Questions before tables
What each report must answer, who reads it and which decisions depend on it.
A data warehouse earns its keep when figures live in several systems and someone spends days each month stitching them together in Excel. It does not replace e-conomic, Business Central or your online shop. It collects data from them into one shared picture that Power BI can present.
We start from the questions management wants answered. The data model is designed only once the list of reports is agreed.
What each report must answer, who reads it and which decisions depend on it.
Your finance system via its API, a Shopify or WooCommerce shop, the CRM, Google Analytics 4 and those spreadsheets nobody can live without.
Azure SQL or Microsoft Fabric in a European region, so personal data stays inside the EU and access is governed through Entra ID.
Customers, products and departments are matched across systems so a customer is only counted once.
Data is pulled overnight, keeping the load off source systems while staff are at work.
Reports and dashboards with permissions per department, so each manager sees their own numbers.
The first report is usually ready within four to six weeks. Later ones are faster because the foundation is already in place.
We start with the two reports that currently eat the most hours of manual work.
Data model, matching of customers and products between systems and a fixed refresh schedule.
Loading, the report itself and a reconciliation against the figure finance has so far calculated by hand.
New reports are added on the same foundation as needs arise.
The first test is whether the number matches the old spreadsheet. If the report is even slightly off, people quietly keep their own files. So we always reconcile against the previous calculation and explain every difference before the report is shared across the business.
Built-in reports only see the system's own data. Once sales need to be viewed alongside web traffic, campaigns or CRM leads, they fall short. Heavy queries against the live database can also slow the system down for everyone.
Many organisations manage with Power BI Pro for the people who view reports and a small Azure SQL database. Fabric capacity becomes relevant with larger data volumes or many readers. We cost both options before you choose, and build hours are billed as agreed or at DKK 895 per hour excl. VAT.
For management reporting one overnight refresh is usually enough. Near real-time loading costs more and puts pressure on source systems. Begin with a daily cycle and increase frequency only where there is a genuine need, such as stock levels in an online shop.
List the reports that are assembled by hand today and the systems the figures come from. We will build the first one and reconcile it with your own calculation.
Thank you for getting in touch
One of our consultants already has it. Expect a reply within the working day; anything urgent goes straight to an engineer.
That city is not on our list. Check the spelling or pick the nearest larger town.