You have /5 articles left.
Sign up for a free account or log in.

Retain pulls student data from various software platforms used across campus to provide predictive analytics and a dashboard of intel for advisers and faculty members.
SmileStudioAP/iStock/Getty Images Plus
Student success professionals at National Louis University in Chicago will have a new tool to help them guide student advising and persistence work, enabling real-time data-based decision-making assisted by artificial intelligence.
Retain, an in-house software that pulls from the university’s data lake, identifies risk factors in student persistence and gives practitioners actionable steps to address and improve learner outcomes. The tool is a reflection of the university’s larger data strategy and the latest example of how AI can impact the student experience in higher education.
Survey Says
A 2024 survey of campus chief technology officers by Inside Higher Ed and Hanover Research found fewer than 10 percent of respondents believe higher education is prepared to handle the rise of AI, but closer to 30 percent said they believe their institution is at least somewhat prepared.
Just over one-quarter of CTOs say they’re using artificial intelligence for predictive analytics of student performance and trends, and one-quarter say they’re using it for research and data analysis.
How it works: Retain is an AI-enabled system that pulls relevant data from campus systems to create a “single pane of glass” of student information, says CIO John Mazariegos.
The dashboard, which is presently designed to support advisers or success coaches, allows staff to view all their advisees’ data, including their course grades, flags, degree completion and attendance. Based on the data presented, the AI generates a persistence likelihood on a scale of one to 10 (one being low, 10 being high) and categorizes student performance in terms of red, yellow or green flags.
For example, a student whose GPA has fallen below 2.0 will receive a red flag, and a student whose grades have declined slightly week over week may receive a yellow flag.
The system does not use an algorithm to determine student likelihood but instead weighs the different factors of student success such as GPA and attendance, Mazariegos clarifies.
Retain pulls data daily from the different systems and processes the data twice a week with AI, meaning the insights available to advisers are new and updated each morning.
Within the system, advisers can also send emails or text messages to learners as an aggregate or individually. At present, this is a one-way notification system, but campus leaders are looking for a vendor to implement two-way texting so students can respond, Mazariegos says.
The background: NLU has used EAB’s software for the past four and a half years, and as the end of the contract term drew closer, administrators began to consider other solutions for data aggregation and presentation. Tangentially, the rise of generative AI tools pushed leaders to consider ways to integrate artificial intelligence into campus operations to streamline student supports. The two factors led to the creation of Retain.
Retain was built entirely in-house using NLU’s programming team and has taken about one year. The tool will officially replace existing software in summer 2025.
NLU’s president, Nivine Megahed, has led the university’s data strategy since 2011. As of 2014, NLU has managed a shared data repository, which houses data from the various systems across campus including Desire to Learn (D2L, the institution’s learning management system), Ellucian Banner Student information system and Degree Works.
“That’s one of the things that’s made this easy for us … The aggregation of data really made this possible,” Mazariegos says. Instead of having to build integration between software platforms, staff can reach directly into the data depository, making the process faster and more efficient.
Building better: Campus leaders relied on advisers’ and other practitioners’ insights to develop the tool.
One piece of feedback advisers shared from using the existing platform was that autogenerated messaging was harder to get rid of than to write an original message, so Retain does not suggest messaging or create prescriptive next steps for the student.
The software will also create an aggregate of notes that staff or faculty have left on a student’s account, rather than housed in each individual software application, so anyone who needs to interact with the flag can do so.
Ensuring data privacy was key to the software’s development. “That was one of the biggest worries that we had initially,” says lead programmer Franko Hysaj.
Students’ personal information (including name or email) is not disclosed to the large language model, and all information is anonymized to protect their information. Staff also reviewed privacy policies of the LLM to make sure that they aligned with the university’s, including not retaining data, just processing and serving it.
What’s next: Starting in 2025, NLU will begin to train faculty members on how to use Retain, collecting feedback based on their experiences, and then will solicit insight from students.
“The biggest hurdle we’ve had to overcome is [users sharing], ‘This is how I’m used to doing it’ and ‘This is where I go to get the data,’” Mazariegos says, so leaders are helping staff reimagine how they’re using technology to inform interactions with students.
The university is also considering ways to provide insightful data to career service professionals, who are less concerned with a student’s course progress and have greater interest in their checklist toward an internship or graduation, Mazariegos says.
Mazariegos says there have been conversations about licensing the software to other institutions, but for now the secret sauce is just “having the data and how we’re presenting it.”
Get more content like this directly to your inbox every weekday morning. Subscribe here.