OKRs tell you what you want. They do not tell you what is true.
OKRs are the latest turn in a forty-year pattern: goal alignment repackaged as performance measurement. The case below is the one a measurement architect would make — that OKRs, like the Balanced Scorecard before them, close a different gap than the one they are often bought to close.
What OKRs do well
Objectives and Key Results are a goal-setting framework, and they do what goal-setting frameworks do well: focus attention, align teams around shared priorities, and create a common language for what an organisation is trying to achieve in a period. For that purpose, they work — and their advocates are right that clarity of intent is worth a great deal.
The tension begins when Key Results are read as a measurement system — assumed to tell you whether the processes are healthy and whether a crisis is building. That is a different question from whether the objective was met, and it is the reading this piece takes issue with.
OKRs tell you whether you reached your objective by the end of the period. A measurement system tells you whether your operational processes are performing within the bounds required to remain viable — continuously, not in hindsight. Both questions are worth asking. Only the second detects a problem before it becomes a crisis.
The structural argument
The claim OKR advocates would contest is this one: that goal-alignment systems share an architectural ceiling. Key Results are metrics attached to objectives. They are not calibrated against process history, they carry no natural process limits, and they do not aggregate upward into a confirmed strategic claim. On this account they produce completion records at the end of a period, not signals during it.
An organisation can reach most of its Key Results while its core processes deteriorate — retention falling, supply reliability dropping, reserves eroding — if none of those processes are structurally connected to the framework. A defender of OKRs would answer that this is a misuse, not a flaw: OKRs were never meant to carry operational monitoring, and an organisation that expected them to was holding the wrong instrument. That answer is fair, and it is also the whole point.
Objective: enter three new markets. Key Result: sign fifteen distribution agreements by period end. At review: eleven signed, graded a success. Meanwhile retention fell sharply and the existing network degraded. The framework did not measure those, because they were never set as objectives.
A stability area monitored continuously. A retention indicator crosses its lower limit early in the period. An investigation opens, a cause is found, a response is deployed before the review meeting. The deterioration is caught while it is still small.
The same problem, a different brand
OKRs came out of the semiconductor industry in the 1970s and spread through technology firms in the 2000s. They refine Management by Objectives, the 1954 framework that began the goal-alignment tradition. The vocabulary changed; on this reading the architecture did not.
Goal-alignment frameworks tell an organisation what to aim at. They do not, by themselves, tell it whether its foundations are holding while it aims. A runner fixed on the finish line who cannot feel a failing knee finishes injured, or not at all. The finish line is not the problem. The absence of anything watching the knee while the runner runs is the problem.
The counter-argument deserves its place: a well-run OKR cycle includes health metrics, and a disciplined team will notice the knee. True. The structural point is only that nothing in the framework requires it — the watching is left to judgement, and judgement is exactly what fails under pressure.
Not mutually exclusive
This is where the honest version of the argument lands, and it is not a dismissal. OKRs communicate strategic intent and focus teams on the destination. A continuous measurement stack confirms whether the foundations required to reach it are holding. These are different functions, and both are legitimate.
An organisation can set direction with OKRs and monitor roadworthiness with a stack. The OKR tells the team where to go. The stack tells the manager whether the vehicle can make the journey. Neither replaces the other, and anyone selling you one as the other — in either direction — is overselling.
A regional organisation ran OKRs for two years with achievement rates consistently high. In the third year a programme collapsed — not because the objectives were wrong, but because the staff capacity needed to deliver them had been eroding quietly for eighteen months. Two senior field officers left. Their knowledge was never documented. The next cohort lacked the relationships that community trust depends on.
None of this was visible in the framework. Retention was not a Key Result. Knowledge retention was not an objective. The OKRs measured deliverables, not the conditions required to produce them.
A continuity area with a module for staff capacity would have flagged the pattern within the first few months of an eighteen-month slide, leaving well over a year to respond. Instead there was a collapse and a funder review. A committed OKR practitioner might have caught it too — but nothing in the framework obliged anyone to look.
อาหารสั่งกลับบ้าน
OKRs are a goal-alignment system, not a measurement system. They tell you what you achieved by period end. They do not, on their own, tell you whether the foundations held while you worked.
Key Results are targets, not calibrated thresholds. Without calibration from process history and natural limits, they surface failure — not the drift that precedes it.
OKRs and a measurement stack serve different functions. One sets direction. One monitors whether the organisation is in condition to travel it. Both legitimate. Neither a substitute for the other.
The foundations OKRs leave unmonitored are the ones that cause collapse. Staff capacity, supply reliability, reserves, institutional knowledge. A stack watches them continuously; a goal framework is not built to.
The full argument is in Book III.
It sets out why goal-alignment systems and measurement systems are different instruments — and why an organisation needs both.
Get the Trilogy on Amazon →MetriqOne · หลักฐานที่ปราศจากสัญญาณรบกวน · metriq.one