Senior Product Manager · Consumer Products · Growth · Monetization · AI
I find where users drop before they reach value, and fix the product before spending more to bring them back. About 7 years in product management, across EdTech and real-money gaming.
01Documented impact
Selected outcomes from professional product work, charted with exactly the precision available: ranges stay ranges, approximations stay approximate.
Headline outcome
Day-7 retention went from 12.2% to 25.4% in a controlled rollout of the redesigned first 60 seconds of onboarding at Witzeal.
Scale
100K+
Learners reached
Across Edfora’s learning and engagement products, not a single feature.
Growth
~10% week over week
GMV growth trajectory at Witzeal, sustained over 11 months. Not attributed to any single change.
Assignment completion, D0 gameplay and Day-7 retention (changes in level, in percentage points), GMV growth (a weekly rate) and the 48% retention level use different units, so they sit in the table rather than on this axis.
| Outcome | Result | Precision | Context |
|---|---|---|---|
| Learners reached | 100K+ | Approximate | Reach of Edfora’s learning and engagement products overall, not of a single feature. |
| Assignment completion | 18% → 45% | Exact | Edfora, 2-year academic-cycle dataset: 18% under the static learning path, 45% after the adaptive system was introduced (+27 percentage points). Not attributed to the adaptive system alone. A change in level, so it is not plotted. |
| Day-7 retention | 12.2% → 25.4% | Exact | Witzeal onboarding redesign: control vs. treatment in a 3-week controlled rollout, ~50K users. |
| D0 gameplay | 12% → 33% | Exact | Share of new users who played a game on their first day, before and after the Witzeal onboarding redesign. A change in level, so it is not plotted. |
| DAU growth | 12–15% | Range | Quiz and gamification layer at Edfora, after DAU had been flat for two months. |
| Average session time | ~15% | Approximate | Quiz and gamification layer at Edfora. |
| Student retention | 8–12% | Range | Real-time engagement dashboards for faculty at Edfora, replacing a monthly spreadsheet pull. |
| GMV growth | ~10% WoW | Approximate | Witzeal Technologies: a ~10% week-over-week GMV growth trajectory sustained over 11 months, alongside an experimentation roadmap across pricing and reward loops. Not attributed to testing alone. A weekly rate, so it is not plotted. |
| Bonus spend reduction | ~20% | Approximate | Witzeal Technologies: bonus allocation moved from flat tiers to expected ROI per player segment. Retention held steady. |
| Long-term retention, stabilized | 48% | Exact | Lifecycle messaging moved from one blast to segmented cohorts at Witzeal Technologies. A retention level rather than an uplift, so it is not plotted. |
02Selected work
Each told as a decision: what we saw, what we chose, what it cost and what happened.
Two cases in depth, then two Baazi Games decisions in brief.
Only 12% of new users played a game on their first day, and Day-7 retention was ~12%. A redesigned first session, tested against a 30% control, lifted D0 gameplay to 33% and Day-7 retention to 25.4%.
Read the case study
A fixed practice sequence gave every learner the same next question. A 3PL IRT-based engine matched difficulty to each learner instead. Across a 2-year academic-cycle dataset, completion was 18% on the static path and 45% after the adaptive system.
Read the case study
PokerBaazi · Real-money poker
<60 mspeak server latency
FanBlaze · Fantasy sports · a feature sunset
<6%of match-day users used it
~2 minlonger sessions
“Users came for fantasy execution, not passive score consumption.”
03Experience
Built software first, then ran releases, then owned products. Each step shows up in how I scope, prioritize, and make trade-offs with engineering.
2014–2018
Engineering foundation
2019
Program & release management
2019–2023
Product management
2023–2026
Senior product management
Gurugram
Owned product strategy and roadmap for learning and engagement experiences on web and mobile, working with engineering, design, content and business teams.
OwnedProduct discovery and PRDs · UI/UX design collaboration · Sprint planning · Post-launch analytics: retention, DAU, feature adoption
In the decision logFaculty signalsGamification
Gurugram
Growth, onboarding, monetization and lifecycle for a real-money gaming platform.
In the decision logBonus allocationTesting roadmap
New Delhi
Experimentation, segmentation and risk across a multi-game platform including PokerBaazi, Lagai Khai and FanBlaze. Product work on the first-deposit funnel, churn, game discovery and monetization.
In selected workPokerBaaziFanBlazeIn the decision logSegmented journeys
Gurgaon
Program and release management for enterprise web applications, including EwayBill and an LMS. The step between engineering and product: owning delivery across teams and clients.
Gurgaon
Sole Android developer on a collaboration platform for the handmade industry, connecting makers, buyers and designers. Built the app from scratch and worked directly with the CEO and CTO on what to build.
In the decision logOne app, three roles
04How I work
A lot of my work has started with a number that stopped moving: flat DAU, inconsistent monetization, retention data that arrived a month late. The job is to find the behavior underneath it, choose the one change worth making, and measure it so the team learns something even when the result is flat.
Low completion looked like a content problem. It was a fit problem.
Evidence · Adaptive Assignment Engine · ProblemOnly 12% of new users played a game on D0: an activation problem, with a very different fix from retention.
Evidence · Onboarding Funnel Redesign · ProblemPick one outcome tied to user value, and treat activity metrics as diagnosis rather than success.
Evidence · Adaptive Assignment Engine · OutcomeRead the result by segment. Every result, including the flat ones, narrows the next bet.
Evidence · Decision log · Testing roadmapThen target the incentives you keep by expected return.
Evidence · Decision log · Bonus allocationKeep a stable default everywhere else. Past a point, adaptation adds unpredictability.
Evidence · Adaptive Assignment Engine · ReflectionDecision log
The signal, the call, what it cost, and what happened. Trade-offs are product reasoning; marked outcomes are documented.
Product Manager
Reward costs were eating into margin without a clear retention payoff. Bonuses were allocated in flat tiers.
Replace flat bonus tiers with expected ROI per segment, optimizing for incremental NGR (net gaming revenue) per rupee of bonus spend.
Cutting incentives in a real-money gaming product can quietly hurt retention, and that was the main risk going in. The change only reads as a win because both numbers moved the right way: spend fell and retention held.
Senior Product Manager
Students were disengaging well before faculty found out. Retention data reached faculty through a monthly spreadsheet pull, so by the time anyone saw it, it described students who had already drifted.
Replace the monthly pull with real-time engagement dashboards, so faculty could step in while a student was still reachable.
The next layer in the product documentation, myAdvisor, moves from dashboards to alerts. It specifies high and medium priority alerts, alerting at module level, a history view, filters by module and date, and deep links that open the part of the product where the teacher can act. Unread alerts are handled deliberately, and notifications are timed around a teacher’s schedule instead of firing the moment a threshold trips.
A dashboard shows everything and leaves the teacher to find the problem. An alert picks the problem for them, which only helps if it is the right problem and it arrives when the teacher can do something about it. Send too many and teachers stop reading them, so priority, timing and unread handling matter as much as the signal itself.
The retention result belongs to the dashboards. myAdvisor is shown as product design from the documentation, and no outcome is claimed for it.
Senior Product Manager
DAU had plateaued for two straight months. The response was a quiz and gamification layer, and the risk showed up early: teachers flagged the early prototypes as “too game-y”.
Keep the quiz and gamification layer, and work with design to keep the mechanics from feeling gimmicky to teachers.
The most attention-grabbing mechanics are often the ones most likely to lose teachers. In a classroom product, teacher trust is part of the engagement loop, so it can be worth trading some raw engagement for credibility.
Product Manager
Monetization results were inconsistent from one change to the next, and the diagnosis was a lack of structured testing.
Build an experimentation roadmap across pricing and reward loops, with each change written as a hypothesis and run as an A/B test.
Testing is slower per idea than shipping on conviction. The return is that every result, including the flat ones, narrows the next bet.
Supporting context: the GMV trajectory spans the period the roadmap ran in. It is not attributed to testing alone, and no starting GMV is shown.
Product Manager
A single default journey was underperforming across different player segments.
Use behavioral clustering to find the segments that actually behaved differently, then redesign the journey for each of them.
Every extra journey is something more to build, test and maintain. Segmentation pays off when the segments are few and clearly different in behavior.
Android Developer
The platform connected three kinds of users in the handmade industry: makers, buyers and designers. The alternative was separate apps for each role.
As the sole Android developer, I proposed a single app where people choose their role, then built it from scratch. Requirements came out of discussions with the CEO and CTO, and the UX was worked through with the designer before implementation.
One app means more role logic inside the product and slightly less tailoring per role. Against that: a small team, limited time and budget, one codebase to maintain, one app to run and market instead of three, and a lower technology bill.
This was an Android Developer role, not a product management role; the product contribution came from working directly with the CEO and CTO on what to build. The 400+ maker shops and 100+ designers describe the platform’s scale. They are not claimed as a result of the one-app decision.
Growth
Monetization
Personalization
Adaptive Assignments · AI Diagnostic (independent prototype) · Segmented journeys · Bonus allocation
Experimentation
Consumer Products
Onboarding Funnel · Adaptive Assignments · AI Diagnostic (independent prototype) · Gamification · One app, three roles
AI & Data Products
Adaptive Assignments · AI Diagnostic (independent prototype) · Faculty signals
Product Analytics
Onboarding Funnel · Adaptive Assignments · Faculty signals · Segmented journeys
05AI product lab
The product work is everything around the model: where rules are enough, how confidence is shown, what happens when it’s wrong, and who can overrule it.
Assess
Collect a small, representative set of learner evidence.
Diagnose
Identify the skill gaps the evidence actually supports.
Explain
Show the learner and educator why, with evidence before the verdict.
Recommend
Propose a learning path the educator can accept or override.
Practice
Generate targeted practice for the diagnosed gap.
Evaluate
Score the outcome against a rubric, not a demo prompt.
Adapt
Feed the result back into the next decision.
A conceptual AI product system, and what exists where
User signals
Diagnostic engine
Model / rules
Recommendation
Evaluation
01
Much of a learning product can run on deterministic logic. A model belongs where judgment is needed and a wrong answer is recoverable.
02
A representative test set, a rubric, named failure categories and a launch threshold, agreed before anyone argues about the demo.
03
Show the evidence, say how confident the system is, define what happens when it isn’t, and let a person overrule it.
04
If it can’t run at every learner checkpoint at an acceptable speed and cost, it is a demo feature, not a product.
06Open to Product Manager and Senior Product Manager roles
I’m open to Product Manager and Senior Product Manager roles, especially in growth, consumer products, monetization and AI, in Delhi NCR and, where it makes sense, Mumbai. If user behavior and business outcomes have to move together on your problem, I’d like to hear about it.