Simplifying registration to convert more users

A third of people never finished signing up. I found out why, then fixed it.

Role: Product Designer
Company: Caxton
When: Feb 2024 – July 2025
Result: 10% reduction in drop off rate

Problem

34%

of people dropped out during registration.

1Heuristic review

5Usability tests

100Session recordings analysed

What I found

The product choice screen stopped everyone. Users either dropped off here or spent minutes trying to decode it.

The original Caxton product choice screen, showing four card options side by side

Small input details created dead ends. A red strength meter on strong passwords, a date picker that fought you, an address field that loaded above where you were looking, and no error states to explain any of it.

The password field showing a strength meter reading Very Strong, coloured in Caxton red

The verification screen read as a failure. People created second accounts rather than following the instruction to check their email, which the error styling buried.

The identity verification screen, styled as a red error message

What we changed

Cut the product choice screen down to the job people came to do.

The original product choice screen, four card options side by side — Before

The redesigned screen asking what would you like to do now, with two clear options — After

Fixed the inputs. Green meter for strong passwords, a plain date field, address results below the postcode, error states everywhere.

The password strength meter reading Very Strong, coloured in Caxton red — Before

The same field after the change, with the strength meter now green — After

Rewrote verification as a next step, not an error.

The original verification screen, styled as a red error message — Before

The reworked screen asking the user to verify their email, with a clear next step — After

Result

The drop off rate fell by 10%

24%Drop-off after

1000+More completed sign-ups in first two months

Final thoughts

This is a fantastic example of how small changes can have a big impact. I opted for this approach as my first project exactly for this reason. Other obvious changes to improve the UX of Caxton required substantial backend work that we couldn't prioritise at that point in time.

Working at Caxton taught me to be patient with heritage products, and to build my process around the company whilst also advocating for the end user. There was a lot of ground work to be done. When I joined, we were not doing user research. I had to build trust within the company, and find the right stakeholders who could help me progress. Within a short time I had ongoing research with our users and I was able to move the needle on the bigger projects.

These included an end-to-end redesign of the consumer app, an overhaul of the B2B desktop offering, and an internal fraud detection tool for the compliance team.