When to use

When building or reviewing feeds, lists, media-heavy screens, dashboards, search, maps, chat, or async feature flows.
When users may experience slow network, large data sets, repeated refreshes, image loading, or expensive rendering.
When UI feels correct but slow, jumpy, blocked, or battery-heavy.

Core analysis model

Separate actual performance from perceived performance.
Define first useful paint, interactive readiness, refresh behavior, and completion feedback.
Choose loading UI: skeleton, placeholder, progressive content, cached content, or blocking spinner.
Review list rendering, pagination, virtualization, image loading, caching, prefetching, and memory pressure.
Check animation cost, gesture latency, tap feedback, layout shifts, and avoidable re-renders.
Plan offline, stale cache, retry, background refresh, and optimistic update behavior where relevant.