To improve analytics and decision-making, the Client segmented its various operational data sources into five key business verticals. As part of a convergence initiative, data from these sources had to be accommodated into a unified data model that catered to these verticals’ needs. This would involve decommissioning existing error-prone Excel and SQL Server Reporting Services (SSRS) reporting.
Converging on Better Analytics
These systems would be replaced by a consolidated data warehouse that uses Oracle Business Intelligence Enterprise Edition (OBIEE) and the Tableau business intelligence platform as reporting solutions.
To thoroughly test such a large and important environment, the Client needed an experienced solutions partner with strong process understanding, thorough technological collaboration abilities, and deep expertise in the reporting and analytics field. Prolifics Testing was the ideal fit for these requirements, leading to our engagement to take on the testing of its reporting environment.
Tuning Up Reporting
As an integrated part of the project team, Prolifics Testing’s Independent Verification & Validation (IV&V) experts performed testing and validations on data migration and reporting models. The large size of the data sets being tested made a simplified testing approach crucial to a successful delivery. Our Effecta Accelerator eased data comparisons and validations.
Extract Transform & Load (ETL) Testing was introduced into the Client's environment for the first time. Our stagewise testing successfully validated each phase of the data flow, eliminating data loss.
Reports were delivered partially in code drops as part of the Client's continuous delivery and continuous integration model. Validating reports after every code drop was difficult. In response, the IV&V team created test data that was specific to each functional code drop and validated it accordingly.
In the end, the IV&V team was able to identify defects in the early stages, which streamlined the process of implementing fixes. Its method of validating data in the staging and data warehouse tables against the data sources eliminated most of the data-related issues before they could be discovered in the report validation step, thereby making it easier and cheaper to rectify them. This led to a 99% defect removal efficiency (DRE) rate.
The result is a meticulously tested system that the Client can use to perform robust, reliable analysis. It can identify and capitalize on business opportunities with more agility and confidence.