WORK/

The company had just gone through a PE acquisition, transitioning from 40+ years of family ownership to investor-backed growth. The mandate was modernisation: move off local servers, migrate to the cloud, and improve the platform experience along the way.

That was the brief. What the job actually required was different.

There was no design process. No system. No shared language between design, product, and engineering. Before any screen could improve, the conditions for good design work had to exist. Building those conditions and proving their value to a business that had never had them was the real job.

The first year was about establishing credibility as much as shipping work. I introduced UX into a product culture that had not had it, ran the first proper user research the company had done, and started building a design system from scratch.

Over time the practice grew. A team of six designers at its peak. A design system covering around 1,000 components, with variants, documentation, and clear standards, replacing a fragmented UI built up over a decade. DesignOps introduced so the team could work efficiently rather than reinvent the wheel on every project. Cross-functional product squads implemented at the leadership level, replacing the siloed handoff culture that had slowed everything down.

None of it existed before. All of it outlasted me.

The legacy platform migrated to the cloud. The core product was rebuilt from the ground up, mobile-friendly, simplified, and designed around the people who actually used it rather than the internal logic of the assessments themselves.

One problem that had always bothered me: candidates completed psychometric assessments and often received no feedback at all. Results were gated behind trained practitioners or HR leads who might never pass them on. We fixed that. A dedicated application gave every person in the platform, admins, managers, candidates, employees, direct access to their own results, with feedback and data visualisations that made the science readable without a manual.

The assessments themselves were rebuilt for the new platform. Teams integration added. An AI coach layered in. A product that had been a back-office tool for HR started to become something employees actually wanted to use.

Towards the end of my time there, the product direction shifted from assessment tooling towards something broader: using the insight generated by assessments to improve how people work together, not just at the point of hire, but day to day.

I was part of shaping that direction from a product and UX perspective. What the platform now describes as Connection Intelligence, using behavioural and psychological insight to help managers and teams perform better together, was work we were doing before it had a name.

Significant leadership change during my seven years meant product direction shifted more than once. A roadmap built under one CPO looked different under the next. Strategic priorities changed. The platform had to flex with them.

Making UX work through that kind of ambiguity, keeping design quality consistent when the destination keeps moving, is something most case studies do not talk about. It is also the thing that is hardest to do, and the thing design leaders get tested on most in practice.