Ankit Raj
Ankit Raj

Ankit Raj

VP and Head of Product, Kapiva

I lead product and growth at Kapiva, a health‑tech D2C brand, from the first visit to the customer lifecycle.

Growth is where I started in product: I ran growth, product and marketing at Wink & Nod through its acquisition by Livpure, after consulting at KPMG and Kearney. I still build: most new products start as an AI prototype of mine, and I wrote some of our internal tools myself with Claude Code.

Read the case studies EmailLinkedInGitHub
01 / Work

Case studies

Problems I owned end to end: what was wrong, who it was for, what we built, how we proved it, and what changed.

  • 01Growth · Experimentation

    I ran Kapiva's conversion-rate optimisation programme: 40+ A/B-tested launches across PDP, checkout and payments, drawn from an analysis of 20M+ sessions and 100M+ events in BigQuery. Offers were tested hard and kept profitable with coupon personalisation and cart-level discount capping. Revenue per session rose 69% in 6 months.

    Quick view Full case study ↗
    01 / Growth · Experimentation

    A growth engine built on experiments

    Kapiva · VP and Head of Product · Owned

    I ran Kapiva's conversion-rate optimisation programme: 40+ A/B-tested launches across PDP, checkout and payments, drawn from an analysis of 20M+ sessions and 100M+ events in BigQuery. Offers were tested hard and kept profitable with coupon personalisation and cart-level discount capping. Revenue per session rose 69% in 6 months.

    Revenue per sessionBaseline → +69%
    Discount spendRising as offers scaled → A stable, healthy share of revenue
    Welcome offerSpilling over to repeat users → Use by repeat users fell
    02 Approach

    Ran a conversion-rate optimisation programme of 40+ A/B-tested launches across PDP, checkout and payments, including many offer tests. The experiments came out of an analysis of 20M+ sessions and 100M+ events in BigQuery across web and app: navigation between products, the homepage, traffic channels, Instagram and Facebook in-app browsers, and the app.

    03 Outcome

    Revenue per session rose 69% in 6 months.

    Read the full case study
  • 02AI · Health tech

    Quality checks on our doctor consultations covered a ~10% manual sample. I moved them to AI audits of every consultation, proved the scoring against consultations scored by real doctors before going live, then replaced the vendor tool with our own system. Coverage went to 100%, cost fell 70%, and the effectiveness score became accurate enough to feed the doctors' incentives.

    Quick view Full case study ↗
    02 / AI · Health tech

    AI audits for every doctor consultation

    Kapiva · VP and Head of Product · Owned

    Quality checks on our doctor consultations covered a ~10% manual sample. I moved them to AI audits of every consultation, proved the scoring against consultations scored by real doctors before going live, then replaced the vendor tool with our own system. Coverage went to 100%, cost fell 70%, and the effectiveness score became accurate enough to feed the doctors' incentives.

    QA coverage~10% sampled → 100% audited
    Cost of the audit systemVendor tool → 70% lower, in-house
    What the score is trusted forSpot checks → Doctors' incentives
    01 Problem

    Doctor-consultation QA relied on manual sampling of ~10% of consultations.

    02 Approach

    Moved consultation QA to AI audits of every consultation, then replaced the vendor tool with an in-house system. The audits produce an effectiveness score (CES) and a satisfaction score (CSS) from 14 parameters. Before going live, validated the AI against golden sets of consultations scored by real doctors; CES was off on two medical-knowledge parameters, fixed by reworking their rating scale, expanding the model's medical vocabulary and adding real cases as examples, then re-tested on the golden sets and by language (about a quarter of consultations are in regional languages). Every score cites the transcript line behind it; doctors see their scores, audits and strong and weak points in Care Studio and can dispute feedback there; humans still review 5% of consultations.

    03 Outcome

    QA coverage went from ~10% to 100%, and the in-house system costs 70% less than the vendor tool. CES became accurate enough to feed the doctors' incentive and performance system.

    Read the full case study
  • 03Lifecycle · Health tech

    Retention and LTV were Kapiva's biggest problems, and the usual commerce levers barely moved them. The real problems were adherence and expectations: people weren't sticking to the product, and expected results in weeks when it takes months. We rebuilt the post-purchase experience around doctors, diet plans and adherence journeys, on a platform built for it. LTV rose 30% and retention rates 20%.

    Quick view Full case study ↗
    03 / Lifecycle · Health tech

    Retention is an adherence problem

    Kapiva · VP and Head of Product · Owned

    Retention and LTV were Kapiva's biggest problems, and the usual commerce levers barely moved them. The real problems were adherence and expectations: people weren't sticking to the product, and expected results in weeks when it takes months. We rebuilt the post-purchase experience around doctors, diet plans and adherence journeys, on a platform built for it. LTV rose 30% and retention rates 20%.

    Lifetime valueBaseline → +30%
    Retention ratesBaseline → +20%
    Consultations self-booked0% → 51%
    01 Problem

    Retention and LTV were the biggest problems. The usual commerce levers (next-order coupons, milestone journeys, restocking offers personalised by cohort, restock nudges on WhatsApp and the app, restock calls from health coaches) helped only a little. The real problems were adherence (people weren't sticking to the product and dosage) and expectations (people expected results in 2–3 weeks; it takes 3–4 months).

    02 Approach

    Ran targeted lifecycle programmes: doctor consultations, diet plans and personalised adherence journeys with gamification. Doctor calls became guidance, not a sales pitch: setting the right expectations and answering the small questions people won't take to their own doctor. Built into the post-purchase experience, this became the health-tech charter: deliver health outcomes by solving adherence.

    03 Outcome

    LTV improved 30% and retention rates rose 20%.

    Read the full case study
  • 04Growth · Mobile

    Customers were buying on the web. I moved them into the app with three reasons to switch, coins, delivery tracking and doctor consultations, promoted on the order confirmation page and in WhatsApp delivery-tracking nudges. App attach rate rose from 8% to 35%.

    Quick view Full case study ↗
    04 / Growth · Mobile

    Moving customers from web to app

    Kapiva · VP and Head of Product · Owned

    Customers were buying on the web. I moved them into the app with three reasons to switch, coins, delivery tracking and doctor consultations, promoted on the order confirmation page and in WhatsApp delivery-tracking nudges. App attach rate rose from 8% to 35%.

    App attach rate8% → 35%
    01 Problem

    Customers were buying on the web; the goal was to move them to the app.

    02 Approach

    Moved customers from web to app through purchase-funnel and post-purchase interventions: coins for downloading the app, delivery tracking in the app and doctor consultations in the app, all communicated on the order confirmation page and through WhatsApp nudges, especially delivery-tracking nudges.

    03 Outcome

    App attach rate rose from 8% to 35%.

    Read the full case study
  • 05Monetisation · D2C

    Most orders were single packs, which held AOV down. I moved selling from the 1-pack to the 2-pack with a set of UX nudges, and watched revenue per session and LTV as guardrails so a bigger basket didn't cost conversion. AOV rose from ₹650 to ₹1,200.

    Quick view Full case study ↗
    05 / Monetisation · D2C

    Selling the right pack size

    Kapiva · VP and Head of Product · Owned

    Most orders were single packs, which held AOV down. I moved selling from the 1-pack to the 2-pack with a set of UX nudges, and watched revenue per session and LTV as guardrails so a bigger basket didn't cost conversion. AOV rose from ₹650 to ₹1,200.

    Average order value₹650 → ₹1,200
    01 Problem

    Most orders were single packs, which held AOV down.

    02 Approach

    Moved selling from 1-pack to 2-pack with UX interventions: default selection bias, "recommended" and "best results" tags, and a value-added freebie with the highest packs. Added a checkout nudge asking customers to upgrade their pack size and save. Tracked revenue per session and LTV as guardrails, since stretching AOV further reduces conversion.

    03 Outcome

    AOV rose from ₹650 to ₹1,200.

    Read the full case study
02 / Hands-on

Projects

Things I build myself with AI, for my team and for the business.

  • affiliate-platform

    Affiliate platform for gym trainers

    Built an affiliate sales platform for gym trainers solo with Claude Code.

    Claude Code
  • pm-os

    PM OS

    Built a PM OS of AI automations on Slack, Gmail, ClickUp and Calendar that my team also uses.

    AI
  • daily-briefing

    Daily briefings for my pods

    Built an AI daily-briefing system for my three pods solo with Claude Code.

    Claude Code
  • self-serve-data

    Self-serve data analysis

    Query our BigQuery event and order data myself through AI connectors.

    AI
  • prototyping

    Prototypes before engineering

    Prototype most new products with AI before engineering builds them.

    AI
03 / Path

Experience

  1. Kapiva · VP and Head of Product

    Sep 2020 – Present

    Own product, growth, CX and the customer lifecycle for a health-tech D2C brand across web, app, CRM and coaches. Lead three pods (Commerce Tech, Health Tech, Biz Tech) of 15 people across product, design and analytics.

    • Built the Health Coach channel from 1 to 112 coaches; within a year it reached ~25% of D2C revenue.
    • Launched and scaled the health app to 1M+ downloads with zero paid install spend; it reached 30% of D2C revenue within 15 months.

  2. Wink & Nod (now Livpure) · Head of Product and Marketing

    Dec 2018 – Aug 2020

    Ran growth, product and marketing for a D2C sleep products company, acquired by Livpure in 2020.

    • Scaled Wink & Nod, a D2C sleep products company, from $250k to $1.8 mn ARR in 12 months, leading product and growth; the company was acquired by Livpure in 2020.

  3. Kearney · Senior Business Analyst

    Aug 2017 – Dec 2018
    • Delivered INR 25 Cr in annual savings for a fertilizer client through digital solutions, including quality-prediction analytics.

  4. KPMG · Associate Consultant

    Jun 2013 – May 2015
    • Assisted in the technology upgradation of 70 mines of Coal India Limited (capacity of 150+ MTPA).
04 / Foundations

Education

Indian Institute of Management (IIM) LucknowMBA · 2017
Indian Institute of Technology (IIT) KharagpurBachelor of Technology · 2013
05 / Contact

Let's talk.

Email is the best way to reach me.

ankitraj.kgp@gmail.com LinkedInGitHub
esc
↑ ↓ move↵ open