“Tell me how you measure me, and I will tell you how I will behave.”
Eliyahu Goldratt placed this warning in The Haystack Syndrome during a discussion of performance measurement and organizational goals. A metric does not merely observe behavior. Once rewards, status, or scrutiny attach to it, people adapt their decisions to improve the number. If departments are measured on local utilization, they may produce inventory the company cannot sell. If support teams are judged only on speed, they may close cases before problems are solved. Measurement becomes a control system, whether leaders acknowledge that power or not.
The practical lesson is to begin with purpose. Define the system’s goal, identify the few conditions necessary to achieve it, and choose measures that reveal whether decisions improve the whole. A balanced set can expose tradeoffs among throughput, quality, time, safety, cash, and customer outcomes. Metrics should help people learn and coordinate, not merely rank them. When a number is close to a target but distant from the real objective, managers need permission to investigate rather than celebrate.
Goldratt’s formulation should not be read as accusing employees of dishonesty. People often optimize a flawed measure because the organization has told them that the measure is what matters. Goodhart-like effects can arise without manipulation: attention narrows, unmeasured work is neglected, and local success harms the larger system. Some vital qualities, including trust and judgment, resist precise quantification. A dashboard must therefore remain evidence for conversation, not a substitute for responsibility.
Modern analytics make the warning more important. Companies can measure keystrokes, response times, clicks, and conversion at extraordinary detail, creating the illusion that what is countable is what counts. Leaders should test every important metric by asking what behavior it encourages, what it hides, who can be harmed by optimizing it, and how it might be gamed. They should also review measures when strategy changes. Goldratt’s insight is simple but demanding: design the measurement system as carefully as the work system, because the two will soon become one. One practical safeguard is to pair every result measure with a balancing measure and a narrative review. Faster delivery can be considered alongside defects and employee workload; revenue growth alongside returns and customer retention; utilization alongside queue time. The narrative asks what changed in behavior after the metric gained attention. If teams begin postponing difficult cases or redefining categories, the signal is not merely misconduct; it may show that the measure and purpose have separated. Revising a metric can be an act of control, not an admission that measurement has failed.
Goldratt presented the maxim in his 1990 book The Haystack Syndrome while examining information and control in organizations.
He argued that local performance measures must connect logically to the goal of the whole system.
Metrics influence attention, incentives, and decisions as soon as consequences attach to them.
Distortion does not always reflect bad faith; people may faithfully optimize exactly what management chose to reward.
Digital workplaces can measure behavior at unprecedented scale and precision.
Responsible leaders test what each metric encourages, hides, or transfers before using it for judgment or compensation.
Explore more "Quotes of The Day"
Discover more notable quotes from influential voices across politics, science, business, technology, sports, and culture. Each quote offers insight into how ideas, beliefs, and decisions shape the world around us.