2 posts
An engineer's runbook for fixing AWS SES email deliverability issues before AWS places your account under review or pauses your sending — exact thresholds, the CLI to check them, a triage playbook for the first hour after a review notice, and the bounce and complaint handling pipelines so it doesn't happen again.
Amazon SES is excellent at sending email, but nearly silent about what happens after. There are no meaningful built-in alerts for rising bounce rates, complaint spikes, delivery degradation, or template failures. As a result, many teams either spend weeks stitching together SNS, Lambda, EventBridge, and CloudWatch pipelines, or they discover problems only after customers stop receiving emails. This post breaks down the operational side of SES: the four metrics that actually matter, the thresholds that should trigger action, and how to design an alerting system that surfaces issues before they become incidents. It also explores why observability and team-wide visibility are often the missing layer in otherwise reliable SES setups.
