Selected work, in depth.
Each case is a complete account: the problem, constraints, decisions, trade-offs, and honest reflection on what shipped and what didn't.
The work I'm most proud to defend in a room.
Dalvoy, 0→1: an EdTech product retaining and growing organically
The product I'm currently building from zero. Now past 1 Lakh+ downloads (12K DAU / 94K MAU), with Day-30 retention tripled (3% to 10%), growth doubling month-over-month, and 30%+ organic acquisition at a CPA under $0.10. The newly shipped Dalvoy Shorts feature lifted retention and engagement.
UPI-first checkout: turning India's payment habit into a 3× conversion lever
Overall checkout converted at 5%. UPI users converted at 12–14%. I rebuilt the flow around UPI as the default, a short product video over the price, two taps to paid: open UPI app, or change UPI app.
Activation, year over year: taking day-zero engagement from 25% to 60%
Twelve months of compounding work on onboarding, intent capture and first-lesson recommendation. Day-zero activation more than doubled on a paid base while CAC tripled around it.
Every case. Searchable by domain.
Dalvoy, 0→1: an EdTech product retaining and growing organically
The product I'm currently building from zero. Now past 1 Lakh+ downloads (12K DAU / 94K MAU), with Day-30 retention tripled (3% to 10%), growth doubling month-over-month, and 30%+ organic acquisition at a CPA under $0.10. The newly shipped Dalvoy Shorts feature lifted retention and engagement.
UPI-first checkout: turning India's payment habit into a 3× conversion lever
Overall checkout converted at 5%. UPI users at 12–14%. Video over the price, two taps to a paid order, open UPI app, or change UPI app. The default became the habit, not the gateway.
Gamifying the read-along lesson: a 4-day A/B win
Added points, streaks and per-word feedback to the read-along surface. The lift was modest, the test was clean. +2.7 points D0 activation held across all 4 days at ~2K users per bucket.
Activation, year over year: 25% to 60% on a paid base
Twelve months of compounding work on onboarding, intent capture and first-lesson recommendation. Day-zero activation more than doubled on a paid base. M0 minutes ~3×. DAU/MAU nearly doubled.
Live-session booking: from zero to 16% trial booking
Designed and shipped an in-app live-session booking flow from scratch. ~1 in 6 trial users booked a session within their first day, without cannibalising lesson activity. Calendar-first, slot-aware, paced for tutor supply.
Home Page V2: from catalogue to launchpad
The home page wasn't a catalogue, it was a launchpad. Once we treated it that way, lesson starts went up 62% and order creation followed.
Result Page: turning the end of a lesson into the start of the next one
Four shipped A/B iterations on the page every user hits. Practice up 19.5%, orders up to 37%, and payment success up 30.8% on the headline release.
Reminding users what they were buying: the checkout features bottom sheet
Surfacing premium features at the moment of payment, without blocking the CTA. Order creation rate jumped 79.8%, payment success 7%.
Building a design system from scratch
Scattered Sketch files, inconsistent tokens, no shared components. Token-first architecture built over 6 months, foundation, component library, governance, while the product kept shipping.
Redesigning the membership page to cut churn at SpeakX
20% of subscribers cancelled on Day 0, before completing a single lesson. Redesigned to surface learning progress and clarify subscription value, making the case to stay before the cancel button.
More case studies in progress
Two additional write-ups are being finalised, covering growth loops and a consumer engagement project. Check back soon.