Using BI to Audit Data Quality
Data Talks, Episode 15
May 14, 2019
Data Talks is Exago's podcast on all-things business intelligence, analytics, and application software. When most people think of BI, they think about extracting insights from their data, not about their data. This month, George Firican explains why BI can be a surprisingly effective tool in data quality auditing — particularly for organizations looking to save money by investing in multipurpose software. Firican’s expertise is informed by his work as the University of British Columbia’s Data Governance and Business Intelligence Director as well as his passion project Lights on Data.
Segment 1: Building Visualizations for Clients
(1:10) Distinguishing between “data governance” and “data quality.”
(4:20) Who in an organization should own data governance initiatives.
(6:50) The importance of data quality.
(12:14) Data quality factors (timeliness, integrity, accuracy, etc.) and defining data quality
(15:30) How to develop a data governance program.
(20:20) The data governance trifecta: identify, fix/prevent, communicate.
(23:55) Using BI to transform data in the display layer.
(26:21) The virtues of using BI to audit data quality, as opposed to a dedicated tool.
(27:40) BI report confidence intervals.
(30:40) Data quality scorecards and data profiling scorecards.
(33:50) Using BI to navigate the politics of data quality.
(37:40) Staging a data quality intervention.
Segment 2: What We Are Nerding Out About
(54:55) Nicole: Data sonification!
(57.15) Alex: Meetups!
|George Firican is Data Governance and Business Intelligence Director at the University of British Columbia and author of Lights on Data, a data management blog for non-profit, governmental, healthcare, and higher-ed organizations. With a background in web development and project management, he made the transition to data governance seven years ago and never looked back. You can find him on Twitter and LinkedIn.|