As one of the world’s major copper producers, First Quantum Minerals (FQM) operates long-life mining assets across multiple continents, supporting industries that depend on copper for everything from power grids and wind turbines to electric vehicles and consumer electronics.

FQM describes itself as operating with “an agile approach to work that encourages innovation and helps our projects, and our people, to reach their full potential.” That mindset was reflected in how the company approached modernizing its reporting environment.

Hours Before Insight

For the commercial team at FQM, the workday started by checking if their reporting data was ready. Daily model refreshes took between four and six hours, failed refreshes required manual intervention, and even on a good day, some Power BI visuals took up to two minutes to load.

Although their reporting platform functioned, it was slow and increasingly difficult to maintain.

In mining operations, delays in operational visibility can create ripple effects across procurement, warehouse operations, inventory planning, and broader commercial coordination. FQM needed an analytics foundation that could scale with the business, one that delivered faster, more reliable access to operational insights from day one.

To modernize the platform, FQM partnered with SoftServe to redesign the analytics solution around Databricks and Microsoft Power BI. The goal was a more scalable and resilient operational intelligence foundation.

When Legacy BI Becomes a Bottleneck

For organizations that built their analytics foundations in an earlier era of BI tooling, there comes a point where the architecture has simply reached its ceiling. Incremental fixes stop delivering returns. Redesign becomes the only real answer.

FQM's reporting environment had reached that point. What started as a practical solution had accumulated years of complexity, more data, more business requirements, more reporting relationships, until the weight of it became the problem itself.

Because everything lived inside one large monolithic model, a problem anywhere affected everything. Introducing new business-requested data points required significant effort from specialist resources.

As each layer grew more complex, the whole stack slowed down.

Rebuilding the Analytics Foundation with Databricks and Power BI

SoftServe redesigned the reporting architecture with one goal in mind. Get the heavy computational work out of Power BI and into Databricks, where it belongs.

The solution was built on Databricks' Medallion Architecture, a layered data design pattern that organizes data into progressively cleaner, more refined stages. FQM's team continued to manage the Silver layer, which handled raw data ingestion and initial structuring.

SoftServe's work began at the Gold layer, building clean, optimized, business-ready transformations in Databricks that would serve as the authoritative source for reporting across FQM's three mining sites: FQMO, Kansanshi, and Trident.

The result was a meaningful shift in how work was distributed across the stack.

Calculations and transformations that previously lived inside the SSAS semantic model were moved into Databricks, which took on greater computational load than the legacy system while still reducing refresh time by up to 96%.

Gold tables were exposed through Databricks SQL Warehouse and used as the foundation for two focused Power BI mini-models, a Purchase model and a Stock Inventory model, covering procurement, inventory, warehouse operations, purchase orders, sales orders, and work orders across all three sites.

Two priority Power BI reports were reconnected to the new datasets with minimal disruption to business users.

Instead of one large model that everyone depended on and nobody wanted to touch, FQM now had two domain-aligned datasets that were easier to understand, easier to maintain, and built to scale.

From Hours to Minutes

Modernization significantly improved both platform performance and the day-to-day experience for FQM's commercial teams across all three sites.

FQM gained measurable improvements across every reporting dimension that mattered:

Modernizing also made the platform significantly easier to maintain and scale. Unnecessary tables were removed, problematic relationships were fixed, and transformation logic was moved to where it could be managed properly. New business-requested data points that would have required significant effort to add before can now be incorporated through a straightforward, scalable process.

Beyond the performance numbers, the organization gained something harder to measure but equally important. A reporting environment that once demanded constant manual attention is now one that business users can simply rely on.

A Blueprint for Modern Analytics

The FQM engagement established a repeatable modernization pattern that any organization running complex legacy BI environments can follow.

The architecture is straightforward by design:

• Ingest and structure data into a governed Silver layer
• Transform into business-ready Gold datasets in Databricks
• Serve through Databricks SQL Warehouse
• Consume through optimized, domain-aligned Power BI mini-models

Each layer has a clear responsibility, and each handoff is clean. And because the heavy work happens in Databricks rather than inside the reporting tools, the entire stack becomes easier to maintain and more resilient under load.

The immediate scope covered reporting across three mining sites and multiple operational domains. But what FQM built is not limited to those domains.

The same pattern applies to any reporting area, any site, and any new business requirement that comes along, without starting over each time. That is what separates a modernization project from a modernization strategy.