Dinero was a real, 2-year product. Some data values, quotes and metrics are modified or representative to protect participants and institutional data.

DineroInternal Platformfor Masters’ Union

UX Research Case Study

From an 8-step manual admissions journey to a unified platform: serving 1,000+ students, eliminating 3–5 hours of daily admin overhead and converting an external payment link into 93% in-app completion.

User research
Product design
Double Diamond
EdTech

1,000+
Students onboarded: 87%
Usability task success: 93%
Payment completion (A/B): <1 hr
Admin daily overhead

The Dinero admin dashboard: applicants, fees received, pending dues and overdue accounts at a glance

Framework: Double Diamond; HEART · GSM
Timeline: 2021 – 2023
Team: 1 PM · 3 UX/UI Designers; 8 Engineers · 2 QA
Tools: Figma · Miro · Notion

Part 1

The Problem

Context

Two systems, no connection

Masters’ Union is a tech-first business school that scaled rapidly, but its internal operations couldn’t keep pace. Two disconnected systems ran the entire admissions and fee cycle.

NPF: captured applications and stored student records.

Pinelab: processed fee payments as an external gateway.

Problem

Neither system talked to the other. Every handoff was manual. Every status update was a spreadsheet entry. Students clicking a payment link in an email, for a Rs.3,50,000+ transaction, were landing on an unbranded external page they had never seen before.

Pre-Dinero

The real 8-step manual journey

The 8-step manual admissions journey as a descending staircase: step 1, NPF application → Export to Excel; step 2, Interview slot assigned → Confirmation email sent manually; step 3, Interview conducted + exam scored; step 4, Scholarship decision: Approved / Rejected; step 5, Student clicks payment link in email; step 6, Offer letter sent by business team via email; step 7, Student submits admission fee + tuition fee; step 8, Payment confirmed → Enrollment updated manually across NPF + Excel + CRM.

1
NPF application → Export to Excel
2
Interview slot assigned → Confirmation email sent manually
3
Interview conducted + exam scored
4
Scholarship decision: Approved / Rejected
5
Student clicks payment link in email
6
Offer letter sent by business team via email
7
Student submits admission fee + tuition fee
If the student cannot pay, manual loan approval begins
8
Payment confirmed → Enrollment updated manually across NPF + Excel + CRM

The admissions tracking spreadsheet, one of the manual tools behind the journey — Total overhead: 3–5 hours of non-value-added admin work daily.

Part 2

Research

Timetable

How it all happened

18 weeks, from the first stakeholder session to Figma Dev Mode handoff.

Oct 2022 – Feb 2023

Week 1–2Week 3–4Week 5–6Week 7–8Week 9–10Week 11–12Week 13–14Week 15–16Week 17–18

Stakeholder interviews · Research plan finalisation · Participant recruitment
User interviews: Students (5 sessions) · Admin (3 sessions)
User interviews: Faculty (3 sessions)
Competitive analysis
Card sorting
Affinity mapping · JTBD framework · Persona development · Journey mapping
Wireframing · Lo-fi to mid-fi prototype (35 screens)
Round 1 usability testing, 5 participants
Design iteration based on Round 1 findings
Round 2 usability testing · Heuristic evaluation
Hi-fi UI · Design system (30 components) · Figma Dev Mode handoff

5
Students
User interviews · Card sorting · Usability testing
3
Admin
Contextual observation · User interviews · Usability testing
3
Faculty
Stakeholder interviews · Workflow validation

Dhiraj mapping the application flow on a yellow glass board

Alignment

Three tensions that reframed the brief

Before talking to end users, we aligned with product, admissions and finance. The goal was not to gather requirements. It was to understand how each of them defined success.

1
Leadership believed the problem was “students not completing payments”. Research revealed it was a trust design problem, not a student behaviour problem.
2
Admins said their manual effort was “1–2 hours” a day. Contextual observation revealed the real number was 3–5 hours.
3
No stakeholder had ever mapped the full 8-step journey end-to-end. It had never been documented in one place.

Landscape

What already existed

We audited what was already in daily use, then scanned the market. No one tool owned the journey from application to enrollment.

Used for

Gap that drove Dinero

Used forApplication intake · student record storage

Gap that drove DineroNo workflow operation: admins exported to Excel for every action

Used forManual tracking of interview scores · scholarship decisions · payment status

Gap that drove DineroNo single source of truth · multi-hour daily reconciliation

Used forExternal fee payment gateway

Gap that drove DineroNo Masters’ Union branding · 14–15 day manual reconciliation · ~25% abandonment

Optimized for

Gap we observed

Optimized forLead capture and applicant CRM for Indian institutions

Gap we observedStrong on intake; limited on post-admission workflows

Optimized forGlobal tuition payments with multi-currency and compliance

Gap we observedStrong on payment infra; not built to own admissions lifecycle

Optimized forStudent information systems and reporting (US/UK focus)

Gap we observedStrong on records; required heavy integration for Indian context

Synthesis

Affinity mapping

11 interviews and 2 contextual observation sessions produced ~190 data points. Every observation went onto its own sticky note, then we grouped bottom-up, with no predefined categories.

Total: 188 notes clustered across 5 themes
Fragmentation 48 · Visibility 41 · Info overload 41 · Counselor invisibility 32 · Payment trust 26

Workarounds were the unmet need. Every role fell back on personal spreadsheets, WhatsApp or email drafts when the official system failed them.
Needs changed by time of day. Admins needed a cohort view in the morning and a single-student view in the afternoon.
Trust was a design problem. Students and admins both described moments of “is this real?” Branding, confirmation and receipts mattered as much as function.

Part 3

Design

Who we designed for

3 research-backed personas

Built from interview data, not assumptions. Every goal and pain point was mentioned by at least 2 participants in that role.

Aditya Verma

MBA Student · Age 22

A driven MBA student focused on a great job post-graduation, who finds the admission and financial process confusing. He wants one place where everything just works.

Goals

Track interview progress and feedback in real time
Pay fees without worrying about third-party links or whether payment went through
Access counselor reports and meeting history in one place

Pain points

Unclear next steps, no visibility into interview stage
External payment link felt untrustworthy, no receipt
Counselor feedback scattered across sessions

Hears: You need to pay your fees via this third-party link.
Your interview status will be updated soon.
Sees: Multiple emails with unclear instructions. Other students discussing the process in WhatsApp groups.
Says & does: Where do I check my interview status?
Did my payment go through? How do I get a receipt?
Thinks & feels: I wish there was one place to track everything.
Am I missing important deadlines?

Priya Sharma

Admissions & Finance Administrator · Age 42

Manages admissions, financial records and interview tracking across multiple tools daily, handling 2–4 data discrepancy errors a day across NPF, Pinelab, Excel and email.

Goals

Reduce manual work in the admission approval process
See student finances, transactions and interview status in real time
Automate fee processing and refunds

Pain points

Too many manual approvals: slow and error-prone
Data scattered across NPF, Pinelab, Excel and email
Delayed fee verification leads to daily student complaints

Hears: We need to approve 50+ applications today.
Students are complaining about missing payment confirmations.
Sees: Multiple spreadsheets tracking fee payments. Unorganized email requests for refund approvals.
Says & does: I need to check multiple systems to track student applications.
Compiles manual reports each morning.
Thinks & feels: I spend 4 hours a day just reconciling data.
Why can’t I see a student’s full journey in one place?

Dr. Rajeev Menon

Faculty & Student Counselor · Age 40

Has mentored students for over a decade. Without a structured system his feedback gets lost in emails, and he has no way to flag at-risk students early.

Goals

Give timely, structured feedback after interviews and sessions
Track student history and progress in one system
Reduce manual note-keeping

Pain points

Feedback often gets lost in email drafts
Manual documentation takes too much time between sessions
Hard to ensure students follow up

Hears: Your session notes weren’t attached to the student record.
A student from last month missed their follow-up.
Sees: No structured view of students he has counseled. Students re-explaining context he should already know.
Says & does: My session notes live in my professional email drafts.
Manually tracks follow-ups in a personal notebook.
Thinks & feels: I’m losing context between sessions.
I want to flag at-risk students but there’s no channel.

Motivation

Jobs to be done

User	When I…	I want to…	So I can…
Admin	start my workday	see every pending action in one view	stop opening NPF, Pinelab and Excel
Admin	update a student status	do it in one click with confirmation	move to the next task immediately
Faculty	finish a counseling session	log notes right there in the platform	not lose context or forget follow-ups
Student	check my admission status	see it instantly, no emails	stop anxiously checking my inbox
Student	pay my semester fee	complete it inside one trusted platform	have proof and peace of mind

End to end

User journey

Every role mapped across seven stages, including where the flow breaks.

StudentAdminFaculty
1 · Intake
- Submit applicationvia website
- Complete?
- Incomplete-docs email
- Upload CSV from NPFbatch · weekly
- CSV validation error
2 · Scheduling
- Wait for slot
- Slot invite receivedemail + Calendar
- Schedule + send invitesbulk · 1-click
3 · Interview
- Attend interviewvideo / in-person
- No-show
- Track interview statuslive dashboard
- Faculty no-show: reassign
- Score candidate
- Score overdue >48h
4 · Decision
- Admissions decision
- Full 100%
- Partial 25–75%
- No scholarship
- Rejected
- Reschedule
- Decision Boardpanel + admin
- Tie-break: founder
5 · Offer
- Receive offer letterBranded · MU
- Accept?
- Decline / lapse 7d
- Auto-trigger offer letterbranded · 1-click
6 · Payment
- Payment path
- Pinelab (full)
- Loan / EMI
- Deposit 1/3 · 1/6 · 1/12
- Payment failed (x3)
- Abandoned >24h
- Track paymentsreal-time
7 · Enrollment
- Enrolled: Dashboard
- Auto-syncNPF + CRM + Excel
- Sync conflict: manual merge

Step
Decision
Success end
Edge case
Critical edge

5 critical handoffs, where student and admin lanes interact

H1 · Application
Admin CSV → Student Portal
Duplicate / invalid row
H2 · Slot Invite
Scheduler → Student inbox
Wrong tz / bounce
H3 · Decision
Faculty + Admin → Student
Same-day SLA miss
H4 · Offer Letter
Admin → Student (Pinelab)
Un-branded (legacy)
H5 · Payment confirm
Pinelab → CRM/NPF/Excel
Sync race condition

Information architecture

User flow

Categories emerged through participant-led grouping during interviews and affinity mapping, then were validated in a design-team workshop before wireframing.

Whiteboard sketch of the main admissions flow

Three sticky-note boards grouping features into categories for each role

Sketch to hi-fi

Result: Replaced 4 morning-reconciliation tools with one live dashboard

1 / 5

Validation

Usability testing

Two moderated rounds, four weeks apart: mid-fi first, then the hi-fi build. The goal was to find what was broken while fixes were still cheap.

64% → 87%
Task success, Round 1 to Round 2: 7
Critical issues found before any code

You’ve just received your offer letter. Please pay your first semester fee using Dinero.
Scenario 1Student Payment Flow
Payment completed within 5 minutes
You just completed an interview with a candidate. Please update their status to “Shortlisted”.
Scenario 2Admin Candidate Management
Status updated within 30 seconds
A student just completed their counseling session. Please log your notes and set a follow-up reminder.
Scenario 3Counselor Follow-up
Note saved and reminder set within 60 seconds

Part 4 · Outcome

Final UI

The screens shipped because the research justified them, the wireframing tested them and the design system made them coherent.

Loading video

Measured

Impact

Five months of research and design changed how Masters’ Union runs admissions.

Task success rate

87%

+23 pts

Baseline: 64%R2 result: 87%

Payment completion

93%

+19 pts

Baseline: ~74%

Admin manual effort

<1 hr/day

−80%

Baseline: 3–5 hrs/dayTarget: < 1 hr/day

Tool consolidation

3 → 1

−2 platforms

Baseline: NPF + Excel + Pinelab

Reflection

What I took away

01
Stakeholders misattribute problems, research locates them
Leadership believed students weren’t completing payments. Research showed a trust failure: students were willing to pay, but the experience felt unsafe. The fix wasn’t a reminder email, it was embedding the gateway.
02
Observation reveals what interviews cannot
Admins said 1–2 hours of daily overhead. Observation showed 3–5. They had normalized the effort. I now default to observation-first for workflow-heavy users.
03
Test early enough to fail cheaply
Round 1 found 7 critical issues before a line of code. Fixing them in Figma took hours; after development it would have taken weeks.
04
The design system is a research artifact
Three role-based dashboards with different mental models still had to feel like one product. The 30-component token system encoded that requirement.

DineroInternal Platformfor Masters’ Union

The Problem

Research

Design

Aditya Verma

Priya Sharma

Dr. Rajeev Menon

1 · Intake

2 · Scheduling

3 · Interview

4 · Decision

5 · Offer

6 · Payment

7 · Enrollment

5 critical handoffs, where student and admin lanes interact

Final UI

Task success rate

Payment completion

Admin manual effort

Tool consolidation

Stakeholders misattribute problems, research locates them

Observation reveals what interviews cannot

Test early enough to fail cheaply

The design system is a research artifact