Before the Dashboard: How to Set Up Business Data

A trustworthy dashboard is the last step of setting up your business data. Here is the whole process, and the order that makes it work.

Set up your data before the dashboard

The chart is the last step, not the first.

Management asks for graphs and clean end numbers, and that is a fair thing to want. But a graph is only a mirror of the data feeding it. When two people record revenue in two different ways, or half the figures live in someone's inbox, the mirror shows a distorted picture.

Here is the split that causes most of the friction. Management sees the top, the chart, and assumes that is where the work happens. The person building the report knows the truth, that the real work sits in the steps underneath, where the numbers are defined, stored, and cleaned.

A dashboard cannot fix data that was never set up to be trusted.

The fix is not a better chart. It is a process, built in order.

The process is bigger than one dashboard

Getting to a number you trust takes six steps, not one.

The formal versions of this are longer than most founders expect. The data lifecycle is often described as eight stages, and the business intelligence cycle as six. Stripped down to what a small business actually needs, it comes to six steps. Management only ever sees the last of them. Everything before is the foundation.

Here is the whole process, in order.

Notice where the dashboard sits. Dead last.

The step everyone skips

Most bad numbers are not an analysis problem. They are a cleaning problem.

Between storing your data and analysing it sits the step almost every small business skips: cleaning and validation. In the formal models it has its own name, the transform in extract, transform, load. It is where duplicates, typos, and half finished records get caught before they reach a chart.

Skip it and your dashboard will faithfully show you the wrong number, beautifully formatted.

Clean data is not optional. It is the difference between a chart and a guess.

Look again at the agency. Its revenue looked wrong because one client sat in the data twice, once under a full company name and once under a short version. No tool would have caught that on its own. A few minutes of cleaning did.

Which is why the order matters far more than the tool.

Why the order is the whole game

You cannot analyse your way out of a broken foundation.

Each step depends on the one before it. Numbers you never defined cannot be captured consistently. Data scattered across five places cannot be cleaned. Data you have not cleaned cannot be trusted in analysis. A dashboard built on any of those breaks quietly, and usually at the worst moment.

This is why buying a fancier platform rarely fixes anything. The tool is step six. The problem is almost always somewhere in steps one to five.

So do not start with the tool. Start with the first step.

Where to start, and what comes next

You do not need the whole system today. You need step one.

Here is the smallest useful start you can make this week.

That is steps one, two, and four in miniature. The rest builds on top of them.

Over the next weeks we go through each step properly, one post at a time, from deciding what to measure, to capturing and storing it, to cleaning it, to building the dashboard that finally tells the truth. This post is the map. The next one starts at step one.