OKRs — objectives and key results — have become one of the most popular management methodologies of recent decades, adopted by companies of all sizes seeking to align the organization around what matters. The idea is elegant: define ambitious, qualitative objectives that inspire a direction, and associate with each measurable key results that say, without ambiguity, whether we are getting there. But there is an enormous difference between adopting the OKR structure and using it well — and that difference lies, almost entirely, in the quality of the data underpinning the key results.
An OKR is only as good as its ability to be measured honestly. The objective gives the inspiration and direction; it is the key results, with their numbers, that turn that inspiration into something concrete and verifiable. And it is precisely here that most OKR implementations fail: the key results are vague, are not measured rigorously, or measure what is easy instead of what matters. A data-driven OKR — in which each key result rests on a reliable and meaningful metric — is what separates the methodology that transforms an organization from the one that merely adds another layer of bureaucracy.
This article is about how to make OKRs really work, ensuring their key results are metrics that matter and that can be measured honestly.
The anatomy of a good OKR
An OKR has two parts with distinct and complementary functions. The objective is qualitative, ambitious and inspiring — something like "make the customer experience truly memorable". It is the "where to" and the "why", the direction that mobilizes people. It is not measured directly; it serves to give meaning and energy. On its own, however, an objective is just a good intention, because it does not say whether we are getting there.
This is where the key results come in: two to four measurable indicators that define, with no room for interpretation, what it would mean to achieve that objective. If the objective is a memorable customer experience, the key results might be a concrete rise in measured satisfaction, a reduction in response time, an increase in the retention rate. The key results are the "how we will know", and it is their precision and reliability that give substance to the OKR. Without good key results, an inspiring objective is a dream with no way to know whether it came true.
The fatal mistake: key results that are not real metrics
The most common failure in implementing OKRs is having key results that seem measurable but are not, in practice, honest metrics. This happens in several ways. Sometimes the key result is an activity disguised as a result — "launch the new feature" instead of "reach a certain level of feature adoption". Other times it is vague to the point of being useless — "improve satisfaction" without defining how it is measured or the target value. And often what is easy to measure is measured instead of what truly reflects the objective, giving a false sense of progress.
This is the point where data quality becomes decisive. A key result is only useful if it rests on a metric that is reliably measurable, that genuinely reflects the objective, and that the team can influence with its work. Defining good key results is, at heart, the same challenge as defining good metrics — with a clear owner, comparison and a real link to what matters. An OKR with poorly defined key results is not an OKR; it is a wish list with the appearance of rigor.
The principles of a solid key result
- Actually measurable: it rests on a metric you can measure reliably and on time, not on an impression.
- A result, not an activity: it measures the effect we want to achieve, not the task we will do to get there.
- Linked to the objective: it genuinely reflects what it means to meet the objective, not just what is convenient to measure.
- Actionable: the team can influence it with its work, which makes it motivating instead of arbitrary.
Ambition and honesty at the same time
One of the most valuable and most misunderstood characteristics of OKRs is that they should be ambitious to the point of not being fully achieved. The original philosophy is that reaching a hundred percent of all key results probably means the objectives were too modest. This has an important implication for data: a data-driven OKR requires a culture in which not fully reaching an ambitious key result is not a failure to hide, but valuable information about what is realistic and where the limits are.
This combination of ambition and honesty is only possible if the measurement is true. If people feel they will be punished for not reaching a hundred percent, they will set modest, easy key results, or manipulate the measurement — and the power of OKRs is lost. The reliable data and the culture that accompanies it are what allow real ambition: stretch goals, measured honestly, where reaching eighty percent of an ambitious goal is celebrated as progress, not hidden as failure.
A concrete case
A company adopted OKRs with great enthusiasm, attracted by the promise of alignment and focus. But after two or three cycles, the enthusiasm had given way to cynicism — people saw OKRs as a form-filling exercise that changed nothing. When they analyzed why they were not working, the cause was clear: the key results were poorly defined. Many were activities disguised as results ("implement the new system") instead of results; others were vague to the point that anything counted as success; and several measured what was easy instead of what reflected the objective. At the end of each cycle, no one knew for sure whether the OKRs had been met, because there was no reliable data to say so — and that ambiguity made the whole exercise empty. The company reworked its approach, focusing on the quality of the key results. For each objective, they forced themselves to define key results that were real metrics — measurable with reliable data, genuinely linked to the objective, and influenceable by the team. They invested in ensuring each of those metrics was actually measured rigorously, instead of estimated haphazardly at the end of the quarter. The transformation was remarkable. Suddenly, OKRs stopped being a form and became a concrete conversation about real progress: each team saw, with data, how much it was advancing toward its key results, and that clarity generated genuine focus and energy. Cynicism gave way to commitment, because people trusted that the numbers told the truth and that effort was reflected in them. The methodology had not changed; what changed was the quality of the data underpinning it.
OKRs as data discipline
At heart, implementing OKRs well is, to a large extent, an exercise in data discipline. It forces the organization to precisely define what it wants to achieve and how it will measure it, to have reliable metrics to track progress, and to look honestly at the results. These are exactly the skills of a data-driven culture, and that is why OKRs and a company's analytical maturity go hand in hand — each reinforces the other.
Seen this way, OKRs are not just an imported management methodology; they are a way to institutionalize the habit of linking objectives to metrics that matter, and of deciding and adjusting based on what the data shows. A company that does its OKRs well is, almost by definition, exercising the discipline that makes it data-driven in everything else.
In practice
If your company adopted OKRs but they became an empty form-filling exercise, the problem is probably in the key results. Look at them with critical eyes: are they real metrics, measurable with reliable data, linked to the objective and influenceable by the team? Or are they disguised activities, vague or easy to measure but irrelevant? Improving the quality of the key results — making them data-driven — is usually the difference between OKRs that transform and OKRs that only decorate. Do your OKRs measure what matters with data you trust, or are they wishes with the appearance of rigor?
