

DriveScore
Your driving, transparently rated and rewarded
Context
Drive and performance tracking through mobile device GPS, accelerometers and gyroscopes.
Improve your driving (and your score) -
get discounted insurance, car finance and servicing offers from selected partners.
(MVP launched to a beta pool of 1,000 existing ClearScore users before I joined)
My areas of focus
Previous designs
The initial MVP launch (prior to my joining) validated the concept and technical feasibility. It lacked design clarity, feature roll-out and functional intent.




Making recording a drive more simple and reassuring
Bottlenecks
User hesitance around granting invasive permissions (Always-on GPS and notifications) before they’d even got into the app or taken a drive. Disallowing these during onboarding didn’t resurface when using the app.
Users had to cover 150 miles before their score would be unlocked and generated.
Both these limitations resulted in high drop off early on.
(Hard) technical constraint
The app requires these permissions to automatically detect drives and record performance without the user ever having to interact with the screen.
Solution
Succinct 1-page permissions information at first sign-in that clarifies what they’re for, and what is and isn’t recorded, and when.
We also eliminated the 150 miles rule. Instead just requiring the user to complete one drive, then building the score over time.
Show users more detailed score history
In the MVP:
No score detail available. Only a current ‘snapshot’ of that moment’s score.
Development
The score ring now indicates the user score and how close they are to reaching the next level of insurer threshold.
This was a new feature I introduced where more insurers who typically offer better deals are ‘unlocked’ as higher score tiers are reached. (e.g. air mile tier points) This increased return visits by 22%.
Access to score history and relative-to-all-users stats now reveals deeper insight into individual driving performance.
Provide more meaningful journey insights
In the MVP:
Only a running count of events
e.g. ‘6 speeding events this week’.
No further insight.
Development
New Insights area creates a statistical overview of how a driver’s events have accumulated over a recent period.
Allows a user to see real data about their driving performance and habits - creating a more continuous engagement cycle for improvement.
Individual events are viewable through this curated ‘most recent’ summary when deep-diving into an event type.
They can also be accessed from a recorded drive’s listed events.
Simplifying classification
In the MVP:
Users assumed their score is a running total of points.
But it’s actually graded by event frequency, and weighted over time against an established baseline - the same as a credit score. So, one incident of speeding does not equate to a one point deduction.
Development
It was important to transparently surface this to users when classifying and reclassifying their trips with expectations of big score changes.
Review and challenge recorded events
In the MVP:
No formalised dispute process in place. Only ‘Contact us’.
A frequent driver complaint from V1 was having no way to view or challenge a recorded event. 18% of users expressed high frustration at mis-classified drives not being amendable.
Development
The user now has full visibility of all events and their specific context, and control over challenging an event.
The Challenge flow establishes clear expectations and sets honest limitations from the start, and transparently reports the status throughout the process. This led to users feeling a much greater sense of trust in the product and empathy for the recording system.
Insurance that’s worth stopping (gently) for
In the MVP:
Users had to reach a score of 750 or over to qualify for discounts - or even be able to start an insurance quote. Around 16% of drivers hadn’t reached a score of 750 within their first month and felt despondent about not being able to get any quotes sooner.
Once users had reached 750 their app usage typically dropped by about 50%
Development
Working closely with the commercial team we proposed a lower entry score of 600, combined with tiered thresholds where reaching higher scores would ‘unlock’ more insurers who typically offered more competitive quotes.
We also completely removed the requirement to obtain a minimum score to complete an insurance quote. It was shortsighted to deny people access to a feature they could use on any other comparison service. It just meant that any quotes received when a user’s score was below 600 wouldn’t have any discounts applied.
Garage and car management
In the MVP:
The broad functionality and connectivity remained more or less as it was from the first iteration.
Development
I made subtle optimisations to the add / remove car flows, as well as improving the date notifications, and creating an additional route into the insurance quote flow since Insurance had now been removed as a primary nav item.