Case study · Reward by Design · Khatabook · 2021

A form nobody filled, turned into a shop.

Khatabook wanted to enable loans, and loans needed KYU: Know Your User. The form converted at under 2%. So we made the form disappear: merchants built their own shop on screen, answer by answer, and inaugurated it. 17,000 merchants completed it in two weeks.

My roleLead Designer
Users10M+ merchants across India
TeamDesign · product · analytics · motion
Timeline2021
ScopeKYU onboarding · Behavioural design · Motion
Outcome17,000 merchants in 2 weeks · 78% engagement
01 · The business situation
The most important screen was the most ignored.

Khatabook wanted to enable loans for its merchants, and to underwrite a loan you have to know the user. KYU asked the essentials: where the business is located, what kind of business it is, whether it is GST registered, how many staff it employs. Standard, boring, necessary.

As a plain form, it converted at under 2%. The research told us why: low digital literacy, cognitive overload, and a trust deficit. In the field interviews merchants said it plainly: "It asks too many things at once. I get confused." And: "Why do you need all this data? Will it be safe?" The brief said improve completion. The question I asked first was: why would a merchant care?

02 · The strategy · Reward, don’t manipulate
We didn’t gamify. We acknowledged effort.
The merchant wasn’t filling in a form. They were opening their shop.

The obvious play was gamification: points, badges, progress bars. We rejected it. Gamification treats users like they need to be tricked into finishing, and that is manipulation wearing design’s clothes.

The real insight was simpler. Small business owners in India are proud of their shops. The shop is the identity. So we made the product reflect that: as a merchant answered, their shop visibly took shape on screen, and when the last answer landed, they cut the ribbon and inaugurated it. Grounded in four behavioural principles: reward anticipation, cognitive ease, completion bias, and micro-feedback, and carried by the craft the research demanded: progressive disclosure instead of a wall of fields, plain language instead of legal jargon, and error recovery that never restarts the journey.

The reward, in motion
Every answer builds the shop. The last one cuts the ribbon.

The build is personal. The kind of business a merchant runs decides the shop that rises on their screen: a retailer's storefront, a wholesaler's depot, a manufacturer's unit, a distributor's warehouse. The GST answer dresses the shop with its registration, and completing KYU earns the inauguration: ribbon, garlands, celebration.

The main scene: the shop assembles answer by answer.
01 · Retailer
02 · Wholesaler
03 · Manufacturer
04 · Distributor
The GST answer
The inauguration
03 · The team, and how I ran it
A tight pod, run like a growth experiment.

A small pod: design, product, analytics, and motion. The empathy came first, from field visits to real shops, interviews about how merchants see digital finance, and observation on the old devices and weak connections they actually use. Every choice after that was a hypothesis with a metric attached, instrumented before it shipped, then A/B tested and validated in Hindi, Kannada, and other local languages.

The leadership work was mostly restraint: holding the line against stakeholder pressure to add points and badges, and making the case that respect converts better than tricks. The results settled the argument.

04 · The design moves
Four principles, made tangible.
01Reframe the jobNot filling a form. Opening a shop. The mental model changed before any screen did.
02Ask one thing at a timeProgressive disclosure replaced the wall of fields, in plain language instead of legal jargon, so nobody had to feel confused to answer.
03Build progress you can seeLocation, business type, GST, staff: each answer puts another piece of their own shop on screen, so effort has a visible, personal payoff.
04Earn the inaugurationFinishing KYU is not a success toast. It is a ribbon cutting on the shop they just built.
05Answer the trust question"Why do you need all this?" got a straight answer at the moment of asking, and errors never sent anyone back to the start.
06Measure respectfullyCompletion and engagement, instrumented honestly. No streaks, no dark patterns, no guilt.
05 · Business outcome
Under 2% became 17,000 merchants in two weeks.
17,000Merchants completed KYU in 2 weeks
under 2% to 78%Engagement with the KYU flow
10M+Merchants on the platform
0Dark patterns shipped

The form that converted under 2% ended at 78% engagement, and 17,000 merchants completed KYU in the first two weeks. Abandonment fell, completion got faster, and support costs came down with it. It became the most successful project I worked on at Khatabook, and it unlocked the data the lending business was built on.

The unmeasurable outcome was the more important one: merchants saw themselves in the product. A bookkeeping utility earned an emotional connection, which is the hardest currency in consumer fintech.

06 · What I carry forward
Design should reward behaviour, not manipulate it. The difference is respect.

Gamification asks: how do we get users to do what we want? Good design asks: how do we make doing the right thing feel good? I have applied that distinction to every product since, at every scale.

I wrote up the KYU engagement work in more detail here: How we improved KYU engagement to 78% at Khatabook.

Next story
All stories.