Fixing Application Drop-offs in University Admissions

Project Snapshot

Owning the end-to-end admissions experience

Ken42 is a unified SaaS platform used by universities to manage admissions, academics, placements, and alumni. Within that platform, I took full ownership of the admissions module — the most complex surface in the product, spanning student-facing application flows and administrator-facing verification dashboards.

The core problem: a workflow that looked like a form was actually a multi-step system involving lead capture, document verification, interview scheduling, and offer management. It was breaking down at every stage — for students and administrators alike.

• Owned the end-to-end application flow — forms, journeys, and dashboards
• Identified and mapped drop-off points across the full admissions funnel
• Redesigned the flow to shift from data collection to guided completion
• Introduced data persistence and document reuse to reduce repeated input
• Designed for multi-university scale — configurable without becoming complex
• Explored an AI-assisted application experience to further reduce friction
• Delivered the full concept in a focused 2-day design sprint

Business Impact

Completion as a conversion lever

Drop-offs in the application flow are not a UX problem — they are a revenue problem. Every student who abandons mid-application is a potential enrolment lost. The redesign was positioned as a conversion initiative, not a usability cleanup.

* Metrics are predicted targets based on identified friction points, not measured outcomes.

Context

A workflow masquerading as a form

University admissions is not a single form — it is a multi-step workflow involving different rules, roles, and decision points at every stage. Lead capture flows into application submission, which triggers document verification, which gates interview scheduling, which precedes offer and acceptance.

Ken42 serves multiple universities from a single platform. Each institution has different eligibility criteria, application steps, and decision workflows. The design challenge was building a system that adapts to institutional variation without fracturing into unmaintainable custom builds.

The system was collecting data, but failing to complete applications efficiently. The redesign shifted the goal from data collection to guided workflow completion.

The Problem

Where the system was breaking down

Product & Design Ownership

Designing the system, not just the screens

I operated as the sole designer — responsible for problem framing, flow architecture, interaction design, and concept documentation. The work required thinking across two user types simultaneously: students navigating an unfamiliar process, and administrators managing high volumes with limited tooling.

• Mapped the full admissions workflow before designing any single screen
• Prioritised completion over comprehensiveness — fewer steps, guided progression
• Designed reusable structures that could adapt to multiple institutional configurations
• Balanced student experience with administrator operational needs in every decision
• Explored AI-assisted concepts to address the deepest friction points in the flow

What Was Designed

Shipped for completion and scale

Trade-offs Considered

Every decision involved a constraint

This project operated under real tensions — between institutional flexibility and platform scalability, between automation and user control, between shipping fast and solving deeply. Each trade-off was made explicitly.

Predicted Success Metrics

How success would be measured

Target metrics based on the friction points identified in the existing flow. They represent the hypotheses the redesign was built to test — not measured outcomes.

• Reduction in application drop-off rate — primary north star metric
• Decrease in average time to complete a full application
• Reduction in support requests raised during the application flow
• Increase in multi-programme application rate (data reuse adoption)
• Reduction in administrator time spent on manual verification tasks
• Improvement in applicant-to-offer conversion rate

Reflections

What I'd do differently

• Introduce data persistence earlier in the design process — it was the highest-impact change and should have been the first constraint, not an addition
• Design the configuration layer for institutional variation upfront — retrofitting flexibility into a standardised flow creates edge cases
• Add explicit feedback loops at every stage — students need to know not just where they are, but that their progress is saved
• Validate drop-off hypotheses with real behavioural data before finalising the redesign — the funnel projections were based on qualitative signals, not analytics
• The biggest shift was moving from designing forms to designing systems that help users complete workflows

Key Learnings

What this project demonstrates