Becoming "data-driven" is increasingly a top priority for higher education executives across the country and a major initiative for chief information officers and their teams.
This study outlines the best practices of how to:
- Develop a sustainable data governance initiative
- Proactively detect and fix data quality problems
- Get the right data to the right users
- Mass-customize reporting and analytics to drive user adoption
- Structure BI efforts for sustained impact
Through the use of case studies, content examples, and best practices identified by leaders in higher education, these resources will help accelerate your institution’s analytics strategy.
Download the study
Intuition-driven decision making no longer sufficient
Increases in higher education costs and greater competition for students, donations, research grants, and other revenue sources have necessitated data-informed decision making at many institutions. Pressure to be more analytical comes from many places, from state legislatures to boards of trustees, and often winds up at the CIO’s door because business intelligence is perceived by campus members to be a technical capability.
Data goverance is crucial to BI success, but often low in maturity
Data governance, the process of creating standards for data elements (e.g., data definitions, potential values, security levels), promotes consistency that enables reliable data comparison across an organization—a fundamental input to BI. Many organizations, however, suffer from low accountability for data governance responsibilities and poor campus engagement in related discussions. Institutions with successful data governance efforts have created sustainable models for ongoing data governance efforts, achieved consistently defined performance metrics across the institution, and created more mature BI efforts.
Data quality everyone's problem, no one's job
Higher education institutions collect large amounts of data, such as student grades, swipe card records, faculty service activity, and procurement purchases, but much of the data is too low quality to be useful in data analysis. Poor quality data may impede analyses or guide campus members to misinformed conclusions. Few campus members are incented to fix data problems (e.g., missing data, incorrect data, misplaced data) in source systems, so data cleaning occurs only (if at all) in frozen sets outside source systems. Institutions that have increased data quality in source systems have achieved gains in BI staff productivity and campus member efficiency.
Role-based access holds the solution for secure, just-in-time data privileges
Users must have access to data for analysis, but typical case-by-case access provision practices fail to balance access, efficiency, and privacy. Institutions, trying to balance security with openness, typically create processes in which individuals request access on a case-by-case basis. However, desire for data (and more objective security) has outpaced the adequacy of one-off access requests. Institutions have turned to models of role-based security to provide quicker and more principled data access to categories of campus members.
Campus members need help identifying what data is relevant and useful
The jump from intuition-based decision making to data-informed decision making has outstripped many campus members’ ability to use data to inform decisions. These campus members struggle to identify data sets and reports that may help them make better decisions; their frustrations often lead to low adoption of analytical resources and central data sets. Institutions that have been successful in increasing BI adoption have focused on making reports more relevant to end users through methods such as data spotlights, recommended report functions, and report enhancement mechanisms.
BI efforts require enterprise support and dedicated leadership to manage change
Many existing analytical initiatives across campus occur without coordination, causing duplicative work, redundant spend, and untapped institutional expertise. Coordination of analytical efforts may require a strong central effort to achieve institutional standardization and a “single source of truth” data set. Progressive institutions have also merged institutional research and BI units into one analytics team, and some institutions have begun to staff up data management and analytical efforts by hiring directors of data governance or chief data officers.
Download Our Related Toolkit
Access additional information on each of the 20 hallmarks of a data-driven university and explore resources and guidance to implement the strategies at your institution.
Access the Toolkit
Analytics as an Enterprise Process