To enhance the proactive monitoring of applications, Sentry is evolving its alerts system by introducing Change and Crash Rate Alerts. These new alert types aim to provide early warning signs of issues before they escalate, helping developers identify and address problems promptly. The system now compares metric alerts or issue alerts between periods to detect significant changes in error rates or transaction rates, triggering alerts when these changes exceed historical thresholds. This allows teams to roll back changes or investigate issues more efficiently, reducing notification fatigue and preventing issues from falling through the cracks.