Case Study 01 · Insurance · CRO
Small Tweaks,
Massive Gains
How small design changes backed by proper testing moved revenue.
Optimising the Home & Contents funnel
The challenge
After a home insurance quote flow relaunch, the team wanted to see where we could get quick wins and improve the experience for customers. We dug into the analytics and mapped where the flow was losing people.
Discovery & journey mapping
I co-hosted a 'fresh-eye' review with a number of stakeholders of the quote flow. We each called out what felt broken, and decided what was worth testing versus what needed a bigger fix.
Fresh-eye UX review of the Home Direct funnel on Miro
One of the items that was flagged an easy fix with the potential for impact was a 10% discount banner meant to push the home and contents bundle. It looked flat, sat in the wrong spot, and users were basically scrolling straight past it. I redesigned it and moved it right next to the relevant option, making it clearer which option the discount was relevant to.
Before & after
The original design buried the discount message beneath all the product options. In the redesign, the promotional banner appears directly adjacent to the "Buildings & contents" option, creating an immediate visual association between choosing the bundle and saving money.
Before: banner below, easy to miss
After: banner beside the relevant option
The CRO experiment
Before we ran the 11-week A/B test, I worked with the CRO team to define upfront what a win, a loss, and an inconclusive result would actually mean.
Small design changes weren't enough to solve the real problem. We'd need to go back to the research and dig deeper into user intent to understand what was actually getting in the way further up in the funnel.
Small design changes maybe aren't enough to shift behaviour on their own. We'd need to look at bigger, more structural changes to the page or the flow and come back to retest before drawing any conclusions.
Small design changes can make big wins. If the numbers moved, it means we were solving the right problem in the right place and that it's worth looking for more quick wins like it.
The variation delivered a statistically significant lift in final conversions, driven heavily by desktop users.
- +2.87%Overall CVR increase
- +369Incremental sales during the test period
- +4%Desktop uplift in sales completions (>95% confidence), 256 of the total incremental sales
- +2%Mobile uplift in sales completions
Product mix stayed identical across variants. People weren't buying differently, they were just actually noticing the offer. The redesign didn't shift behaviour, it just stopped letting the discount disappear into the page.
Low-friction personalisation for returning users
The challenge
Analytics showed returning users had real intent. They'd been here before, they knew what they wanted. But the homepage had no idea who they were. It was treating everyone like a first-time visitor, which meant all that intent was wasted.
Technical constraints & strategy
The CRO team wanted to target returning users, but the backend couldn't pull up previous quotes. All we had was which product line they'd looked at last time: home or motor.
Working with that constraint, I brainstormed what the possibilities were. Returning users have already seen the marketing spiel. They don't need to be convinced, they need a direct route back. Simply redesigning the home page with a personalised 'Welcome back' banner and a CTA that directs them to start the quote of the product they were looking at previously.
Results & impact
It worked. We weren't just shuffling people around, we were actually helping them finish their task faster.
- +7%Lift in overall motor sales starts and completions (95% confidence)
- +20%Lift in desktop motor sales (99% confidence)
- ~Desktop and baseline conversions remained flat overall
- +23%Lift in mobile home sales (98% confidence). On mobile, cutting straight to the point landed better
What I learned
UX and CRO brings a strong combination. Having a framework in place before we ran the tests meant we could actually act on what we found instead of debating what it meant. Small changes can be just as impactful as a large project.