UK Pharmaceutical Company
A major UK-based pharmaceutical company aligned data in various source systems with their existing master data hub, cleansing their master data and improving data quality across their organization and branches around the globe.
Business objective & project requirements
The pharmaceutical company aimed to determine how closely data in their source systems (including clinical studies, chemical compound and substance information, customer data, and more) corresponded with the information in their existing master data management hub.
- Verify the alignment of data elements across 7 source data systems in various formats and one master data hub to determine how closely source system data matched that of their MDM solution.
- Data Profiling and DQ Risk Mitigation were mandatory for all IT projects.
- Prepare a framework that would make it easy to integrate data quality monitoring into every future data-related project in the company.
- Although not an original requirement of the project, in the process of implementing a data quality solution, Ataccama uncovered data quality issues not only in terms of differences between attributes in source systems and their MDM hub, but also within data stored in the company’s MDM hub, including duplicated information, invalid values, and more.
Solution & benefits
The comprehensive data quality solution enabled company data curators to locate and repair any data issues in their source systems and MDM hub. Business users are now able to view data quality improvements over time and raise awareness of the importance of data quality on all levels. The solution included features such as:
Summary reports with an overview of the current state of data
Detailed reports to allow data curators to perform manual corrections in individual source systems
Export to XLS functionality
Color-coded XLS templates that could be populated directly from Ataccama data quality tools, and accessed via Ataccama’s user-friendly GUI
Drill to detail possibilities, including entities (clinical study), parent attributes, and child attributes