Comparison
ToilvsToil automation
Toil
you did the same twenty-minute manual runbook every Tuesday for a year.
Manual, repetitive, automatable operational work that scales with the size of the system and produces no lasting value. It is worth naming because it hides: each instance is small and doing it feels like being helpful. The SRE convention of capping it — measuring what fraction of a team's time it consumes and treating a breach as a signal to invest — turns a slow drain into a visible, arguable number.
Full entry →Toil automation
the twenty-minute manual thing you did every Tuesday for a year became a controller that does it without telling you.
Removing repetitive manual operational work by encoding it, whether as a script, a controller or a self-service action. The economics are the point: work that scales with the size of the fleet consumes the team's capacity for improvement, which is what makes it worth automating even when each instance is small. The precondition is that the manual procedure is written down and consistent, which is why the runbook usually comes first.
Full entry →