Building “One Auburn” on the Snowflake Data Cloud

Founded in 1856, Auburn University has developed into one of the largest universities in the South, remaining at the educational forefront with its traditional blend of arts and applied science. Changing with the needs of today while living with a respect for the traditions and spirit that are Auburn, the university offers over 150 majors and has a total enrollment of more than 31,500 students with 5,493 full time employees (as of 2023).

Previously, Auburn University relied on reporting solutions directly from its Banner ERP, an administrative software solution designed specifically for higher education systems.  As a result, the central data team at Auburn spent most of their time customizing reports for business units. Even minor tweaks, such as adding a column or changing its name, required their assistance. According to Misty Carroll, Associate Director of Information Technology at Auburn University, “We believed that a self-service model, with strong data governance in place, could better service business units to meet their reporting and analytics needs.” 

Auburn explored several solutions but were faced with resource contention, concurrency issues and lagging data access. To meet their aspirations, according to Charles Hunt, Director of Enterprise Systems at Auburn University, “We needed a modern data platform in the cloud that would move Auburn to a future where we can be quick and resilient.”

One Auburn, built on the Snowflake Data Cloud, centralizes stakeholder data. Its charter is to enable data-driven decision-making, delivering personalized experiences to university stakeholders. In the higher education space, Auburn needs to continue to leverage Banner. “Before, we were very limited in the custom API integrations to Banner. But now that the data is also available in Snowflake, it's very easy to connect new software within Azure and the process is scalable,” says Carroll. Starting with Banner data, Snowflake on Azure will become the single source of truth for Auburn’s data on finance, athletics, alumni, human resources, students, faculty and more.

Snowflake’s native abilities to separate compute and storage, near-zero maintenance and rigorous data governance were among the reasons why it was chosen to support this core pillar of Auburn’s strategic plan. “Like many higher education organizations, we store PII data, and that comes with many security concerns,” Hunt says. Snowflake met higher education compliance requirements and included governance features, such as role-based access and dynamic data masking.

When Auburn decided to build One Auburn on Snowflake, they brought on Snowflake Professional Services (PS) as well. “Because this is part of Auburn’s strategic plan and many key departments are coming on board, we wanted to make sure it was done right,” Hunt says. “When we moved from our on-premises solutions to Snowflake, it wasn’t a lift-and-shift; it was starting from scratch. We’re limited in our knowledge and resources, and PS was able to help us co-design key items, such as data cataloging.”

The Office of Institutional Research (IR) serves as the official source of factual, research-based information about Auburn University's students, faculty and staff; its programs, resources and outcomes; and its performance in relation to peer institutions and the external environment. As one of the most data-intensive departments, the IR and central data team collaborated on Auburn’s transition to Snowflake, helping establish data definitions, documentation and more.

The IR team often gets requests from stakeholders with broad questions. The faculty committee for student retention, for example, needed broad information to understand which factors most impact student retention. Previously, this project would have taken weeks or months between the initial data requests and the eventual consolidation of various data sources. Snowflake’s self-serve design plus overnight refresh rate allowed for the finished product weeks ahead of schedule. 

To answer these broad questions like student retention, the relevant data sets (such as student engagement, finance, course information, academic performance) were too large to deliver through traditional flat files, which also lacked proper security. Typically, the IR team would have had to ask stakeholders to narrow their request to be more targeted so they could share smaller data sets. 

Now, with Snowflake powering Qlik dashboards, the IR group could quickly fulfill the request in a week. Stakeholders are now granted access to dashboards connected to all the relevant data sources where they can freely tinker to answer their questions. The self-serve model empowered both the IR department and the users of IR dashboards, reducing the administrative burden on central IT.

As more departments participate in One Auburn and the university’s Snowflake maturity increases, Snowflake’s data-sharing capabilities may open more opportunities for greater collaboration. A few examples: further automation of data cleansing and reporting, increased transparency and collaboration loading Snowflake from Azure Data Factory, and increased user engagement. This evolution will continue to allow for extra time to focus on data analysis rather than data management.

“We’re moving toward a future where data-driven decision-making is infused at every level, and the university will get closer to realizing a student 360 view,” Carroll says.