From Multiple Data Sources to Azure Databricks: A Zero-Downtime Power BI Migration for Future AI Readiness

Most growing organizations do not set out to fragment their data. It happens one decision at a time. A new ERP for finance. A CRM for sales. An HR system for people. An accounting tool for a specific entity. Each one solves a real problem – and each one becomes another separate source of data.

This project started with our own reporting. Like many growing services organizations, CloudFronts runs its business across several systems: Business Central and Finance & Operations for ERP, Dynamics 365 CRM for clients and engagements, Zoho People for HR, and Zoho Books for accounting. Leadership, finance, HR, and delivery teams rely on more than 40 Power BI reports built on these systems, from management dashboards and HR reports to the Steering Committee report we use to track quarterly meetings with client accounts.

Over time, each system had been connected in its own way. Some data moved through Azure Data Factory, some through SQL Database, and some through Azure Functions. Reports connected directly to their sources, with business logic built inside each one, and they sat across separate workspaces for HR, management, and other teams.

The idea behind the project was simple: instead of continuing to add pipelines and reports system by system, bring all business data into one trusted platform on Azure Databricks, migrate every existing report onto it without disrupting the people who use them, and build a foundation that is ready for AI in the process.

None of the old setup looked broken on any given day. But the costs were real, and they compounded as the business grew.

One data platform is not just a new database or a bigger data warehouse. It is a change in how the organization treats its data – from something each system and report manages on its own, to a shared business asset managed once, for everyone.

Functionally, it comes down to four commitments:

For leadership, the value of one data platform shows up in outcomes rather than architecture.

We built our one data platform on Azure and Databricks. Here is how it fits together, from five source systems to a single, governed Power BI workspace.

Step 1: Connect the Source Systems
Business Central, Finance & Operations, Dynamics 365 CRM, Zoho People, and Zoho Books are registered once in a central ingestion list that tells the platform what data to bring in.

Step 2: Bring Data In Automatically
Azure Logic Apps read that list and pull the data from every system into Azure Databricks. All raw data lands in one place, the Bronze layer.

Step 3: Keep Only What Matters
In the Silver layer, only the data needed for reporting is kept, then cleaned and standardized so every source looks consistent.

Step 4: Apply Business Rules Once
In the Gold layer, calculations, groupings, and connections across systems are applied once, creating trusted, business-ready data.

Step 5: Run One Pipeline for Everything
A single Databricks job moves data from all sources through Bronze, Silver, and Gold. There is no separate pipeline for each system.

Step 6: Move Reports onto the Platform
Existing Power BI reports are pointed to the Gold layer instead of their original sources. Layouts, filters, and visuals stay exactly the same.

Step 7: Validate Before Switching
Each migrated report is compared with the original to confirm the numbers match, so users are never disrupted.

Step 8: Control Access by Role
All reports live in one Power BI workspace, and Microsoft Entra ID groups such as HR, Management, and Finance decide who can see which reports.

The hardest part of moving to one data platform is rarely the technology. It is the risk of disrupting the people who rely on existing reports every day. A traditional migration rebuilds each report by hand – slow, inconsistent, and visible to every user.

Because our framework is configuration-driven, migrating a report became a repeatable routine rather than a rebuild: register the source data, select the fields the report needs, define its business logic once, run a single pipeline, and repoint the report to the curated data. Each migrated report was validated against the original before switch-over, so the numbers matched exactly.

Take our Steering Committee Meeting Report, which tracks quarterly steering committee meetings across client accounts. It originally connected directly to Dataverse, with its logic built inside the report. After migration, it reads the same information from the Gold layer in Azure Databricks. Same layout, same filters, same visuals – users noticed nothing, except that the report now sits on a foundation shared with every other report.

Here is the part that matters most for the years ahead. Most AI initiatives do not fail because of the model. They fail because of the data. An AI assistant cannot answer a question about customer profitability if customer data lives in the CRM and invoice data lives in two different accounting systems. A forecasting model cannot be trusted if the history it learns from is calculated differently in every report.

One data platform removes that barrier. When data from every business system is already unified, cleaned, governed, and connected in the Gold layer, AI use cases can build directly on top of it. There is no separate data consolidation project to fund first, and no debate about whose numbers are right.

In other words, the work done to fix reporting is the same work that makes the organization AI-ready. Every report migrated onto the platform is also another trusted dataset available for AI.

With data from ERP, CRM, HR, and finance systems in one governed platform, a new set of use cases becomes practical. These are the kinds of questions a unified foundation makes it possible to answer:

This is not theoretical for us. Our Project Risk Engine is an example of what becomes possible once business data sits on Azure Databricks: it reads project, billing, and communication data to score project health and answers questions in plain English through Databricks Genie.

If you are deciding whether your organization needs one data platform, these questions are a useful starting point:

Scattered data is a quiet cost. It shows up as reports that disagree, logic rebuilt in every dashboard, governance that depends on individuals, and AI ambitions that never quite get started.

One data platform changes the equation. By bringing every business system into a single, governed foundation on Azure and Databricks, we unified five systems, migrated 40+ Power BI reports without disrupting a single user, and created a trusted base for the AI use cases that come next.

We built it to solve our own problem first. The lesson we would share with any leadership team is simple: the path to AI runs through your data foundation. Get the data into one trusted place, and the use cases follow.

Watch the Webinar

See the framework in action. In our live session, we walk through the platform architecture, how reports are migrated without disruption, and a live demo moving a Power BI report from Dataverse to Azure Databricks.