When a target becomes the task
A metric is usually a proxy: an observable stand-in for something harder to measure. A hospital might track how quickly patients are seen because timely care matters; a school might count test results as one indication of learning. The proxy is useful precisely because the real goal—good care or durable understanding—is broader and less easy to tally.
Trouble begins when rewards, rankings, or penalties hinge on that one number. People naturally pay attention to what their organization notices. They may reorganize work around the measured part, even without anyone deciding to cheat. A school under pressure to raise scores, for instance, can devote more time to tested material and less to subjects the score leaves out.
Why the blind spot grows
Once a target carries consequences, the number stops being a neutral report and becomes a signal about what matters. Staff learn which actions move it. Meanwhile, unmeasured work can become harder to defend: it takes time, but does not improve the score that determines recognition or resources.
This can also change what gets recorded. If a measure depends on how cases are classified, borderline cases may be categorized in ways that improve the result. That may satisfy the formal rule while making comparisons less informative. The system then sees an increasingly polished picture of performance, not necessarily a more complete one.
Keeping the number in perspective
Metrics are most useful as clues to investigate, not as complete verdicts. Pairing a target with other indicators, checking how work changes in response to it, and examining individual cases can reveal what a single score misses. The practical question is not only “Did the number improve?” but “What changed to make it improve—and what important work might that change have displaced?”
That question matters because a clean trend line can reflect genuine progress, a narrower definition of success, or both. Without looking beyond the target, those outcomes can appear identical.
