All work

Case study · AXA and Swiftcover

Motor and Personal claims

A heuristic review and early research pointed at sign-in as the weak link in getting people to self-serve their claims. I set up the unmoderated A/B test in Optimizely, building the test variants myself, then analysed and observed the results to find what actually fixed it, and used what it taught us to make the case for what comes next.

Company

AXA and Swiftcover

Role

UX Designer

Timeline

2019 – 2021

Team

End users, PM, BA, Researcher, Digital Optimisation team, Engineers

Overview

The claims journey existed to do one job well: let someone register a motor or personal claim themselves, quickly, ideally right after an incident, rather than calling in. Every extra step was a chance for someone to give up and phone in instead.

What shipped

A tested sign-in variant that outperformed the existing experience, an updated call-to-action, and a mobile-optimised login — the channel most people were actually arriving from mid-claim.

Outcome

A specific tested variant measurably improved sign-in completion, and that evidence shaped the call-to-action and mobile login changes that shipped. The password-recall and account-import findings also fed directly into the passwordless and native-app sign-in thinking that followed.

The throughline: find the friction with lightweight research, test real alternatives rather than debate them in a room, and ship the one with evidence behind it.