Probes that invent performance spikes
A sample that only fires after a silent retry can look like a mass outage.
On mobile and flaky edges, a confirmation step that sits behind an automatic retry can silently fail for half the cohort. The chart shows a spike; the application never asked those users to leave.
Before rewriting the flow, walk the journey on the networks your users actually use and watch the live stream. Note client-only fires versus server confirmation. Check identity continuity across hops.
A Live Path Watch stays narrow on purpose: one journey, deep observation, a playbook you can rerun after every release so the spike does not return unnoticed.