Collecting data and knowing what to do with it are two completely different jobs.

The average book has around 300 pages. A city contains roughly 4 million windows. A person sighs about 500,000 times over a lifetime. The average Nobel Prize winner stands 178 centimetres tall.

These are real, verifiable facts — and utterly useless for any decision any team will ever have to make. Not because they're false. But because they answer no question worth answering.

What gets counted doesn't always count

In 1963, the sociologist William Bruce Cameron wrote something that stuck: "not everything that can be counted counts, and not everything that counts can be counted." The line is often attributed to Einstein — which tells us something about our tendency to trust big names when an idea challenges us. But who said it isn't what matters. What matters is that it's still true, and we still ignore it systematically.

We collect more data than ever. Analytics tools are everywhere, cheap, and easy to use. And we make decisions of questionable quality at a rate that should worry us more than it does.

Measuring is not the same as knowing

Confusing measuring with knowing is expensive. A team that piles up dashboards full of numbers gets the feeling it's in control of the business. There are meetings where charts are presented, everyone nods, and by the end no one knows what to do next. Not because the data is missing. Because the wrong data doesn't answer the right questions.

The problem isn't a lack of information. It's the absence of purpose in choosing what to measure. Any analytics system produces dozens of metrics effortlessly. But a metric that informs no concrete decision is just noise with good presentation.

Data without purpose isn't data, it's decoration

There's a fundamental difference between measuring because you can and measuring because it matters. The first is easy: the numbers show up on their own. The second demands a prior, harder question: what do we need to know in order to decide better? If there's no clear answer to that question, any number you collect is just noise dressed up as information.

The trouble is that today's tools make collecting data so easy that the choice is no longer "what will we measure" but "what will we ignore." And ignoring data that already exists feels like waste, even when that data is good for nothing. The result is the steady accumulation of metrics that no one uses but no one removes, because removing them feels irresponsible.

This build-up has an informal name inside organisations: the dashboard that grew. It started with three indicators that mattered. Over time, others were added, by different people with different intentions. Today it has thirty-two. And no one quite knows what to do with most of them.

The cost of measuring badly

Measuring without purpose isn't harmless. It has concrete costs that rarely appear in any report.

The first is the cost of attention. Every metric on the panel competes for the attention of whoever reads it. The more metrics there are, the less attention is left for the ones that actually matter. Miller's Law, formulated by the psychologist George Miller in the 1950s, shows that the human mind can actively process around seven items at once. A dashboard with thirty indicators isn't giving you thirty times more information. It's guaranteeing that none of them gets the attention it deserves.

The second is the cost of trust. When teams spend time collecting, cleaning, and presenting data that informs no decision, they learn — rightly — that measuring is bureaucracy. And when the moment comes to measure what really matters, no one believes it's worth the effort anymore.

In short

Collecting data keeps getting easier. Working out which data matters remains hard. And most organisations invest in the first without investing in the second, piling up metrics that create the illusion of control without informing a single real decision.

The problem isn't the tools. It's the absence of a prior question: what do we need to know in order to decide better? Anything that doesn't answer that question is noise, however pretty it looks on a chart.

Next time you're about to define a new metric, start with this question: if this number changes, do I know what to do next? If the answer is no, the number is good for nothing.

‍

This article was originally published on Tek Notícias