Headlines and company announcements involving artificial intelligence and machine learning are making waves in industries as diverse as banking, healthcare and ecommerce. But both AI and machine learning are built on top of impeccable data quality, which is easier said than done. In fact, Experian found that 75% of businesses are wasting 14% of revenue due to poor data quality, and IBM estimate that bad data costs companies $3.1 trillion annually in the US alone.
Why does this matter to software testers? Because if your organization isn’t already recruiting testers for data warehouse and big data testing, it certainly will soon.
Our webinar, Data Warehouse Testing —The Next Opportunity for QA Leaders, brought together Wayne Yaddow, a data warehouse and business intelligence consultant and Raj Kanuparthi, a senior quality leader who drove the big data testing initiative at one the world’s leading financial service companies, to share why and how they are seeing project managers putting data testing initiatives front and center as part of their quality processes.
The costs of manual testing often outweigh the benefits, in the context of data testing or otherwise. But when it is concerning data transformations, data cleansing or data warehousing, the end of result of manual testing is uncertain data quality, and that is a major problem for businesses. Obviously, automating data testing is crucial, but Wayne also stressed the importance of testing data at its source and during the ETL phases. Errors or discrepancies caught in these stages can be dealt with more efficiently than waiting for the compromised data to surface in BI reports.
Drawing from his experience driving a big data testing initiative at one of the world’s leading financial service companies, Raj laid out the process, culture and technology challenges faced by organizations trying to establish data testing automation frameworks of their own. Starting with the establishment of a data governance program and a shared understanding of the different aspects of data (validation, profiling, accuracy, integrity, consistency, completeness etc.), Raj shared his recommendations for organizations setting up or overhauling their data testing practices.
Dig further into this diagram, get Wayne’s personal book recommendations on designing and managing data warehouse testing projects and more by watching Data Warehouse Testing —The Next Opportunity for QA Leaders on-demand. You’ll also see how easy it is for testers with Tricentis Tosca experience to use Tosca BI for data quality testing and BI report testing.