How can I handle a high bounce rate with emails that I send using Amazon SES?
The bounce rate of my Amazon Simple Email Service (Amazon SES) account is higher than normal. It's putting my account at risk for an account review or a sending pause. How can I identify what's causing the increase in bounce rate and resolve it?
Use a bounce and complaint monitoring system
To avoid a high bounce rate or quickly address increasing bounces, it's a best practice to implement a bounce and complaint monitoring system. The following are mechanisms that you can use to monitor bounce and complaint sending activity and rates.
You can use Amazon Simple Notification Service (Amazon SNS) to monitor bounces, complaints, and deliveries. You must configure Amazon SNS notifications on each identity that you intend to monitor. To view example notifications, go to Amazon SNS notification examples for Amazon SES.
To use event publishing to monitor email sending, you must create a configuration set. Event publishing works only when you send the message using a configuration set. You can configure a default configuration set for each active sending identity, or specify your configuration set when you send the message.
Also, set up an Amazon SNS event destination or an Amazon Kinesis Data Firehose event destination. Setting up an event destination allows you to get detailed information in JSON format. See example event records for Amazon SNS and Kinesis Data Firehose.
Email feedback forwarding
Email feedback forwarding monitors only bounces and complaints. Email feedback forwarding is turned on by default, but you can turn it off if you're using an alternative monitoring system. The information that you receive from feedback email is limited compared with Amazon SNS notifications or event publishing.
To get the most information about your bounces and complaints, it's a best practice to use Amazon SNS or event publishing as your monitoring system.
You can use the Reputation metrics dashboard in the Amazon SES console to monitor bounce and complaint rates. Create reputation monitoring alarms using Amazon CloudWatch. It's a best practice to set up CloudWatch alarms and AWS Lambda to automatically pause email sending when the bounce or complaint rate exceeds a certain threshold. Pausing allows you to investigate the cause of the increased bounce or complaint rate. It's a best practice to maintain a bounce rate that's under 5% and a complaint rate that's under 0.1%.
Identify what's causing the increased bounce rate
To stop the bounce rate from increasing, you must first identify the email addresses that are causing the bounces. Use the information from the Amazon SNS notifications or Amazon SES event data to find the email addresses.
To identify what's causing the increased bounce rate, review the bounceType and diagnosticCode fields in the Amazon SNS bounce notifications or event data. For information on bounce events, see Bounce types.
Stop sending email to the addresses that are increasing your bounce rate
After identifying the email addresses that are resulting in bounces, stop sending messages to those addresses so that your bounce rate doesn't continue to increase. Be sure to remove the email addresses from your email recipient lists.
Turn on the Amazon SES account-level suppression list
When the account-level suppression list is activated, the email addresses that result in a hard bounce are added to the suppression list. Subsequent messages to these addresses aren't sent, resulting in a lower bounce rate.
Resolve the root cause of the increased bounce rate
The following are some of the most common reasons for an increase or sudden spike in bounce rate:
- You send messages to an email list that contains recipients with invalid mailboxes. This can happen if you use an incorrect or out-of-date list of recipients. You must remove any invalid email addresses from the list before sending messages again. For more information on managing email lists, see Amazon SES best practices: Top 5 best practices for list management.
- Based on your reputation metrics, mailbox providers can block the IP address that you're sending email from. If there's a large number of subscriber complaints, then a mailbox provider might block your messages. Follow the guidance that you find on the mailbox provider's postmaster site to remove the block.
- Mailbox providers can use a third-party blocklist to filter email. If you continue to send messages to a mailbox after your messages are blocked, then your bounce rate can increase. You can contact the mailbox provider or blocklist provider to get information on why your messages are blocked. After you're removed from a blocklist, you must address the issues and change your sending habits accordingly. Be sure to review the policies for removing Amazon SES IP addresses from blocklists.
Review the design of your sending application
After you identify the root cause of the increase in bounces, review the design of your sending application. The design can impact your bounce metrics. For example, your application sends a confirmation email to a user when they sign up, and the user enters an invalid email address. The confirmation email then results in a hard bounce. Another example is setting up your application to retry sending emails whenever there's a failure. If there are issues on the recipient's mailbox, then you can continue to get bounces.
For more best practices and actions that you can take for successful email sending, see Building and maintaining your lists. For more information on ways to keep a low bounce rate, see What can I do to minimize bounces?
Automate how bounces are processed
You can also implement a solution to automate how bounces are processed. See the following examples:
- Handling bounces and complaints
- Maintaining a healthy email database with AWS Lambda, Amazon SNS, and Amazon DynamoDB
Best practices for sending email using Amazon SES
What is considered a soft bounce on Amazon SES, and how can I monitor soft bounces?
Configuring notifications using the Amazon SES console
- AWS OFFICIALAktualisiert vor 2 Jahren
- Wie verhindere ich die Fehler ThrottlingException oder „Rate überschritten“ bei der Verwendung von AWS Glue?AWS OFFICIALAktualisiert vor einem Monat
- AWS OFFICIALAktualisiert vor einem Jahr