Case Study

Digital Federal Credit Union: Building a Modern, Serverless Infrastructure

Backstory

As a not-for-profit financial cooperative owned by and operated for its members, Digital Federal Credit Union, better known as DCU, is one of the top 20 credit unions by assets in the United States and the largest in New England. The credit union serves more than 950,000 members and their families in all 50 states.

The Challenge

Born from its parent company, Digital Equipment Corporation, DCU prides itself on being a high-tech financial services company offering products and services with a mission in mind: to help members achieve their financial goals collaboratively. Members access their DCU accounts via mobile apps, Online Banking, ATMs, telephone, DCU’s 23 branch locations in Massachusetts and New Hampshire as well as co-op shared branches across the United States.

Member information, as a result, is distributed across different channels, which prevented business stakeholders from making data-driven recommendations to the business in a timely manner. Business teams sought for insights on their members, membership pipeline, fraud collections, operation efficiency, and more. Without real-time access to data for analytical purposes, managers relied on time-consuming and manual copying, pasting and merging data from multiple systems into Excel spreadsheets. The data was stale by the time the data was ready to be used. The merging of data also often proved that the same information from two different systems were not consistent.

What’s more, the Data Analyst team spent most of their time supporting over 150 tickets per month on ad-hoc and problem resolution requests submitted by business stakeholders. Although the IS team has grown over the years, their productivity is slowed largely due to having to constantly maintain and work around on-premises systems and juggle hundreds projects.

“Our Information Systems department grew from 40 to 130 people in 5 years to support the high-growth phase the company is undergoing,” said Eric Giorgio, DCU’s Manager of Data Services, “By transforming our IT infrastructure, we can help our company exponentially grow at an even faster rate than what we are seeing.”

The Solution

To eliminate the costly and ongoing maintenance of their existing on-premises systems, DCU is moving to a completely serverless infrastructure to modernize the analytic platform. Their serverless infrastructure strategy followed a new design principle of moving all their machines and compute to the cloud. With the serverless infrastructure strategy was in place, data engineering was free to self-pace their data migration and integration roadmap. “By following our serverless design principle, we will no longer have to maintain machines as we build out a new state of the art enterprise data warehouse,” said Giorgio. “Instead, we can focus on creating a highly secure, performant and decentralized environment for business stakeholders to access the data whenever and wherever they need it.”

Once the serverless infrastructure strategy was in place, DCU sought for an integration platform to help them move their on-premises systems and data to the cloud. The evaluation of requirements in selecting an integration platform included support for their serverless strategy on AWS Fargate, application integration, ETL, and robust security as the credit union manages their members’ private information. The team chose SnapLogic as their integration platform of choice as other integration solutions did not support their serverless design principle or they required labor intensive building of data pipelines which is a costly resource to maintain. “We chose SnapLogic to support our serverless infrastructure strategy and to partner with us beyond application and data integration. The partnership has enabled DCU to further support their long-term growth strategy,” said Giorgio.

Business Results

Within 6 months, DCU has migrated 30 months worth of data, or over 5.8 billion of records, into Amazon Redshift. The data is refreshed each night so that the Executive team can see an up-to-date dashboard built in Tableau. Now that the data model for the executive dashboard is accepted, the team will roll out self- service dashboards on member, member onboarding, transaction, fraud, and other types of data so that more business stakeholders can access the data.

“Since migrating data with the SnapLogic platform, unified data is now put in the hands of business stakeholders to help propel our business in this exciting high-growth phase we are seeing. And the key value SnapLogic brings to DCU is that we don’t need to spend time writing code to build integrations,” said Giorgio.