AutoScout24, part of the Scout24 group, is Europe’s largest online marketplace for new and used cars, with 10 million users per month and more than 2.4 million vehicle offers. Their marketplace is accessible through their apps for iOS, Android, and also web. AutoScout24 hired Novoda, a digital product agency, with a clear objective in mind — to form a team focused on the product’s growth, with three key metrics to improve; user acquisition, user activation, and user engagement.
I got involved in their journey as part of Novoda, together with two other designers and a few developers. The Growth Team operated in a fast-paced environment, with the primary task of generating ideas that could potentially improve the three key metrics, and heavily use A/B testing frameworks to monitor the success of those ideas. This took place only in the Android app, which served as a testing ground, with all successful ideas eventually making their way to the iOS app.
A solid starting point
Prior to our involvement, the AutoScout24 team had already compiled a backlog with high-level ideas that could potentially lead to better metrics. My role in the project was to take these high-level ideas and turn them into tangible user experiences, collaborate with developers on building them in code, and work with product owners to define their success once they went out into the public.
Starting to work on a new idea was the one of the most exciting parts of the process. Forming teams with product owners and designers from both AutoScout24 and Novoda, we would have quick sessions of sketching and wireframing, typically in a Crazy-8 format, and then build up a high-level user experience through our collective ideas.
From ideas to tangible solutions
From then on, my role was to refine the rough user experience into fully fledged wireframes and user flows, with all possible states and scenarios considered. More often than not, these were made with good old pen and paper, which allowed me to review the potential solution with the wider team quickly while limiting their understanding to the core experience, instead of unnecessary polish.
Our established process permitted me and the rest of the designers to rapidly assess technical specifications and limitations with developers. Along the entire process of designing a feature, there was frequent communication with developers to advise on feasibility, difficulty of development, and specifications.
The most entertaining part for me was creating prototypes of the features, as I really enjoy prototyping. My tool of choice is Flinto, because it allows me to start prototyping fast without much prior preparation and file organisation, like most other prototyping tools do. The prototypes were used to assess the usability of the design by the designers, but also among people in the office to get an idea of how customers might react. The real tests, though, were the A/B tests that went out to actual AutoScout24 customers!
Creating a killer onboarding experience
The process of designing an onboarding experience was the most insightful part of this entire project for me. One of our hypotheses was that we could increase user activation by promoting the most valuable features of the app in a one-time onboarding experience visible to new users. After a few rounds of internal user testing, I decided on a final direction for this experience, which promoted three main features, each accompanied by a relevant illustration. But the real fun was in the animations!
I took this design a step further by coming up with an intricate animation that would occur when the user interacted with the interface. I worked very closely with a brilliant Android engineer to make sure that what I had in mind was achievable, and once we were sure that it was, I started storyboarding an animation. Within a couple of days, using the onboarding illustrations created by another designer, I used After Effects to bring the animation to life and nail the details.
Airbnb had recently released Lottie, a tool that made it ridiculously easy to display any sort of animation on iOS and Android. The Android engineer and I spent a few days closely working together to leverage this tool into displaying the animation in its intended way. We built something so good that we decided to make it open-source for everyone to use! I’ve also written extensively about working with Lottie on Medium.
Within the first two weeks of the experiment, we saw 52% of the users completing the onboarding experience, while uninstall rates and reviews remained stable. This was an achievement for AutoScout24 and eventually the onboarding experience was released to the full user base. Similarly, most other features the rest of the team and I built proved to be effective and eventually were fully released.
The most valuable thing I learned from this project was to think of the success of a product based on clear and measurable metrics. I understood key aspects of a product’s growth, such as user acquisition, activation and engagement, and how A/B testing can be used to validate potential solutions that can improve the experience of the customer base. On top of that, I learned a lot about using a powerful tool like Lottie to infuse products with appealing and engaging animations.