Your data already knows who's at risk. Lumen tells your faculty in time to act.
Every modern university collects more student data than it can read. Most of it serves administration. Almost none of it reaches the people making pedagogical decisions, in time to act.
A Portuguese university gave us 33 years of academic records, sociodemographic data, and questionnaire responses across 10,397 students. We didn't go in looking for a product. We went in to listen.
This is the first number we found.
Most universities can't answer this with one number. Tuition lost. Completion-tied funding gone. Ranking exposure. Alumni base eroded. Each lives in a different system, calculated by a different team, if calculated at all.
Lumen does not compute that number for you. It tells you who you're about to lose, while you can still act on it.
Four inputs. The math is plain. Adjust the left-hand side to your reality, the right-hand side updates immediately. Most institutions never compute this number. The ones that do tend to invest differently in retention.
Not raw tables. Not BI dashboards built for finance. A working interface designed around pedagogical decisions, validated by the people who'd actually use it. That's what we built next.
Lumen reads the data your institution already collects. It turns it into a working interface for faculty, course coordinators, and academic leadership.
Lumen is not four screens and done. Every new pattern surfaced in your data becomes a permanent view. Every new question your team brings, a new module.
"Lumen could be a framework, not just a platform. A replicable model for the application of AI in higher education systems."
Six things you need to know before a pilot. The architecture answers most of the questions IT and procurement teams ask before they sign.
Lumen was designed from day one to redeploy across institutions. The data structures it reads are common across European, North American, and Latin American universities. The ML models recalibrate on your historical records. The interface localises.
Map your SIS to Lumen's star schema. Secure ETL. Anonymisation at the data layer. Quality profiling.
Retrain the ensemble model on your historical academic records. Validate against known outcomes.
Localise course taxonomy, language, indicator names. Adapt the dashboards to your institutional context.
Two or three pilot programmes. Hands-on sessions with the professors and coordinators who'll use it.
Real students. Real cohort. Faculty interacting with the platform across the semester. Mid-pilot and end-of-pilot review.
Lumen routes different signals to the rector's office than to the faculty member. The data layer is shared. The dashboards are role-aware. The student at the centre is the same.
Your next dropout is already in your dataset. Lumen finds them while you can still act.
A 30-minute call to walk through your student information system, the retention questions your faculty actually ask, and what a one-semester pilot would look like at your university. We come prepared. You leave with an honest answer either way.
Book a 30-minute call