How do you know your best-performing team member is your best-performing team member?

If the answer involves words like "dedication", "energy" or "always busy", you are answering a different question. You are describing what is visible, not what is working. That substitution is one of the most reliable failure modes in business, and it is worth stating as a plain truism: when success can't be measured, organisations fall back on signals that are visible — activity, hours, spend, confidence — and end up rewarding failure dressed as effort. This article is about why that happens, how to tell a vanity metric from a business metric, and the measurement lens we use to keep the two apart.

The measurement vacuum

When no agreed measure of success exists, something fills the gap — and it is never a fair replacement. Three things rush in:

  • Busyness metrics. Hours logged, tickets touched, meetings attended. Easy to count, mostly measuring presence. A team can be visibly busy and commercially worthless; the dashboard will not show the difference.
  • HIPPO decisions. When evidence is absent, the Highest Paid Person's Opinion decides. Sometimes that opinion is good. But an organisation that cannot tell good outcomes from bad cannot tell a good opinion from a confident one, and confidence is the easier quality to fake.
  • Sunk-cost attachment. The longer an unmeasured initiative runs, the harder it becomes to stop — because nobody can say it is failing, and stopping it would admit the spend produced nothing. Unmeasured work does not just waste money; it protects itself from ever being judged.

Vanity metrics vs business metrics

The vacuum persists because many numbers look like measurement while behaving like decoration. The test is simple: does the number connect to something the business actually wants — enquiries, orders, margin, retention — within a timeframe a decision can respond to? "Impressions" fails that test for most businesses. "Enquiries received per week, split by source" passes it. Neither number is worthless, but only one of them can end an argument. Honest measurement is the discipline of leading with the second kind, and keeping the first kind in its place: context, not verdict.

The PHAT lens

At Cooper Agile we organise measurement around four layers, memorable as PHAT:

  • Precision. Agree the exact numbers and acceptance criteria before work starts. What will count as success, by when, measured how — written down and signed. Precision removes the retrospective renegotiation that unmeasured work always attempts.
  • Heuristic. Fast, practical signals a human can scan weekly: enquiry counts, conversion counts, queue depths, error rates. Heuristics are deliberately rough — they exist to surface questions early, not to answer them precisely.
  • Analytical. Periodic deep dives: trends over quarters, comparisons across cohorts or channels, root-cause analysis when a heuristic moves. Analysis is scheduled, not heroic — it happens because the calendar says so.
  • Tracking. Continuous instrumentation so evidence accumulates without effort: the dashboards, logs and feeds that feed all three layers above. Good tracking is invisible when it works and its absence is discovered only after the fact.

Where does the plumbing run? On the delivery stack the industry has long standardised around — the combination of social, mobile, analytics and cloud tooling, commonly abbreviated as SMAC. The vocabulary matters less than the habit: measurement is a property of the system, not an after-the-fact report.

The caution that keeps measurement honest

There is a named warning every measurement programme should carry into its first meeting. Goodhart's law — articulated by economist Charles Goodhart, and popularised in the sharper formulation often attributed to Marilyn Strathern — holds that when a measure becomes a target, it ceases to be a good measure. The moment a number decides bonuses, people optimise the number. Support tickets closed becomes support tickets rushed. Enquiries captured becomes enquiry forms nagging. This is not cynicism; it is what systems do. The defence is to pair every target with a second measure that punishes gaming, and to review both together.

Illustrative example: a services firm runs a customer-success team and judges it on one number: tickets closed per agent. The number improves steadily for three quarters. Meanwhile renewal conversations, unmeasured, quietly deteriorate — agents have learned that closing quickly beats resolving thoroughly, so issues bounce back or go quiet. An honest measurement pass re-bases the team on paired metrics: tickets closed and issues reopened within thirty days, plus a quarterly renewal trend reviewed analytically. Within two cycles, the gaming route is closed, reopened issues halve, and the renewal line recovers. Nothing about the people changed; the measures stopped rewarding the wrong behaviour.

Constraints and trade-offs

Three honest limits:

  • Not everything valuable is measurable. Brand trust, goodwill, judgement — real assets with no reliable number attached. Measurement should inform decisions, not pretend to settle them. The unmeasurable deserves honest discussion, not forced quantification.
  • Measurement can be gamed. Goodhart's law is always in force. Every measure you elevate becomes a target someone may optimise superficially; budget the pairing and review habits that keep numbers honest.
  • Start small. A weekly heuristic scan of three business numbers beats an analytics platform nobody reads. Measurement systems fail by over-building first and under-using forever.

Your action checklist

  • Name the three most important things your business does, and the number — if any — attached to each.
  • Separate your current metrics into vanity and business piles, honestly.
  • Pick one initiative and write its acceptance criteria before it continues.
  • Stand up a weekly heuristic scan: three numbers, one human, fifteen minutes.
  • Pair every target metric with a second metric that punishes gaming.
  • Schedule one analytical deep-dive per quarter, on the calendar.

If your business runs on visible signals instead of honest ones, our reporting, analytics and monitoring service builds the tracking, heuristics and review cadence that replace them — Precision first, PHAT throughout. And if you want an independent check on the numbers you already trust, our technical oversight reviews measurement systems the same way it reviews everything else: against evidence, not confidence.