What inspired you to go with Starburst?

We turned to Starburst because it allowed us to corral all our different data sources – like Redshift, MySQL, Heap, and S3 – into a single source of data access. Our data consumers get all of their information through Starburst without needing to know the details of how or where the data is stored. It just looks like a single virtual warehouse, so we’ve taken to calling it our logical data warehouse (LDW), which is the common term in our company. Starburst has also eliminated our need to make copies of all our data which in turn saves us time!

Where does Starburst fit in your overall data strategy?

Starburst is central to our data and analytics strategy. For all of our data consumers or any of our analytical use cases that are using BI tools, doing queries, or using data in any way, they access the data through Starburst. This way everyone who is accessing the data is getting it in a direct fashion, and we, as administrators, can have the luxury of moving data around as needed.

Which critical technologies are you integrating with Starburst?

We are currently integrating BI tools like Tableau and Geckoboard. Also, our whole corporate infrastructure, everything that services our customers and providers is on AWS, and AWS and Starburst work well together.

What are your three favorite Starburst features?

My three favorite features are:

• Materialized views: We can see all our data queries for all our data in one place.
• Query federation of disparate data sources: Nothing like having a single source of secure access to all our data!
• Vibrancy of the community: The enhancements and augmentation that have been done by Starburst developers or others in the Trino community have been amazing.

Which business outcomes are you most proud of?

Our number one business outcome was that we could eliminate outdated systems of analytics that involved copying data over every night because of Starburst’s single source of data access. No one even mentions this old process anymore because no one misses it! We can seamlessly move data from one storage technology to another without data consumers’ work being disrupted. This allows for a lot of agility in order to move out of legacy systems and move the business along as needed.

What success have you been able to measure?

Starburst has greatly reduced time-to-insights which has really amped up our ability to address problems in our functionality. We are able to do things now that simply weren’t possible before. Things we had to do manually are now automated, which saves us a lot of time, effort, and manpower.

For example, doxo had recently begun to notice unexpected patterns in customer churn rates and credit card failures. With the ability to see all our data together and query it in one place, we quickly learned that the problem was due to changes happening in the overall banking industry. Starburst allows us to see more clearly into the inner workings of the organization and make more accurate business decisions.

Do you think decentralized data access and Data Mesh are the future of analytics at scale?

My gospel when it comes to data is the fewer copies, the better! Data movement and data copying is simply a solution in search of a problem. Although this has been the case for years (and I have often received some resistance to changing these old ways), it is an inferior solution. By leaving data in its original source, you maintain a lot of flexibility in querying and analytics.