PRODUCT MANAGEMENT · FULL CASE STUDY · 2024–2025
Registech
Moving rural schools near Jaipur off paper - one painful job at a time, starting with the one nobody had time for: enrollment.
- TIMELINE
- Apr 2024 – May 2025
- ROLE
- PM · Research, Design & Build
- STACK
- React · Supabase · PWA (offline-first)
- METHOD
- Jobs-to-be-Done · Field research
HOW THIS STARTED
I didn't set out to build school software
A school near Jaipur. I expected some kind of system - a spreadsheet, anything. There was nothing digital. Admissions, attendance, fees, all in paper registers living in one person's head.
THE REAL PROBLEM
The problem was never that they needed an app
The easy story is "no tech, give them tech." Wrong. Every tool assumed steady internet and someone comfortable exploring an app. Neither was true here.
"Every tool I looked at was built for someone who already knew how to use it. None of them were built for the person actually keeping the school running."
WHO IT'S FOR
One user. Everything else bends around him.
The administrator - accountant, registrar, and IT department, all one tired, capable person who's run this school on paper for years.
Primary
Rural school admins
Primary
First-time software users
Later
Teachers logging attendance
Later
Parents wanting fee receipts
THE BET
The whole system was the temptation. So I didn't build it.
Ten features handed to a first-time user isn't power, it's noise. I picked the single most painful, most frequent job and bet everything else could wait: enrollment.
Chosen: start with enrollment. Earn the rest.
One job, done well enough to build the trust that buys the next feature.
HOW I PRIORITIZED
RICE scores picked v1. MoSCoW explains the rest of the roadmap.
Every feature I was tempted to build, scored and sorted. The top three cleared the bar together and became v1. Everything below waits.
| Feature | Reach | Impact | Conf. | Effort | Score | MoSCoW |
|---|---|---|---|---|---|---|
| One-screen enrollment | 10 | 3 | 90% | 2 | 13.5 | Must |
| Searchable records | 10 | 2 | 90% | 1 | 18.0 | Must |
| Offline-first sync | 10 | 3 | 80% | 3 | 8.0 | Must |
| Attendance tracking | 8 | 2 | 70% | 3 | 3.7 | Should |
| Fee receipts | 6 | 2 | 60% | 2 | 3.6 | Should |
| Report cards | 4 | 2 | 50% | 4 | 1.0 | Could |
| Parent update feed | 3 | 1 | 50% | 3 | 0.5 | Won't yet |
Score = (Reach × Impact × Confidence) ÷ Effort. Rows in accent shipped in v1.
WHAT I SHIPPED
Three things, each designed to feel like less, not more.
One-screen enrollment
Same fields, same order as the paper form.
Searchable records
Type a name, get the student. No page-flipping.
Offline-first
Saves on-device, syncs when a signal shows up.
What it actually looks like
TRADEOFFS
Every cut was a bet about what a school could actually absorb.
Simple over powerful
Mirrored the paper form instead of a 'better' one.
One job over the whole system
A narrow thing that works beats a complete thing that overwhelms.
Familiar over modern
Looking like the register he trusts, not a polished app.
Offline over online
Steady internet isn't a given here. This was the floor.
WHERE IT LANDED
Three schools said yes. That's the part I'm proud of.
One admin enrolled a student digitally without me walking him through it. That small, unglamorous moment was the whole thing.
3
Schools signed on as pilots
0
Manuals needed to onboard
1
Job done well before the next
WHAT I'M TAKING FROM THIS
Building for low tech-literacy rewired how I think about scope.
Learned to subtract
Almost every feature I removed made the product more likely to get used.
Trust is the feature
You earn the second try by making the first thing obviously useful.
Meet people where they are
It looks more like a register than a dashboard. That was the right call.
This one wrapped after three pilots and a lot learned about building for people who'd never used software like this. If you run a school like this, or you know one, I'd still love to talk.