Leadership, strategy, design, product, analytics.
← Back to homeLed a cross-functional effort from initial request through post-launch to redesign the mobile claims experience, freeing up claim handlers’ time through AI experimentation, research, and a refined product vision.
Leadership believed adjusters spent too much time in the field on straightforward claims. We were tasked with finding ways to reduce that time through mobile and AI solutions.
The project required balancing ambitious AI ideas with the reality of complex claims work, ensuring that whatever we delivered actually fit into adjusters’ daily workflows.
Even simple claims involved multiple variables. Early AI experiments didn’t fully address the complexity, and many adjusters’ pain points came from duplicated work rather than field time itself.
Core tension: Early assumptions didn’t align with observed workflows, requiring a shift in product vision.
The challenge was identifying real bottlenecks and focusing on high-value fixes while keeping stakeholders aligned.
I led the design process, managing two designers and collaborating closely with PMs, engineers, data scientists, and field adjusters. I drove user observation, journey mapping, and rapid prototyping to uncover and address real pain points.
Key leadership actions:
We applied a strong product model approach: research, prototyping, and validation at each step. Rather than relying solely on AI experiments, we focused on process simplification and reducing double entry in notes.
Design principle: Observe real work before committing to assumptions. Solve for actual bottlenecks, not assumed ones.
The redesigned mobile experience and process changes led to measurable time savings, wide adoption, and further investment in the mobile platform. The work strengthened organizational support for the product model.
Strategic impact: Created a scalable model for combining AI experiments with practical process design to deliver real operational gains.
This project reinforced the importance of grounding product vision in real-world observation. By validating assumptions through direct user research, we avoided over-investing in solutions that wouldn’t address the real issues.
If I could go back, I’d refocus the team even faster once the initial AI direction proved less viable. The key lesson: real observation drives better outcomes than assumptions alone.