One of the oldest and most respected universities in the United States, Princeton University boasts a generous financial aid program. The Office of Development plays a significant part in that effort, initiating and nurturing donor relationships. Its work also supports other crucial aspects of Princeton’s mission.
“We raise money for scholarships, for fellowships and for professorships,” says Heather Campbell, Associate Director, Development Research Analytics, Princeton University. “Our capital projects and important annual unrestricted funding comes from the support of our alumni, family and friends, organizations and foundations.”
Support Staff in Need of Support
While the Office of Development maintained in-depth transactional giving data and relationship management data, staffers often had to dig through several different reports to answer questions—a time-consuming, manual effort with potential for mistakes.
“They were downloading five different reports and then putting them into Excel and merging them together, manipulating them a bunch of different ways to get the information,” explains Campbell. “There’s absolutely going to be human error whenever you start doing a lot of data manipulation in Excel.”
Viewing data in Excel also made it difficult to notice results that didn't quite make sense—data entry errors or outliers that could indicate significant risks or opportunities. The volume of work posed another, more subtle issue for staffers.
“It can be easy to get into a mindset where you’re simply checking off finding different data elements, rather than trying to understand the overall business question,” says Campbell.
Manual Processes and Hunches
Members of the major gifts team each manage a pool of approximately 200 prospects—far too many people to rely on memory to manage contacts. With no dedicated tools to visualize overall engagement, the office has relied on large list reports and manual review of individual records to make sure prospect and donor relationships were adequately tended.
The research analytics team was responsible for reporting high-level campaign metrics to university trustees and other top executives, but this was another time-consuming, manual effort.
“We would run list reports or custom SQL extracts every other week and then manipulate the data in Excel; then we would put the graphical results into PowerPoint,” Campbell says. The process took at least two days to complete. Not only did this take two full days of employee time, it also meant that executives were making decisions based on stale data.
Finally, the international team had a difficult time finding reliable information about how Princeton alumni across the globe were interacting with University groups in general or the Office of Development, specifically.
“They used their hunches a lot,” says Campbell. “And they were often very good. But we wanted to use data to validate those hunches.”
Taking a Fresh Look at an Existing Tool
The Office of Development didn’t realize it, but it already had the solution installed: Tableau.
“We started using Tableau on our desktops in 2008,” says Campbell. “We just used it for maps.”
In 2012 the Development Office made the decision to purchase several professional licenses providing direct access to the Oracle database.
“I think one of the reasons we didn’t proceed sooner was a lack of understanding of the full benefits of the tool and a wariness of how much database support would be needed. In the end the benefits are far outweighing the cost," says Campbell. "Although we are still playing catch-up with our database architecture and optimization to best support this direct query access, it’s a paradigm shift in the right direction."
Today, the Office of Development has three employees authoring data visualizations in Tableau Desktop and approximately 150 staffers and administrators consuming vizzes through Tableau Server.
In addition to visualizing data from their Oracle database, Campbell’s team also blends data stored in Excel. They have begun experimenting with blending data from the university’s Exact Target digital marketing automation platform. Typically, the team visualizes approximately a million rows, although some dashboards have included up to 8 million rows of data. For security they use Active Directory.