Improved and new insights into business processes and finances
Ammega
Power BI consultant
We set up Power BI around a single data model, including workspaces, refresh and management, as we did for Ammega.
Power BI dashboards and an app gave Ammega a single source of truth
Ammega wanted better insight into its business processes and greater control over its current situation. Working closely with stakeholders across the organisation, including management, planning and finance, we developed several Power BI dashboards and a Power BI app. That meant fewer separate Excel files and everything in one place.
Why doesn’t anyone trust the figures in our Power BI?
Each department has its own pbix file and formulas, refreshes depend on one colleague’s laptop, and no one knows which report is accurate. A Power BI consultant brings everything together in one model, with defined access rights and someone responsible for it.
In a Power BI implementation, we set up the model, workspaces, data refresh and management
One semantic model
A star schema with DAX measures that powers every report. For Ammega, we brought the dashboards together in a Power BI app, with a single source of truth.
Workspaces, roles and apps
Workspaces for development, testing and production, access through Entra groups and an app for each audience, such as management, planning and finance at Ammega.
Gateway and data refresh
A shared gateway for on-premises sources and a fixed refresh schedule. For Vivara, we built the underlying data platform, using data from SAP, Magento and SharePoint.
Management after go-live
The model and reports are version-controlled as a Power BI project, with new versions released through deployment pipelines and an owner for each report. Your team can manage them, or we can.
We get one model right first, then expand
-
Map the environment
Which reports, sources, licences and gateways exist, who uses them and who currently manages them.
-
One model, one report
We build the model and an initial, widely used report, then check the figures with the people who use them to make decisions.
-
Roll out and manage
The workspaces, app and deployment pipeline go live. The other reports then follow, using the same model.
Should we set up Power BI, or can our own team do it?
A good fit if
Your Power BI setup has grown from separate files and a different model for each department, and you want one configured environment with consistent definitions, access permissions and data refreshes that do not depend on one person.
Not a good fit if
The data sources are not yet accessible, or no one owns definitions such as revenue or customer. In that case, start with data engineering or data governance, then move on to Power BI.
What determines the scope
The number of data sources and whether they are in the cloud or on-premises, how many existing reports need to be included, what licences or capacity you have, and whether your team will take over management.
What about our existing reports?
You do not have to discard them. Move the most-used reports to the new model and agree with each owner which reports to retire.
Have Power BI set up, bring someone into your team or build on it yourself?
Have Power BI set up
Choose this if you want a configured environment with a model, workspaces and management, delivered by one provider. We recommend it for an initial setup or an environment that has grown over time.
A BI specialist on your team
Choose this if your team wants to retain ownership but needs extra capacity or expertise in DAX and models. See hire a BI consultant.
Build on it yourself
Choose this if your team knows Power BI well and can handle the setup itself. Microsoft’s Fabric adoption roadmap helps you agree on ownership and support.
As a Microsoft Partner, we set up Power BI and the platform behind it
Data Science Lab has worked with data and AI since 2016 and is a Microsoft Partner for Fabric, which brings together data engineering, data science and business intelligence. We built Power BI reporting for Ammega and a data platform with Power BI for Vivara.
Your current Power BI environment is the starting point for our first conversation
Stef Stoelwinder
Business DevelopmentStef is your first point of contact. In our first conversation, we’ll discuss your reports, data sources and licences, and what needs sorting out first.
E-mail info@datasciencelab.nl
Phone +31 (0)20 3636 724
Which part of our Power BI environment should we improve first?
Tell us which reports, data sources and licences you have. We’ll reply by email or arrange an initial conversation.
Message sent
Thanks for your message. We’ll be in touch as soon as possible.
Frequently asked questions
A Power BI consultant sets up Power BI so that reports are accurate and remain manageable. That goes beyond creating visuals: they design the semantic model, define calculations as DAX measures, configure workspaces, permissions and refreshes, and agree who manages what.
In our Power BI consultancy, Power BI is part of business intelligence, alongside tools such as Databricks Dashboards, Tableau and Dash. We choose what fits your needs and your existing environment.
More than publishing reports. You set up workspaces, for example by domain and by stage (development, test, production), and grant access through Microsoft Entra ID groups using the Admin, Member, Contributor and Viewer roles. You share reports through an app, with a separate audience for each department.
In the Fabric adoption roadmap, Microsoft describes what is needed beyond the technology: content ownership, governance, user support and a core team, the Center of Excellence.
Usually a star schema: fact tables containing transactions or measurements, such as order lines, and dimension tables such as customer, product and date, which you use to filter and group data. Microsoft recommends this because every visual sends a query to the model.
You define calculations such as revenue or margin once as explicit DAX measures. Reports using the same semantic model then show the same figure. With a separate model for each report, each report calculates its own figures, and the results can differ.
Through the on-premises data gateway: a Windows program on your own network that makes only outbound connections to the Microsoft cloud, without opening inbound ports. You can share the standard gateway with multiple users and data sources. Personal mode is for one user, works only with Power BI and cannot be shared.
If refreshes run on a colleague’s laptop, they stop when the laptop is switched off. For sources on an Azure network, there is the Microsoft-managed VNet gateway.
First, check whether the model loads more historical data or columns than it needs. Incremental refresh can then help: Power BI divides a large fact table into partitions and refreshes only the recent ones. To set this up, you filter on the RangeStart and RangeEnd parameters in Power Query.
Watch out for query folding: the source must apply that date filter itself. If it does not, Power BI still retrieves all the rows. Incremental refresh works with Pro, Premium Per User, Premium and Embedded.
With an app for each workspace. In it, you create audiences, such as management and planning, so each sees only its own reports. If someone should see only their own region in a report, you add row-level security (RLS) to the model.
Viewers need a Pro or PPU licence unless the workspace runs on Premium capacity or Fabric F64 or higher. Anyone who creates or updates the app always needs Pro or PPU.
This is about setting up Power BI as a whole: the model, workspaces, gateway and management. If you mainly need one dashboard to support a decision, see Power BI dashboards. If you want someone to build alongside your own team, read hiring a BI consultant to learn how staff augmentation works.
If your organisation is based in Amsterdam, you can also read about our Power BI specialist in Amsterdam.
When you have a few accurate reports, a regular colleague manages them and no one is asking for more. Or when the source data itself is messy: Power BI can only address the symptoms, so you are better off starting with the data foundation or data governance.
If a simpler solution is a better fit, we will say so. Want to compare providers first? See our overview of data analytics consultancies.