Private Access

Please enter the password to view this page.

Case study: 01

The evolution of Cube

Design system journey, from "Colouring it green" to a Variable-driven foundation

In December 2021, I joined Oplo. One month later, we merged with Tandem and Alium. Three financial companies, zero synergy. I was promoted to Head of Design with a mountain of technical debt and a CEO who believed design was just "pretty pictures." My mission was to prove that design is a lever for operational efficiency and speed to market.

My first task was a rush rebrand. We "coloured the app green" and built a basic library (V1). I wanted a universal system for Web and App, but the business saw no value. They killed the project and broke up my team. I fought back, eventually convincing a new Product Director to let us build a new website using a second iteration (V2). The site launched, but the app remained a legacy mess.

Rebellion (V3 - The Ninja Mode)

A year later, the "Green Hub" project arrived: 6 high-stakes eco-banking features. Business still refused to fund a new Design System. So, we went into "Ninja Mode." We spent 20% of our time building Cube 3.0 (Sherwood) in total secrecy. We set new standards (Trinity Framework): Accessible by design, data-driven, and multi-sensory design. We built the infrastructure under the radar because we knew the legacy app couldn't handle the future.

Ultimatum

The moment of truth came when the Tech Lead joined our side. He realised that without a system, the 6 features wouldn't ship. We issued an ultimatum to the board: Build on Cube 3.0, or we will fail to meet the deadline. They finally gave us the green light, but with a brutal two-week deadline to complete the entire system.

The Failure & MAP

In our rush to prove the value, I made a mistake: we over-engineered. We built Cube Sherwood "just in case"—too heavy, too many components without context. With the 2-week deadline looming, I had to pivot. I coined MAP (Minimum Achievable Product): we ruthlessly stripped the fat to deliver only the essential logic. We shipped Green Hub in December, but the "speculation trap" had left the system bloated.

Purification (V4 - Richmond)

May 2025: The business decided to rebuild the entire app from scratch. This was my chance to "clean the trash." I took everything we learned from the failures of Sherwood and started Cube 4.0 (Richmond).

The Goal: Cutting the fat.
The Tech: Migrating to Figma Variables, Aliases, and Tokens.
The Automation: Experimenting with AI and MCP to sync Figma directly to GitHub.

Impact

The "Ninja Mode" groundwork didn't just save the project; it future-proofed the bank. It allowed us to ship Green Hub on schedule and transitioned the team seamlessly into the high-stakes Cash ISA launch.

Engineering Velocity: 40–50% increase in delivery speed (Validated by Tech Lead).
Design Velocity: 75% faster concept-to-prototype cycles.
Cultural Shift: Transformed the Board's perception of design from a "visual luxury" to a core technical asset.

My role

As Head of Design:

My role was about making a strategic bet. When the business said no, I made the executive call to move the project "under the radar." I managed the politics and protected my team’s time so we could build in Ninja Mode. By the time the bank realised they needed a system to scale, we were already 75% of the way there.

As Principal Designer:

I was the architect of the system's logic. I defined the technical standards for the Trinity Framework and led the implementation of Zeroheight, ensuring we finally had a Single Source of Truth for our documentation and code.

Testimonial

"I estimate that for Cash ISA and Green Hub, we accelerated engineering velocity by 40-50% compared to building from scratch. Without the DS, we would have lost months to pixel-pushing and testing individual inputs. Instead, we focused purely on business logic. The ROI is huge."