Incident response: our postmortem process that actually prevents recurrence

our team recently revamped our incident postmortem process, and it's made a huge difference in reducing repeat incidents. historically, postmortems were often just about finger-pointing or quickly closing tickets without real learning. we shifted to a truly blameless culture, focusing on system and process failures rather than individual mistakes. key changes included: mandatory action items with clear owners and deadlines, a monthly review meeting to track progress on all outstanding action items, and a robust classification system for incidents to identify common patterns. we also started distributing the postmortem summaries widely to ensure learning across teams. since these changes, our repeat incident rate has dropped by about 60%. what elements of your postmortem process have you found most effective in actually preventing recurrence? are there specific templates, tools, or cultural practices that have really fostered a learning environment from incidents? i'm always looking for ways to improve this critical feedback loop.