03
Bancontact Pro
A catalogue-based point of sale feature helping Belgian merchants process faster transactions with fewer mistakes at the counter.
A catalogue-based point of sale feature helping Belgian merchants process faster transactions with fewer mistakes at the counter.
Client
Client
Banctontact
Banctontact
Type
Type
P design
P design
Role
Role
Lead Product Designer
Lead Product Designer
Year
Year
2026
2026
The challenge
The challenge
Bancontact Pro users operate in high-pressure environments, festival counters, pop-up markets, event bars. Every transaction is a race against impatience. The existing app required sellers to manually key in amounts, mentally apply discounts, and maintain accuracy under load. In hectic conditions, human error is not an edge case. It's a predictable cost.
~59%
of merchants were still using cash as their primary payment method — no structured digital selling tool in their workflow at all.
3
manual steps minimum to process a single transaction in the existing flow: recall the price, key in the amount, generate the QR. No catalogue, no shortcuts.
8.5
clicks to activate a core feature that should take 3 — the discoverability baseline we had to beat
Research & findings
Research & findings
CASH IS STILL KING
CASH IS STILL KING
Most micro-merchants and small retailers still rely on cash and manual bank transfers as their primary payment method. Digital tools exist in their toolkit — but none are built for speed at the counter.
Most micro-merchants and small retailers still rely on cash and manual bank transfers as their primary payment method. Digital tools exist in their toolkit — but none are built for speed at the counter.
THE FLOW WORKS. THE FRONT DOOR DOESN'T.
THE FLOW WORKS. THE FRONT DOOR DOESN'T.
Once inside the catalogue, users completed multi-product payments in 12.5 seconds with 4 clicks — matching the theoretical minimum. The problem was getting there. Only 14.9% found the feature activation efficiently. 31.7% gave up entirely.
Once inside the catalogue, users completed multi-product payments in 12.5 seconds with 4 clicks — matching the theoretical minimum. The problem was getting there. Only 14.9% found the feature activation efficiently. 31.7% gave up entirely.
DEMAND IS SEGMENT-AGNOSTIC
DEMAND IS SEGMENT-AGNOSTIC
Micro-entrepreneurs and small retailers rated the feature almost identically — a mean value score of ~5.0 across both groups. The pain isn't limited to one merchant type. It's structural.
Micro-entrepreneurs and small retailers rated the feature almost identically — a mean value score of ~5.0 across both groups. The pain isn't limited to one merchant type. It's structural.
HIGH SUCCESS MASKS HIDDEN FRICTION
HIGH SUCCESS MASKS HIDDEN FRICTION
88%+ task completion looked strong on paper. But heatmaps told a different story — scattered clicks, systematic exploration, users succeeding through persistence rather than clear interface guidance.
88%+ task completion looked strong on paper. But heatmaps told a different story — scattered clicks, systematic exploration, users succeeding through persistence rather than clear interface guidance.
Design Principles
Design Principles
The Solution
The Solution
A mobile-first catalogue mode within Bancontact Pro, letting sellers build a structured product list organized by categories, with per-item pricing, and a checkout flow that generates a QR code without manual input. The experience collapses a multi-step mental process — recall price, apply discount, enter amount, verify — into a tap sequence.
I designed and built the clickable prototype used in both Dutch and French research sessions. After the interview data came back, I analyzed the findings alongside the PM and researcher, then translated them into high-fidelity mobile designs. Once the mobile version was stable, I extended the work to web — maintaining consistency while adapting interaction patterns for a larger surface.
A mobile-first catalogue mode within Bancontact Pro, letting sellers build a structured product list organized by categories, with per-item pricing, and a checkout flow that generates a QR code without manual input. The experience collapses a multi-step mental process — recall price, apply discount, enter amount, verify — into a tap sequence.
I designed and built the clickable prototype used in both Dutch and French research sessions. After the interview data came back, I analyzed the findings alongside the PM and researcher, then translated them into high-fidelity mobile designs. Once the mobile version was stable, I extended the work to web — maintaining consistency while adapting interaction patterns for a larger surface.
Accessibility & Inclusive Design
Accessibility & Inclusive Design
The feature serves a wide merchant population: solo micro-entrepreneurs at farmers' markets, small teams running festival stands, retailers with catalogs ranging from 10 to 500+ items. Designs had to work in bright outdoor light, with one hand, at speed. Text sizing, tap target dimensions, and contrast ratios were calibrated for use without ideal conditions — because ideal conditions are not where this product lives.
The feature serves a wide merchant population: solo micro-entrepreneurs at farmers' markets, small teams running festival stands, retailers with catalogs ranging from 10 to 500+ items. Designs had to work in bright outdoor light, with one hand, at speed. Text sizing, tap target dimensions, and contrast ratios were calibrated for use without ideal conditions — because ideal conditions are not where this product lives.
Key Learnings
Key Learnings
Building for a product used by hundreds of thousands of people changes how you think about validation. A hunch isn't good enough — it needs to survive 161 test sessions across two language communities before it earns a development sprint. This project sharpened my approach to prototype-first thinking: validate the concept, surface the friction points, then design with data behind every decision. It also showed the value of respecting existing mental models while still expanding what users believe the product can do.
Building for a product used by hundreds of thousands of people changes how you think about validation. A hunch isn't good enough — it needs to survive 161 test sessions across two language communities before it earns a development sprint. This project sharpened my approach to prototype-first thinking: validate the concept, surface the friction points, then design with data behind every decision. It also showed the value of respecting existing mental models while still expanding what users believe the product can do.
