their OKRs, and goal achievement should be continuously monitored. There are several reasons for this:
-
Discussing the "whys" behind each goal helps the organization learn what works and what doesn't. What's the point of knowing that someone achieved their goals if they don't know which specific actions led to the results?
-
Ongoing monitoring shows that the organization cares about OKRs. Many companies spend time and effort implementing and discussing OKRs only to forget about them until the end of the cycle. If the company is not monitoring its goals, teams, leaders, and subordinates won't monitor theirs either. It's an example that needs to be set by the company, from top to bottom. The company's OKRs should be monitored "all-hands"; team OKRs should be monitored in team meetings. Never skip an OKR check-in meeting. Just as zero tolerance helped reduce violence in New York, it will work wonders for your high-performance culture.
-
OKRs that are too difficult (or too easy)
There is scientific evidence (Locke) that more challenging goals produce better results. That's why "stretch goals" are talked about so much. Carlos Britto, former CEO of AB InBev, says that the ideal goals should be those "you know 80% how to achieve. The other 20% will be learned along the way." But science also suggests that goals should be achievable, i.e., realistic: setting goals that are too difficult frustrates people (who must believe they can achieve them).
Locke and Latham say, once again:
"Nothing breeds success like success. On the other hand, nothing breeds feelings of hopelessness like constant failure. The primary purpose of setting goals is to raise an individual's level of motivation. But goals can have exactly the opposite effect if criteria are adopted that make the individual feel continually incompetent. Consequently, the supervisor should be on the lookout for unrealistic goals, and be prepared to change them when necessary."
Goals that are too aggressive
One of the most common mistakes companies make when adopting OKRs is setting too many stretch goals at the outset. Since many don't have a solid understanding of KPI levels due to the ever-changing nature of young companies, they tend to err on the side of setting overly aggressive goals out of fear that they won't be challenging enough, and they're enticed by Google's unrealistic "roofshots," "moonshots," and other inspirations. Needless to say, if nobody achieves their goals, the initial impression of the practice is terrible, the image of OKRs is tarnished, and people begin to think that not meeting goals is the norm—or worse, they abandon the practice altogether.
The right approach is to set conservative OKRs to start with: levels that aren't a big stretch but aren't ridiculously easy either. After a few cycles of consistently achieving goals, the company can slowly start stretching its goals, in a process of fine-tuning.
Once a high level of maturity is reached with goal-setting practices (where the company as a whole achieves 70 to 80% of its goals, misses 10%, and exceeds 10% consistently), people should start having the freedom to set really stretched goals per cycle, but no more than that.
Google, the world's most famous company practicing OKRs, sends mixed signals in this regard: it essentially preaches that 70% achievement is a goal met (although Google's VP of People Operations disagrees with us, essentially, 70% is the new 100% at the company). Its leaders believe that systematically creating harder goals (aiming for 100% to achieve 70%) throughout the company will produce better aggregate performance. We think that, in practice, people will consider 70% the new 100%, and therefore, the effect will be neutralized. If not, as science also suggests, people become systematically frustrated, and Google wouldn't be the success it is.
Goals that are too easy
Alternatively, goals should not be too easy. Research indicates that very easy goals lead to low levels of motivation and energy due to the lack of challenge they present to employees:
"One of the most consistent findings about the level of goal difficulty is that when goals are too low, people often achieve them, but subsequent levels of motivation and energy typically plummet, and the goals end up being exceeded by only a small amount." In other words, people adjust their efforts to minimize their energy expenditure.