Use case

Stop discovering at-risk students at the end of the term.

Edullent turns the test marks, attendance pattern and concept-level signal into a per-student trajectory — six weeks before the report card.

A student doesn't fail in the exam — they fail in the seventh week of the term. Edullent's student performance intelligence catches the drift at week three.

The problem

What schools live with today

Performance is reviewed twice a year — the mid-term and the final. By the time a teacher sees a slip on paper, the student has been falling behind for thirty class hours. There is no concept-level diagnosis, no risk score, no intervention list. The PTM exists to deliver bad news after the term is over.

The flow

From signal to outcome.

Five steps. The same shape every Edullent use case follows.

Step 1
Capture

Marks, concept strengths, attendance, assignment scores — every signal a student generates in the LMS.

Step 2
Score

Edullent's engine computes a per-student trajectory updated daily.

Step 3
Predict

The Risk Predictor flags Critical / At-risk / Stable / Excelling weeks before a test would.

Step 4
Recommend

Each flagged student arrives with a suggested intervention list — concept, peer, parent or 1:1 plan.

Step 5
Resolve

Teachers act. Parents are looped in via Parent Connect. The next week's data reflects the change.

What you get

Capabilities behind this use case.

Real product surfaces, shipping in the platform today.

AI Risk Predictor

Four-tier risk classification updated as new marks land.

Concept strengths

Subject-level drilldown to the topic that's dragging the student.

Performance trends

Per-student trajectory across terms, with peer-class context.

Intervention list

AI surfaces what to do, not just who is at risk.

Parent loop

Daily intelligence to parents — they see the drift, not the post-mortem.

Branch comparison

For groups: which campus has rising at-risk populations.

Who it changes

Role-by-role impact.

Owner

Network-wide risk view — which branch is producing the most slipping students this term.

Principal

A live list of who needs intervention this week, ranked by urgency.

Teacher

Per-class risk heatmap and the suggested intervention for each flagged student.

Parent

A weekly read of how their child is actually doing — not how they did three months ago.

Outcomes

What the pilots produced.

~6 weeks earlier

first risk-student intervention vs. previous year

4-tier

risk classification — Critical / At-risk / Stable / Excelling

Daily

trajectory refresh — not term-end snapshot

Outcomes measured at Pilot Schools A, B and C in their first term. Anonymized by partner request. See customer stories.

FAQs

Questions schools ask about this use case.

Yes. Edullent carries forward the previous year's signal where available, and starts producing usable predictions after the first two graded data points of the new term. Schools usually see meaningful risk classifications by week three.

It's a composite — recent marks trend, attendance pattern, assignment completion velocity, and concept-strength regression. The composite is calibrated per-school, so it understands what "normal" looks like in your institution, not someone else's.

No — the AI never decides for the teacher. It surfaces the student and the suggested intervention. The teacher chooses. Edullent's design rule: AI ranks and recommends; humans decide and act.

No. Parents see a softer, parent-facing read of trajectory — what to watch, what to ask, what the school is already doing. The internal Critical / At-risk language stays inside the institution.

Continue exploring

See student performance on Edullent.

A 25-minute focused walkthrough on this one use case. Your school structure, your data shape, your real questions.