University of Texas System · Higher Education · Austin, TX · 18 weeks
Financial aid disbursement process automated with AI
The University of Texas System manually processed financial aid disbursements across eight institutions, taking about 14 days each time. We developed an AI-driven automation system that cut the processing time to 36 hours and saved $1.7 million in annual labor costs.
Challenge
The financial aid office handled 180,000 disbursements yearly across eight institutions, each with different verification steps and workflows. The manual process took 14 days on average to complete, delaying funds for students. This delay created uncertainty that directly caused 600 fewer enrollments than expected.
Approach
We developed a system that sorts disbursement applications by complexity. Simple cases move straight to automated processing, while complex ones get flagged for manual review. We connected the system to the federal StudentAid.gov API to verify eligibility instantly. We also created a portal for students to track their disbursement status in real time.
Outcome
We cut disbursement processing from 14 days to 36 hours and slashed annual labor costs from $1.7 million to $280,000. By addressing financial aid uncertainty, we closed the enrollment yield gap by 4.1 percentage points, adding 740 students and $5.9 million in tuition revenue.
We automated 3,600 of the 5,000 weekly applications. That left only 1,400 for manual review.
So, we took two years of disbursement data and threw it into an AI model. What were we after? Spotting the simple applications that didn’t need any extra attention. Out of 5,000 weekly submissions, the model handled 3,600 by itself. That meant the team only had to dig into 1,400 cases manually. It was a simple win, cutting down the workload and speeding up the whole process.
How We Watched Sign-Ups Unfold in Real Time to Pump Up Enrollment
We put together a portal that showed students real-time updates on their financial aid. The impact? Enrollment yield went up by 4.1 percentage points. When students could actually see where they stood, it wiped out a lot of their guesswork. That clear picture made them way more likely to sign up. Pretty simple.
We plugged into the federal API to check eligibility on the spot.
We plugged straight into the StudentAid.gov API to check federal eligibility instantly. No more waiting around 3 to 5 days for manual verification. That tweak cut our processing time from two weeks to under 10 days.
Results
- 14 days to 36 hrs Disbursement processing time
- $1.7M Annual processing labor saved
- 740 Additional enrolled students
- $5.9M Additional tuition revenue
Students were signing up without knowing if they could afford it, and waiting two weeks meant we lost a lot of them. Thinklytics helped us bring that down to about 36 hours, and the next year we enrolled 740 more students.