Gaining immediate data access and reducing infrastructure costs by hundreds of thousands per month

Financial technology company Banco Inter (Inter) became Brazil’s first fully digital bank in 2014. Since then, Inter has been on a mission to simplify customers’ lives. The bank has become a digital platform for investments, insurance, loans, e-commerce, and even food delivery. Inter offers customers digital activities through the company’s “Super App” where users can buy products, book hotels and plane tickets, order food, enter credits for Uber, and more. The company provides over 180 products and services and has a rapidly growing customer base, serving over 24 million customers today. Inter needed a cost-effective way to aggregate vast amounts of customer data and third-party data, to monitor areas of improvement and identify new growth opportunities.

Inter had a big data strategy in which the company would develop a data platform aiming to make good data accessible to the entire organization. Looking to reinforce its data-driven culture, it focused on reducing “time-to-market,” removing technical barricades to the access and use of data, and granting agility to all business areas to make decisions using data. However, Inter faced challenges with the limitations of its previous query solution, Amazon Athena:

• Only allowed a maximum of 20 queries at a time
• Had a 30-minute run-time limit
• Lacked data governance
• Cost $40-$1000 per query, making it too expensive to scale
• Cost over $100,000 in data warehousing (with Amazon Redshift) 

Inter struggled to generate the quick insights that were needed to shape the future of the digital marketplace. 

The company’s data analytics processes left users completely dependent on data teams to serve them their results. With each reporting request, the data team had to spend hours — and sometimes days — to move the data to another zone in the data lake so it could be viewed through their analytics and visualization tools. Inter needed a solution that would allow anyone in the company to be able to generate ad-hoc queries from the data lake for near-real-time insights.

After evaluating three solutions, Inter chose Starburst Enterprise.

Inter wanted a scalable and performant query solution  — and found it in Starburst. Starburst had the strongest potential to become Inter’s main platform for the exploration of data through data democratization while reducing time-to-insight. 

Starburst Enterprise, deployed on Amazon EC2, provides Inter with a single platform to explore more data, maintain data quality and governance, and provide self-service analytics using the visualization tools of choice. Starburst’s federated semantic layer simplifies data integration and access while optimizing query performance across distributed data assets.

Now, users can easily access the platform, and data scientists can connect to Starburst and explore data through their notebook computers. Starburst’s plug-and-play connectors speed up business intelligence (BI) applications and data visualization, enabling the business user to use Trino as a data serving layer without the need for engineering teams to act. Now for the first time in history, users can go to Starburst and do a consistent query to consume the data from the number of logins happening in the last 30 seconds.

Since deployment, Starburst supports Inter’s growth in several ways:

Revenue generation. Data consumers have a single point of access to all of the company and customer data, resulting in greater insights, enhanced marketplace offerings, upsell and cross-sell opportunities, and increased profitability. 

Increased analytics access. Starburst allows thousands of data consumers to access huge amounts of data sets at high query volumes with low latency on their preferred analytic and reporting tools. The amount of data scanned in the general cluster is almost 34 petabytes. The number of queries grew exponentially—over 10,000%—since a single dashboard with just one chart can generate more than 20,000 queries in Starburst per day.

Faster time-to-insight. With 50+ out-of-the-box connectors, Inter can provide immediate access for data exploration, reducing time-to-insight from days to seconds.

Cost savings. Inter can reduce infrastructure costs dramatically by moving to Amazon S3 and Starburst, saving hundreds of thousands of dollars over the next few years.

Through Starburst, Inter has access to a complete set of data from sources such as the Super App, NASDAQ, Brazil’s stock market (B3), and the central bank. The digital bank can create personalized recommendations to provide the right product to the right customer, at the right time. Inter has several other initiatives for anti-fraud and payment reconciliation that are only possible today because of Starburst. Starburst is now at the core of all of Inter’s processes. They’re not just able to query and view the data, but can also build intelligence from it.

Now, data consumers access and query data without assistance from the data team, saving valuable resources. Starburst allows Inter to take advantage of monetization opportunities, pursue revenue-generating projects, and generate sales from personalized recommendations to products and services.