Edvoy Funds
Rebuilding a broken financing operation into a product that generated ₹70 Lakhs profit in its first intake after relaunch.
4.5 Months
Product Designer
70 Lakhs profit on its 1st intake
4,000+ Active Agents
50,000+ Monthly students
A failing loan service, earned back from WhatsApp.
The brief said “fix the loan flow”. What I found was a system where every stakeholder had already built a better workaround than the product. Vendors on WhatsApp. Counsellors on spreadsheets. Students going straight to banks. The real job was earning adoption back from people who had already moved on.
₹70L
profit in the first intake after relaunch
1–2 days
loan processing, down from 1 to 2 weeks
48 hrs
express approval turnaround
6
platforms rebuilt end to end
A live rebuild of the counsellor’s Funds view with sample applications. Vendors update it on their own. Click it to open a loan’s timeline.
02 - The 30-second version
The infrastructure existed. Almost none of it was used.
The challenge
Vendor partnerships, a counsellor team and a vendor portal all existed on paper. In practice, requests were collected by hand, details went over WhatsApp and email, commissions lived in spreadsheets, and some vendors closed Edvoy’s leads and reported nothing back.
What I did
Rebuilt Funds across six surfaces around one question: what would make people willingly move their work back in? Consent-first applications, regional vendor routing, adaptive documents, a vendor portal built for vendors, OCR on approval letters, contextual notes and tasks, and a commission engine that follows the money.
My role
Interviewed students, counsellors and vendors before touching a screen
Reframed the problem from usability to value
Designed all six surfaces, from the student app to cloud admin
Designed the commission engine and multi-disbursal model with engineering
Made the sequencing calls on what not to build yet
The constraints
Better workarounds already existed: WhatsApp is fast, spreadsheets are flexible
Ten vendors, ten rulebooks for commission, region and eligibility
Money arrives in tranches over years, so commission can’t be one event
Six surfaces had to ship together to close the loop
The key insight
Every workaround I found was rational. The system had stopped being worth the friction, so people built better alternatives. That isn’t a design failure. It’s a value failure, and you can’t fix a value failure with better UI.
03 - Key decisions
Three calls that made people come back
Decision 01
Start at the offer letter, with consent.
The loan journey now starts the moment a student has an offer, inside the Edvoy app. Agents can apply on a student’s behalf, but only after the student approves on their own phone. Accountability from step one.
Decision 02
Fewer leads, sent to the right vendors.
A student from Chennai goes to the three or four lenders that cover South India, not all ten. Less noise, more relevant leads, and vendors started taking them seriously.
Decision 03
Pay on disbursal, close the loop.
Commission follows what’s disbursed, not what’s approved. Every tranche is verified, split between vendor and agent with tax, invoiced automatically, and the loop stays open until the full amount is drawn.
04 - Context
The most damaging pattern was the accountability gap
Vendors could, and did, close loans that started with Edvoy and claim the commission themselves, because there was nothing to dispute it with. Revenue was walking out the door and the business had no reliable way to know how much, or where.
What was approved
Approval letters sat in chat threads and inboxes.
What was disbursed
Tranches happened off-system, sometimes never reported.
What commission was owed
Spreadsheets, maintained by memory and goodwill.
Who did what
No audit trail. “I already sent that” had no answer.
So we asked
How do we build a system valuable enough that people willingly move their work back into it?
05 - Try the product
The rebuild, as working pieces
Each card is a real part of Edvoy Funds, rebuilt in code with sample data. They play on their own. Click one to use it.
Adaptive documents
Collateral or not, ask only for what’s needed.
Vendor portal, rebuilt
Four steps vendors actually do.
OCR on approval letters
Values read and pre-filled for review.
Notes and tasks
The WhatsApp replacement, on the profile.
Loan status, for students
Every stage in plain words.
Weeks to days
Where the time went, before and after.
Revenue visibility
Earned, pending and projected, live.
Who did what
Every action attributed and timed.
Counsellor pipeline
Every student’s loan stage, filterable.
06 - Outcome
Everything visible. Nothing off the table.
₹70L
Profit in the first intake after relaunch, across six rebuilt surfaces. Processing dropped from one to two weeks to one to two days, with express approvals inside 48 hours. The reaction that landed hardest internally was real-time commission tracking: for the first time the business could see what was earned, pending and projected without chasing anyone.
07 - Reflections
You stop asking how it should look, and start asking what it has to replace
Better UI wasn’t the answer
The workarounds were genuinely good. The shift was from “how do we make the portal better” to “what does this portal need to be worth to a vendor who already has a faster option?”
The quiet feature did the most
The commission engine is what I’m proudest of technically. The biggest difference came from loan status inside the tools counsellors already lived in.
Don’t make behaviour change the entry fee
The best products make the new way easier than the old one. Funds is a case study in earning adoption back from workarounds that had every reason to exist.