For 2022, Webbula is launching a series of blog posts about email deliverability topics. We have a variety of esteemed authors from the email industry lined up to participate.
|Previous Articles in the Deliverability Series|
Whenever we send an email there are a few options: either it never leaves the sending platform, it is delivered successfully, or it bounces. Email bounces can either be synchronous or asynchronous, which means that the receiving mail server will either immediately reject the mail for an individual recipient or group of recipients while the connection with the sending server is still open or they may send a rejection notification after the connection between the sending and receiving server has been closed. The distinction here doesn’t make a huge difference but it can help explain why metrics don’t always line up when Suppressed + Failed Sends + Bounces + Successful Deliveries don’t equal the expected 100% of the list size.
What is a “good” email bounce rate?
This varies a bit by list type either by B2C/D2C or B2B. Most of the top B2C brands I’ve worked with have email bounce rates below 0.5%, while B2B bounce rates tend to be higher due to employee turnover. For B2B senders I would still hope to see a bounce rate under 2.5%.
Even if your email bounce rate is in the good range, there is still a lot that can be learned by analyzing your bounces, which I recommend doing regularly. For senders with higher volume and more frequent sends I recommend at least a weekly analysis, while senders with lower volume and more infrequent sends should still do a review of the bounce logs at least monthly.
What can I learn from analyzing my email bounces?
Reviewing the messages that receiving mail servers return when rejecting your mail can be crucial to understanding why your mail is being rejected. Is the IP blocked? Is the Domain blocked? Is there something in the content resulting in the mail being blocked? Does the recipient address no longer exist?
Here are some examples of email bounces that could be returned by the receiving mail server:
- The email account that you tried to reach does not exist. Please try double-checking the recipient's email address for typos or unnecessary spaces. Learn more at https://support.google.com/mail/answer/6596
550 5.1.0 sender rejected. Please see https://www.spectrum.net/support/internet/understanding-email-error-codes for more information. AUP#In-1310
- 554 rejected due to spam URL in content
- 554 5.7.1 [VI-1] Message blocked due to spam content in the message.
- 550 Rejected because XXX.72.148.102 is in a black list at bl.spamcop.net Blocked - see https://www.spamcop.net/bl.shtml?XXX.72.148.102
- 550 Failure 550 found in RBL (bl.spamcop.net) XXX.72.148.114
- 554 5.7.1 IP Block-listed in RBL bl.spamcop.net
- 554 5.7.1 The client IP was present in the following DNSBL: bl.spamcop.net
- 550 The sender XXX.72.148.10 is in a black list bl.spamcop.net - https://antispam.nexicom.net/info#bl.spamcop.net
Looking at the examples above we can see that many email bounce messages are very helpful and actually contain URLs pointing to resources that contain additional information. Reviewing the content hosted at the URLs will hopefully help guide the sender on how to unblock their mail. Others are a bit less helpful, such as those indicating that the message was rejected due to containing “spam content” or a “spam URL” but they at least help point the sender in a direction as to where the problem lies. Some are pretty much useless however if they only contain a scrap of information such as a numeric code like “550” and nothing else.
It is also interesting to note that the last 5 bounce messages all contain different text, but represent the same core reason for the mail being blocked, a blocklisting at SpamCop. A frequent question I am asked is whether or not a particular blocklist matters, and reviewing your bounce messages is a great way to contextualize that question and determine whether or not a blocklist matters to your specific mail program.
What can I do about what I’ve learned from analyzing my email bounces?
Before you begin taking action on the information pulled from your bounce logs, it’s important to assess the priority of the different core bounce reasons such as the content or a blocklisting. To do this I like to use a little statistical analysis which you can follow along with at home by downloading this handy spreadsheet. Assessing priority is crucial because there are 1001 ways to waste your time optimizing deliverability. The key is to focus your limited time on the efforts that are going to yield the greatest impact for the business, and ignore issues that are relatively minor or require more effort to resolve than the benefit they yield.
To perform this analysis I typically like to use Excel or Google Sheets and a variety of formulas to count bounces containing a variety of different values. Though be aware with larger data sets this analysis can be performed using SQL if necessary since Excel is limited to 1 million rows and Google Sheets is limited to 5 million cells. Though I will caution against trying to load a large database on a computer that is not designed to handle it, from personal experience I can assure you this will result in Sad Times™.
With the following data set, I focus on the bounce messages contained in column B when producing my analysis in the 2nd tab of the spreadsheet.
Ultimately using a few simple formulas we are able to output an analysis that looks like this:
To create this analysis I first counted all of the bounce messages in Column B using the following formula:
To show the Percentage of Total that this represents I then divided the value in cell B2 by the absolute reference to the cell $B$2 using the following formula:
Using an absolute reference ensures that the cell being referenced will remain the same even if the formula is applied to other cells.
Then to look for specific keywords or phrases I use the following formula to count the number of bounces containing that value:
Though it is important to be aware that the formula above will only look for cells containing exactly the value contained inside the quotes. To allow for the word or phrase to appear anywhere within the bounce message in each cell, it is necessary to use the * before and after the search term or criteria in the formula as in the example below:
=COUNTIF(Bounces!$B:$B, "*spam content*")
In this case we can see that bounces from domains hosted by the ISP Spectrum represents the biggest problem for this sender. Though this data should always be compared to the list weight for a particular mailbox provider to ensure you are not wasting precious time and effort to get unblocked at a mailbox provider that is not materially impactful to your bottom line. We can also see that a significant amount of the bounces are somewhat useless in that they only contain the bounce code 550 without any additional information. When choosing keywords or phrases to search for I recommend scrolling through some of the bounce messages and looking for the shortest phrase possible that uniquely encapsulates the same driving cause.
Bounces can be an absolute treasure trove of information if you have the tools to analyze them, and getting mail unblocked at various blocklists, spam filters, and mailbox providers can provide your program with some easy wins. Though it is always important to address the underlying causes such as data quality or poor engagement that result in the mail being blocked or you are going to be trapped in a never-ending cycle of analysis and unblocking and will never reach email nirvana.
Maximize marketing ROI and connect with your audience on a deeper level with Webbula email hygiene.
Enjoy a free test, today!
About the Author
LoriBeth Blair, founder of the independent deliverability consulting firm Platonic Ideal & frequent writer for Webbula
LB is the founder of the independent deliverability consulting firm Platonic Ideal. Their focus is email intelligence and helping senders maximize the value of their email program through deliverability monitoring and support, as well as analytical insights. LB has worked with senders of all shapes and sizes from those only sending a few hundred emails a day to those that send over 1 billion emails per month and everything in between. Her specialities include email infrastructure design, authentication, DNS configuration, and data analytics.
You can find LB and Platonic Ideal over on Linkedin: