Improved AWS Outposts Monitoring
Led the design of a new metrics tab in the AWS Outposts console to help users monitor hybrid infrastructure performance more effectively. Through user interviews and iterative design, I helped shift users away from DIY dashboards and into a more helpful, integrated AWS experience.
Role: Lead UX Designer
Responsibilities: I worked with a product manager and engineer to scope the solution, conducted user interviews with existing Outposts customers, synthesized insights into monitoring needs, and designed a contextual metrics tab within the console.
Duration: 2 months
Note: Cannot post certain mocks and pictures due to AWS policy.
Users & Audience: Technical roles managing hybrid or edge infrastructure, such as Cloud Infrastructure Engineers / Cloud Architects, and DevOps Engineers.
Problem: Outposts users—primarily infrastructure and DevOps engineers—had no direct way to view performance metrics in the Outposts console. To monitor status, they had to switch to the CloudWatch console, remember specific Outpost IDs, and build custom dashboards from scratch. This created unnecessary complexity, especially during incident response.
Solution: Designed a “Metrics” tab in the Outposts console to surface critical performance data, prioritized by user workflows. Based on interviews, I organized metrics to support both daily health checks and incident diagnosis, reducing reliance on custom CloudWatch dashboards.
Step 1 | Identify user problem
Description
Step 2 | Gather requirements
Description
Step 3 | Research and workshop
Through user interviews, I uncovered two distinct monitoring behaviors: daily proactive checks (to ensure things are working) and reactive troubleshooting (to find what went wrong). Most users had resorted to building custom dashboards outside AWS because the native tools weren’t useful in the moment.
Step 4 | Hi-Fi designs & Prototype
I designed a Metrics tab within the Outposts detail page that prioritizes high-level health metrics at the top, with deeper system metrics available as needed. The design supports both quick daily checks and more in-depth analysis, reducing the need for external dashboards.
Results
After launch, users shared that the new metrics view made it significantly easier to assess system health without leaving the console. Several users noted they no longer needed to maintain separate Grafana dashboards.
Lessons I learned
This project taught me that even power users appreciate simplicity—especially when under pressure. By designing for their mental models and workflows, I could reduce friction without adding complexity.