We have several servers running in AWS that send emails through SES. The number of emails sent is very low, usually fewer than 10 per day across all servers.
Recently, we were informed that our bounce rate has crossed the 10% threshold, putting our account into review. This was caused by a couple employee email addresses being disabled. Some of our older servers were sending automated report emails to those addresses, and they bounced for a while before anyone noticed. We didn't have the SES Suppression List enabled at the time, we've enabled it since.
The problem we're having is that the bounce rate goes up when an email bounces, but doesn't go down when an email is successfully delivered. We're tracking bounces, but I haven't found any documentation that says we need to track successful deliveries. I don't know what configuration I might be missing.
Here is a graph from CloudWatch showing our deliveries, bounces, and bounce rate:
The deliveries and bounces are each graphed as the sum over a 24hr period. The bounce rate is, I believe, cumulative over a long period of time.
The occasional stand-alone bounces came from customers or employees putting an invalid email address into a password reset form, or something similar. We send so few emails that those single bounces were enough to visibly push the bounce rate up.
I expected the bounce rate to be a simple cumulative bounce rate = bounces over time period / sent emails over time period calculation, but then the bounce rate should be going down when the denominator increases and the numerator doesn't. That isn't what we're seeing, it only ever goes up.
Do we need to be explicitly tracking our successful deliveries and reporting them to CloudWatch somehow?
