Home / Collections / The numbers say one thing, the room says another

Pathway

The numbers say one thing, the room says another

When what you're measuring and what you're experiencing don't match.

You’re measuring things carefully. The reports are thorough, the dashboards are live, and the targets are being hit. But something doesn’t match what people are experiencing. This gap between what the data says and what the room feels isn’t a mystery - it’s one of the most common dynamics in any system that measures its own performance. This pathway explores how measurement shapes what it measures, and what to do when the map and the territory disagree.

01

The starting point for understanding why metrics misbehave. When a measure becomes a target, people optimise for the measure rather than the thing it was supposed to represent. The number goes up. The reality it was tracking doesn't. This single dynamic explains more measurement dysfunction than any other concept.

Measurement, signals, and sense

Goodhart's Law

When a measure becomes a target, it ceases to be a good measure

02

Most of what we measure isn't the thing we care about - it's a stand-in. Revenue is a proxy for value. Engagement scores are a proxy for commitment. Test results are a proxy for learning. Proxies are useful and often necessary. The trouble starts when people forget the difference between the proxy and the thing it represents.

Measurement, signals, and sense

Proxy Measures

Using something measurable as a stand-in for something that isn't - useful until people start gaming the proxy

03

What happens when the forgetting becomes permanent. A surrogate measure is a proxy that's been promoted to the real thing - everyone acts as if the number IS the outcome, not a rough indicator of it. The original purpose fades from memory. You end up optimising for a number that's drifted away from what it once meant.

Measurement, signals, and sense

Surrogate Measures

When you can't measure what matters, you measure what you can - and then forget the difference

04

When the measurement system rewards the opposite of what you intended. Not because anyone designed it badly, but because the incentive structure interacts with the system in ways that produce unwanted behaviour. People respond rationally to the incentives in front of them - the problem is what those incentives are pointing at.

Measurement, signals, and sense

Perverse Incentives

Incentive structures that reward the opposite of what you intended - the cobra effect

05

A different kind of mismatch. Lagging indicators tell you what already happened. Leading indicators tell you what's coming. Most dashboards are full of lagging indicators - which means by the time the numbers change, the moment to act has passed. The room often senses what's coming before the dashboard does. That's not a flaw in the room - it's a flaw in the dashboard.

Measurement, signals, and sense

Leading vs Lagging Indicators

Lagging indicators tell you what happened. Leading indicators tell you what's coming. Most people only track the first kind

06

When you do have good measures, you still face the challenge of distinguishing what matters from what's random variation. Most of what looks like signal in a complex system is noise. Reacting to noise creates its own problems - chasing phantom patterns, oversteering, burning energy on things that would have corrected themselves.

Measurement, signals, and sense

Signal vs Noise

The challenge of distinguishing meaningful information from random variation - most of what looks like signal is noise

07

The act of measuring changes what you're measuring. When people know they're being watched, tracked, or scored, their behaviour shifts. The data you collect is partly a reflection of the system and partly a reflection of the system's response to being observed. Neither is wrong - but they're not the same thing.

Measurement, signals, and sense

Observer Effect

The act of measuring or watching a system changes how the system behaves

08

The wider picture. When each part of a system optimises for its own metrics, the whole can suffer. Every department hits its targets. The organisation misses its goals. The numbers say one thing - each part is performing. The room says another - something isn't adding up. This is often where the gap between metrics and experience is widest.

Boundaries, perspectives, and power

Sub-optimisation

Optimising one part of a system at the expense of the whole - every department hitting its targets while the organisation fails

Explore more

These concepts connect to many others across the knowledge base.