Prerequisite: None
Traditional data marts are often mislabeled as self-service because they facilitate ad hoc queries. What happens when the business needs to add a new data source, create new attributes, or change a classification scheme?
Traditionally, this would require a multi-sprint work effort by an ETL team to change the data model, update the data pipeline, update the BI metadata layer, and refresh the ad hoc query environment. Due to these time delays and lack of ETL resources, users will try to do it themselves in Excel. So, the true meaning and intention of self-service is lost.
In this presentation, Devan Muthusamy will discuss how Mass Mutual came up with an innovative solution to address this challenge. Mass Mutual developed an integrated approach that allows end users to enrich data, define new attributes, and even add new data sources directly to existing data marts that reflect in the BI environment instantly. The result is true end-to-end self-service and rapid data deployment. New ETL pipeline development is not required, and the entire process is visible and governed. For example, an investment manager is able to reclassify a data asset instantly using an update business rule—a process that would have taken several weeks in the past. Mass Mutual is now able to achieve SLAs as low as 1–4 hours.