Business Data Models & Reporting

One model where every source system comes together and where the meaning of the numbers is settled, with the reporting on top.

Talk through your data model

What a business data model is

A business data model is the layer between your systems and your reporting where the meaning of the numbers is settled. What counts as revenue, when an order is finished, which two customer numbers in different systems are the same customer.

Without that layer every report hangs directly off the systems. A field changes in the ERP and the number changes. Another system arrives and it starts over. With that layer, a new source system is a connection rather than a rebuild.

A dashboard and a data warehouse are both part of it. The product is the agreement about meaning that is fixed inside it.

Two colleagues going through the numbers together on a laptop

The four layers

Extraction

Every system that holds numbers. ERP, CRM, time tracking, webshop, WMS and the spreadsheets that have been running for years. We take the data as it is, with a timestamp on it, so you can always get back to what it said.

Data warehouse

The place where it all comes together. Modelled on how your business works rather than on how the systems store it, because systems get replaced and your business does not.

Definitions & KPIs

One customer list, one item master, one chart of accounts and one definition per term. What revenue is, when an order is finished, how margin is calculated, which KPIs the board follows. Exactly the layer that fails when two departments each give a different answer and the layer that pays back most.

Reporting

Power BI or Microsoft Fabric on top of the model, or the BI tool you already work in. If none of the three fits the question, we deliver it through our own Customer Reporting Platform, with no licence per user. Dashboards, monthly reporting and where it makes sense an application layer where people can actually record something.

What gets handed over

At the end of an engagement there is more than a report. This is what is handed over:

The data model itself, in your own environment and under your control
The definitions, written down and readable by someone who is not a BI specialist
The load processes from source to model, newly built or taken over, with logging so a failed refresh is visible
The reporting, in the number and size set out in the quote
The documentation, the same documentation we use ourselves while building

There is no key that only we hold, so you are tied to nothing. That is deliberate.

When there is already a data environment

Some companies already have a data warehouse, a set of extractions or a reporting environment. Then we look, layer by layer, at what is already there and what happens with it.

What works and is documented stays and gets connected. What is missing, we build. What is in the way, because nobody knows how it works any more or because it cannot be extended, we replace. Which of the three it becomes, per component, is settled at the end of the Business Data Scan, with the price.

Usually the definitions layer turns out to be the part that is missing. A warehouse set up the way the systems store the data can run for years and still not answer what revenue is.