The SLA you signed and never tested
Every proposal has the SLA table. Four-hour response for Critical. Same-day for High. It reads well in the contract review meeting. Nobody goes back and times how long a Critical ticket takes from alert to human hands on keyboard, not on a slow day, on a bad one, at 2am, with the on-call engineer three tickets deep already.
The first time a client tests that number for real is during an incident, which is the worst possible moment to learn you are twenty minutes slower than the paper says.
The prompt:
You are building an honest read on whether an MSP's SLA holds up in practice.
Context: [the timestamps for alert, acknowledgment, and first substantive action on your last ten Critical-severity tickets]
Build:
Real average response time, calculated against the contracted SLA
The ticket with the widest gap, flagged, so you know exactly where the process breaks