Task success rate improved by 18%
An end-to-end redesign of Garda Oto, Asuransi Astra's digital insurance platform, focused on modernizing the UI and simplifying the claim and submission flow for everyday users.
Insurance
Web & Mobile
Role
UI/UX Designer
Timeline
3 Months
team
2 Designers, 1 PM, 3 Engineers
platform
Web App & Mobile App
Buying insurance should not feel like filling out a tax form
Garda Oto had the right product. Vehicle insurance from one of Indonesia's most trusted names. But the digital experience told a different story. The web and mobile platforms were visually outdated, the submission flow had too many steps with no clear progress indication, and users regularly dropped off before completing their application. Internal teams were facing the same friction on the dashboard side. Data was hard to read, operational tasks took too many clicks, and the overall experience had not kept up with how users expected digital products to feel in 2023.
What the redesign delivered
The redesign covered the full digital surface of Garda Oto. Customer-facing web and mobile platforms were rebuilt with a cleaner visual language and a significantly simplified submission flow. Internal dashboards were restructured for clarity, giving operational teams faster access to the data they needed most.
18%
Task success rate improved across web and mobile
15%
Reduction in submission drop-off rate
20%
Improvement in operational efficiency on internal dashboard
Research
Design Screen
What Asuransi Astra taught me
Eight months in one of Indonesia's most established insurance companies. Here is what stayed with me.
1. Regulated industries have a different kind of constraint
Insurance is not like designing a social app. Every field in the submission form exists for a legal reason. Every piece of data shown on the dashboard has a compliance implication. Designing in this environment taught me to ask a different question before removing anything: is this here because it needs to be, or because nobody ever questioned it? The answer was often the latter, but you have to know the difference.
2. Trust is the primary UX problem in insurance
Users did not drop off because the form was too long. They dropped off because they were not sure they could trust the process. An unclear progress indicator, an ambiguous field label, or a generic error message was enough to make someone close the tab. Designing for trust in a financial product is not about adding security badges. It is about making every step of the flow feel intentional and transparent.
3. Internal tools deserve the same attention as customer-facing ones
The internal dashboard was an afterthought for most of the project timeline. It was not customer-facing, so it was not prioritized. But the people using it every day were processing claims, managing customer data, and making decisions that directly affected the customer experience. A bad internal tool creates a bad customer outcome. That realization changed how I think about every product I work on now.
4. Usability testing in a corporate environment is harder than it sounds
Getting access to real users at a company like Asuransi Astra required navigating multiple approval layers. By the time we got to testing, we had limited sessions and limited time. It forced us to be extremely precise about what we were testing and why. Every session had to count. That constraint made me a sharper researcher.
5. What I would do differently
I would have pushed for a design system earlier. We built components as we went, which created inconsistencies that we had to reconcile later. Starting with a shared component library from week one would have saved significant time and produced a more consistent result across web and mobile.