Mean Time to Detect
Also known as: MTTD
Mean Time to Detect (MTTD) measures the average time between a defect or issue being introduced and it actually being discovered — whether by testing, monitoring, or a user report. A shorter MTTD generally means less time for a problem to cause real-world impact before anyone notices.
MTTD is most commonly tracked for production incidents, where monitoring and observability tooling directly determine the number — a well-instrumented system with strong alerting catches issues within minutes, while one relying purely on user complaints might not detect a problem for hours or days.
Shift-left testing practices push MTTD down within development itself — a bug caught by a unit test seconds after being introduced has an MTTD of seconds; the same bug undetected until a UAT phase weeks later has a dramatically worse MTTD, even though it's the exact same underlying defect.
