Most incidents don't start with a failure - they start with a shift handoff that lost one sentence.
What actually gets lost
Not the runbook. The state: "replica 3 is still catching up", "we upgraded the driver at 14:00", "the vendor ticket is open". State is context, and context is exactly what a checklist template can't carry.
The 5-line fix
End every handoff with five lines, typed by the outgoing engineer:
- What is on fire right now (or "nothing")
- What we changed in the last 12h
- What we're waiting on from someone else
- What I'd check first if it breaks
- The one thing I'm worried about
Line 5 is the magic one - the outgoing engineer's unease is the cheapest early-warning system you own.
It takes 90 seconds per shift and it will pay for itself the first time the next shift doesn't re-diagnose a live fault from scratch.
Templates + the full shift-ops system are in the 24/7 Ops Kit. More free guides: dev.to/hive80lab.
