From 80 reports that disagreed to one set of numbers

DashPoint Analytics is moving an online education provider from about 80 separate reports, each with its own math, to one shared set of numbers, starting with a scorecard of the 14 measures leadership watches most.

14 KPIs leadership metrics on one trusted tracker

The problem

An online education provider had about 80 reports, each pulling and calculating its own numbers from the same handful of systems. The same figure came out differently in different reports, and nobody could tell which one was right.

What we built

One central data platform that collects data from every system, cleans it, and calculates each number one way. Changes are tested in a practice copy before they go live. The nightly update checks itself, and if one system sends bad data it stops and alerts the team instead of quietly showing stale numbers.

Medallion architecture diagram: sources feed ingestion, then bronze raw data, silver cleaned data and gold curated data, then Power BI and analytics, with orchestration and monitoring underneath.
A typical Fabric medallion pipeline: raw data lands in bronze, is cleaned in silver, and becomes business-ready gold tables that reports read from. Ours follows the same layers, with Spark notebooks in place of the dbt steps shown here.Image: Microsoft, Microsoft Fabric documentation, CC BY 4.0, resized.

Outcome

Each number now has one definition. Leadership tracks 14 key measures from one trusted scorecard, and there is a clear, repeatable path for moving the remaining reports.

  • Fabric migration
  • Medallion architecture
  • Data Factory
  • Git + CI/CD
  • Variable libraries

More case studies

Healthcare revenue cycleChecking every insurance payment against the contractRead the case →Professional services (engineering)One source of truth for a national engineering firmRead the case →Healthcare paymentsPredicting what an insurer should pay on a claimRead the case →

Have a similar problem?

Talk to us
↑