From Paper Scraps to Pixels: A Case Study in Digital Habit Formation πβ‘οΈπ±
The story of how we uncovered a deep-seated user behavior and designed a beloved feature that replaced the pen-and-paper habit for millions of Indian merchants.
Embedded Files
Project Overview OkCredit is a leading digital ledger app for millions of small businessΒ owners across India, helping them track credit and payment transactions. During user research, we discovered a crucial gap π: merchants weren't logging all their transactions in the app.
The smallest, most frequent, or short-term credits weren't being recorded digitally. Instead, they were jotted down in a physical kaccha (rough) notebook or on a slip of paper π§Ύ. If the credit was settled within a day, the paper was simply torn up. This informal system was fast but fragile, prone to lost records and manual errors π₯.
The challenge was to digitize this deep-seated kaccha habit, creating a digital experience that was as fast as scribbling on paper but with the reliability and power of a digital tool.
My Role as Lead Product Designer π¨
My Role as Lead Product Designer π¨
As the Lead Product Designer on this project, I was deeply involved in the end-to-end process, partnering closely with our Product Manager, Mukul Taneja. My responsibilities included user research to understand the core problem, ideation sessions, designing the prototypes for testing and final high-fidelity visuals for development.
The Challenge: Understanding the 'Kaccha' Habit π€
The Challenge: Understanding the 'Kaccha' Habit π€
Our initial research set out to answer one question: "Why don't merchants add all transactions to OkCredit?"β The answer was consistent across various merchant types, from local kirana (grocery) store owners to tea stall vendors: they don't record short-term or small-amount transactions.
These transactions were handled in a separate, informal system: a rough notebook π. This was their temporary ledger for customers who would pay back quickly.
Key Insights from User Research π‘:
Need for Speed π: In a busy shop, merchants need to record a transaction in seconds. Speaking is often faster than typing.
Temporariness is a Feature β³: The informal nature of the kaccha book was by design. It separated the "Pakka" (permanent) long-term credit from the "Kaccha" (rough) short-term debt.
The Pain of Paper π«: While familiar, the physical method had clear pains. One merchant told us how he used to lose his parchi or drop coins from his pocket from repeatedly taking it out.
Phase 1: The Hypothesis and an MVP π―
Our goal was to migrate this rough note-taking behavior into the OkCredit app. To compete with the speed of pen and paper, our design for the Minimum Viable Product (MVP) was built on one principle: simplicity β¨.
The initial design, "Kaccha Note," prioritized:
A Prominent Entry Point π: Placed directly on the home page for instant access.
Speech-First Input ποΈ: We hypothesized that voice note would be the fastest method of entry and built our own voice assistant trained specifically on Hindi and common merchant chits (notes).Β
Effortless Management πͺ: Notes could be struck out with a single tap to mark them as complete and deleted with one more click.
A significant challenge with this voice-first approach was a tooling limitation β οΈ. Our primary prototyping tool, Maze, lacked voice prototype testing features. This meant we were unable to create and validate a fully interactive voice prototype pre-launch, making our reliance on post-launch data and user feedback even more critical.
Phase 2: Data-Driven Iteration π¬
We launched the MVP and monitored the data. The results were not what we expected π€¦. The data showed us what was happening, but not why.
The Quantitative Data (The "What") π:
Low Adoption: While 21,500 users opened the feature, only 6,000 added their first entryβa conversion rate of just 28%.
High Drop-off: The funnel from opening the page to adding the first note was only 20.2%. A staggering 37% of users who landed on the page never even clicked the microphone button.
The Qualitative Feedback (The "Why") π¬: To understand the friction, We conducted follow-up interviews with merchants. This revealed the flaws in our initial assumptions.
Voice Input Was Flawed π: A milk supplier was delighted with the feature until Google Voice failed to understand his dialect, forcing him back to his diary. Another merchant mentioned it felt "awkward to speak in front of the customer."
Users Feared Data Loss π±: As all the data was stored locally on the phone, merchants feared losing it permanently. This became a critical issue for users with multiple devices, who discovered their notes did not sync
Lack of Flexibility π€: The biggest request by far was an edit option. Without it, a single speech-to-text error forced the user to delete and start over.
It Didn't Scale π: A tea stall owner who added over 90 notes a day found the feature unusable for reconciliation because entries weren't grouped by customer.
Designing the Solutions π οΈ: This feedback was invaluable. I went back to the drawing board and designed key iterations to solve these core user pains
β Introducing an Edit Option: The most critical fix, allowing users to quickly correct errors from speech-to-text.
β Adding Keyboard Input: We added a manual text option to cater to users who preferred typing or were in a noisy environment.
β Ensuring Data Persistence: We implemented a backend backup to sync notes and alleviate fears of data loss, later adding a Google Drive backup option.
β Also introduced Photo Attachment
Future Ideation ππ‘
Beyond the immediate fixes, we also explored future enhancements to make the feature even more powerful:
Note Categorization ποΈ: We thought of categorizing notes into different types like credit, payments, and customers to improve organization.
Integration with Permanent Ledger : We were also thinking about creating a pathway to move a temporary "kaccha" note into the permanent ledger ("pakka khata") after a certain period.
The Final Business Impact π
The Final Business Impact π
The iterative design changes, driven directly by user feedback, transformed the feature from a novelty into an essential tool.
Successfully Changed User Behavior π: The biggest win was seeing users abandon their physical notebooks. Merchants reported they "stopped using copy completely" and now write "almost everything on kaccha note."
Drove Incredible Engagement π₯: Retained users were highly engaged. One merchant opened the feature 100 times a day, while another peaked at 100 notes added in a single day. We achieved a solid Day 1 retention rate of 8.5%.
Became a Growth Driver π: Kaccha Note became a hook for OkCredit users who started recommending the app to other retailers just to use this feature.
Uncovered New Use Cases π: We saw merchants organically adopt the feature for personal to-do lists, inventory management, and even noting down shopping lists from their spouses.
Learnings and Takeaways π
Learnings and Takeaways π
π Data Tells You What, π¬ Users Tell You Why. The quantitative data showed a drop-off, but it couldn't tell us about the social awkwardness of speaking into a phone. Combining analytics with qualitative interviews is essential to get the full picture.
π‘ An MVP's Goal is Learning. The initial voice-only MVP wasn't a failure; it was the fastest way to learn. Its limitations generated the precise feedback we needed to build a feature users truly loved.
π "Simple" Doesn't Mean "Lacking Essentials." While our goal was simplicity, we learned that once users begin to rely on a feature for important data, essentials like editing and backup become non-negotiable. Their definition of simplicity includes trust and control.
Author: Punna Jagadeesh
Page updated
Google Sites
Report abuse