A Friday systems thinking thread — worth sitting with:

Teams accumulate observability debt by adding metrics and dashboards without a unified label strategy, making it hard to debug incidents. This leads to increased Mean Time To Resolve (MTTR). To fix, define a label schema first and instrument second. This approach helps isolate affected customers and changes during incidents. It's essential to address this issue before it becomes expensive.