Power BI Data Management Platform for Long-Term Research Studies

Background The Client Challenge

Our client is a leading US research university with a long-standing history in interdisciplinary medical and social science research. Over three decades, the university has run a large-scale longitudinal study tracking the long-term health outcomes of more than 15,000 participants recording lifestyle habits, clinical measurements, demographic data, and environmental factors across hundreds of follow-up cycles.

Despite the scientific value of this dataset, it had never lived anywhere close to a single, usable system. Thirty years of records sat across hundreds of disconnected Excel and CSV files, each following its own internal codebook conventions:

• Millions of entries spread across hundreds of files with no shared structure
• The same parameter age, BMI, blood pressure, and others was coded differently from one file or study cycle to the next, making cross-cycle analysis unreliable without manual reconciliation
• No central database existed; the "source of truth" was effectively whichever file a researcher happened to open
• Compiling data for a single report or analysis request could take a researcher hours of manual searching
• Every data query, however small, needed IT involvement, since there was no self-service way in
• Each new study cycle added to the file pile increased the risk of inconsistency and data loss

The university brought in Versich to design and build a centralised data management and analytics platform on the Microsoft stack, working closely with research leads and statisticians throughout.