🔒 IT diagnostics SaaS platform • 2026
Incident Investigation Timeline
Redesigning an interactive event timeline to help engineers find what caused a system crash.
Project brief
Move an old timeline to the new design system, and fix the flow where we can
This is another page of the same enterprise IT tool that support engineers use to investigate problems on single devices. The whole product is moving to a new design system, page by page. The main goal was the new design system, but we also improved the flow wherever it didn’t need big backend changes. I worked on it from discovery to handoff and reviewed key decisions with the design lead and the product manager.
Context
Show what happened on a device, in one place
When a device crashes or slows down, engineers need to see errors, system changes and performance alerts in time order. The timeline page puts them on one chart, with a table of records below. Engineers use it to find the cause and, on Windows, to roll the device back to an earlier restore point.
One chart, five data types and two different click results
The chart was overloaded, and it was not clearly connected to the table below it. The main action, restoring the system to an earlier point, was hard to find.
The same engineers, different data on each platform
Engineers checking Windows devices
They see all five data types and can roll the device back. They need a fast path from the cause to the restore action.
Engineers checking macOS devices
They get less data and no restore points. They need to understand quickly why a track is empty.