Case studyE-commerce & D2C10 min read

Vanaaya: rebuilding a D2C Ayurveda brand's store so it loaded fast, sold subscriptions and stopped losing customers at checkout

An Ayurvedic skincare brand had outgrown its template store: pages were slow on mobile, checkout lost shoppers, and customers who loved the products forgot to reorder. We built Vanaaya's new headless storefront, a one-page checkout with COD verification, subscriptions, a skin-quiz recommender and the operations tools behind them — 26 features built for mobile shoppers in India.

26
Features shipped
3
Apps & platforms
7
Team members
16
Weeks to rollout
Vanaaya storefront on mobile and desktop
Vanaaya: a storefront built for mobile shoppers.
01

A beautiful brand on a slow store

Vanaaya's founders had built something customers loved: small-batch Ayurvedic skincare with a loyal following on Instagram. Their store, built on a popular template with a pile of plugins, had not kept up. Product pages took several seconds to load on a mid-range phone, the checkout had four steps and a surprise shipping fee, and cash-on-delivery orders came back undelivered at a painful rate.

Their best customers reordered — but only when they remembered, which was often after running out and trying something else.

“Our customers love the oil. They just forget to buy it again.”

— Vanaaya's co-founder
02

Finding where shoppers left

We started with the analytics and a round of session recordings. Most traffic arrived from Instagram onto product pages on mobile, and a large share left before the page finished loading. Of those who added to cart, many abandoned at the shipping step. COD orders from first-time buyers had the highest return-to-origin rate.

We also interviewed repeat customers. They wanted two things: a way to reorder without thinking, and guidance on which products suited their skin.

Where the store was losing money
  • Slow product pages on mid-range phones
  • Multi-step checkout with a late shipping surprise
  • COD orders from new buyers returned undelivered
  • No subscription or reorder reminders
Laptop displaying an analytics dashboard
Funnel analysis: speed and checkout were where most shoppers were lost.
03

A headless rebuild

We rebuilt the storefront on Next.js with a headless commerce backend. Product pages are pre-rendered and served from the edge, images sized for each device, and scripts from third-party plugins removed or deferred. The brand's look stayed; the weight did not.

Checkout became a single page: phone number first, address auto-filled for returning customers, shipping cost shown upfront, UPI as the fastest option, and COD confirmed by a WhatsApp message before dispatch.

04

The team

A design-led team, with performance treated as a feature from the first sprint.

1
Engagement lead

Founder workshops, funnel analysis and launch plan.

1
Product designer

Storefront, checkout and skin quiz.

2
Frontend engineers

Next.js storefront and performance.

1
Backend engineer

Commerce backend, subscriptions and integrations.

1
AI engineer

Recommendation model and quiz logic.

1
QA engineer

Checkout, payment and device testing.

7 people in total, working as one team.

05

Decisions we made

Each was agreed with the founders, weighing brand experience against conversion.

01

Stay on the template platform or go headless?

  • Optimise the existing template
  • Headless storefront with a commerce backend

Our call: Headless storefront with a commerce backend. Plugins made the template slow and fragile. Headless gave full control over performance and checkout while keeping a robust commerce backend.

02

How to handle risky COD orders?

  • Disable COD
  • Keep COD with WhatsApp confirmation and risk rules

Our call: Keep COD with WhatsApp confirmation and risk rules. Many first-time Indian shoppers want COD. Confirmation and simple risk rules cut undelivered orders without losing those customers.

03

How to drive repeat purchases?

  • Discount emails
  • Subscriptions plus usage-based reorder reminders

Our call: Subscriptions plus usage-based reorder reminders. Customers reordered when products ran out. Timing reminders to each product's typical usage period matched how people actually shop.

04

How should recommendations work?

  • Generic bestsellers
  • Skin quiz with a recommendation model

Our call: Skin quiz with a recommendation model. Customers asked for guidance. A short quiz combined with purchase data recommended routines that felt personal.

06

The 26 features

Everything in the new store and the tools behind it.

Storefront
Fast on every phone.
  • 01Edge-rendered product pages

    Pre-rendered pages served close to shoppers.

  • 02Responsive image optimisation

    Images sized and compressed for each device.

  • 03Ingredient stories

    Rich product pages with ingredients and rituals.

  • 04Reviews with photos

    Verified buyer reviews and photos.

  • 05SEO-friendly structure

    Clean URLs, structured data and redirects preserved.

Checkout
One page, no surprises.
  • 06One-page checkout

    Phone, address and payment on a single page.

  • 07Upfront shipping costs

    Delivery cost and date shown before checkout.

  • 08UPI, cards and COD

    All common payment methods in India.

  • 09COD confirmation on WhatsApp

    Orders confirmed before dispatch.

  • 10Address auto-fill

    Returning customers check out in seconds.

Retention
Remember to reorder.
  • 11Subscribe and save

    Recurring deliveries with a discount.

  • 12Reorder reminders

    Timed to each product's usage period.

  • 13Skin quiz

    Five questions to a personalised routine.

  • 14Personalised recommendations

    Products suggested from quiz and history.

  • 15Loyalty rewards

    Points on purchases and referrals.

Operations
The tools behind the store.
  • 16Order management

    Orders, returns and refunds in one console.

  • 17Shipping integration

    Labels, tracking and NDR handling with the courier aggregator.

  • 18Inventory sync

    Stock synced with the warehouse.

  • 19Subscription management

    Pause, skip and change subscriptions.

  • 20Promotions engine

    Coupons, bundles and free-gift rules.

  • 21Analytics dashboard

    Conversion, repeat rate and subscription revenue.

  • 22RTO risk rules

    Flags for high-risk COD orders.

  • 23Role-based access

    Permissions for operations, marketing and founders.

  • 24GST invoicing

    Compliant invoices for every order.

  • 25Customer support view

    Order history and status for support agents.

  • 26Content management

    Marketing edits pages without developers.

Vanaaya skin quiz and personalised ritual
The skin quiz: five questions, one personalised routine.
07

Launch

We launched the new store with all existing URLs redirected so search rankings and Instagram links kept working. The first month showed what the funnel analysis predicted: more shoppers reached checkout, and fewer left it. Subscriptions, introduced as "subscribe and save" on bestsellers, quickly became a meaningful share of orders.

  1. Weeks 1–2
    Discovery

    Funnel analysis, session recordings and customer interviews.

  2. Weeks 3–5
    Design

    Storefront, one-page checkout and skin quiz.

  3. Weeks 6–13
    Build

    Headless storefront, checkout, subscriptions and integrations.

  4. Weeks 14–15
    Migration

    Products, customers and URL redirects.

  5. Week 16
    Launch

    New store live with subscriptions on bestsellers.

08

What we learned

Speed is a brand experience. A beautiful page nobody waits for is not beautiful.

Retention is built into checkout. Subscriptions and reorder reminders did more for revenue than any new acquisition channel.

Built with
  • Next.js
  • Headless commerce backend
  • PostgreSQL
  • Razorpay with UPI and COD
  • Shiprocket
  • WhatsApp Business API
  • Recommendation model
  • Vercel edge
CTA Background

Building something like Vanaaya?

Bring the problem as it actually is, constraints included. You will get a straight answer on whether we are the right team.

View Our Work
AI-First Engineering
Secure & Scalable
Built to Deliver Impact
More case studies