eC+ · Fintech · Concept proposal

Smart Repayment Guard

The product is good at “how much can I take?” and silent on “should I?”. The dangerous moment isn’t applying for credit — it’s the repeat withdrawal under stress. This proposes a context-aware safety layer that surfaces the future cost of a withdrawal without blocking the user.

Concept / proposal — not shippedSenior Product Designer — self-directed, end to end.

Role: Sr. Product Designer
Client: eC+
Domain: Fintech
Type: Concept / proposal

Cover — Smart Repayment Guard, concept / proposal, not shipped; a context-aware safety layer that surfaces the future cost of a withdrawal without blocking the user — 01Cover

The problem — good at “how much can I take?”, silent on “should I?”; the dangerous moment is the repeat withdrawal under stress, not the credit application — 02The problem

The reframe — a trust problem, not a compliance problem, positioned between paternalistic blocking and complicit non-intervention, aiming to warn without blocking — 03The reframe

Principles — based on heuristic analysis of the withdrawal flow plus churn/support signals, with three commitments: context, foresight, and choice — 04Principles

Solution overview — a four-stage flow: risk detection, future snapshot, soft alternatives, and acknowledgement, with the user and system action at each stage — 05Solution overview

The intervention — a future-snapshot overlay on a real easyCredit Plus payout screen shows the rate rising from 157€ to 283€ (+126€) after the payout, with softer alternatives offered — 06The intervention

Your rate after this payout — 157€ rises to 283€, +126€ for 36 months — 06The intervention

The trade-off — long-term growth over short-term conversion; a smaller withdrawal today favors a retained, higher-value customer tomorrow — 07The trade-off

How we would measure — projected targets and a test plan, not results, tracking late-payment rate, trust/NPS delta, churn after payout peaks, and payout-frequency stability — 08How we would measure

What I would watch — the risk threshold is a designer's assumption, not a tested rule, so Risk and Data would co-own it before launch — 09What I would watch

Next case study — 03

A calmer way through the job hunt

All work