Hundreds of metrics in one set of reports that everyone reads the same way.

Where the customer stood

A strategy consultancy had drawn up a concept for central reporting. The data came from ERP systems, scanners and other sources. Hundreds of metrics, with very different characteristics, were to go into a single shared data foundation. Our job was to present them so that anyone could see what mattered at a glance.

Two groups would work with it. Management wanted the essentials at a glance, while specialists needed the detail. There was also a question for later: how do you keep changes manageable when so many KPIs depend on each other?

What we did

We began with workshops with the management team, which produced reporting standards: every department now sees the same charts in every report. Alongside that we built a data model that underpins the reports, stays fast and supports self-service. A permissions concept shows each person exactly the data for their own departments. Finally, we trained the business users, who now build their own reports.

The data model needed a light touch. It had to give the variety of the business a tidy structure, carry the visuals, stay fast and allow self-service later on.

The permissions logic recalculates on every request. It works out the signed-in user's area of responsibility and shows exactly that data.

What came out of it

Every department reads its reports to the same standards, so sharing a report no longer needs much explaining. Everyone sees exactly the data for their own area.

The business now answers its own questions. New reports follow the same standards and fit in with the rest straight away.

What proved harder than expected

The permissions concept was the hardest part. For each signed-in user it works out how large their area of responsibility is, then shows or hides data accordingly. It had to stay fast, even though it recalculates every time a report is opened.