Collect

01

MSMEs in India operate on trust-based credit. A jeweller might have 200 customers with open balances — tracked in a notebook, followed up on WhatsApp, disputed over the phone. No automation, no digital record, no visibility. Tracking this manually across hundreds of relationships is a full-time job, not a business function. I spent weeks on field research: visiting distributors in HSR Layout, speaking with jewellers in Bengaluru, understanding the actual paper workflows before I touched Figma.

02

Through 20+ user interviews, I identified 3 core pain points: (1) Forgetting who owes what — memory fails at scale, (2) Awkward follow-up calls — chasing a customer for money damages the relationship, (3) No digital record for disputes — when a customer claims they paid and there's no proof, trust breaks. These three became the product pillars: a real-time ledger, automated non-awkward reminders, and an audit trail.

03

The hardest design problem in a payment app isn't the flow — it's trust. When money is involved, a single confusing screen costs the user real rupees and the business a support call. I designed every confirmation state to overcommunicate: amounts are large and readable, recipient names are prominent, and every action has an explicit success state. Error messages are human — not 'Transaction failed (ERR_402)' but 'This payment didn't go through — check the account number and try again.' Trust is earned one interaction at a time.

04

Collect surfaces the right information at the right time. The home screen shows today's pending collections at a glance — no digging. Sending a payment reminder is two taps and uses pre-written, non-awkward language so the business owner doesn't have to choose their words. Adding a customer is a single short form. The full ledger is always one tap away. The app does the cognitive work so the business owner can focus on the relationship.

Outcome