Making Sense of E-mail Bounces

In my first column I spoke about de-mystifying e-mail deliverability and why e-mail marketers need to master it. I also alluded to a smart bounce management system as a key resource, as it provides the data for performance evaluation, list management, and practice improvement.

Admittedly, our industry needs to come to grips with how that data is used in calculating its key metrics, such as “delivered.” But regardless of the calculation methodology, the fact remains that the reliability of your answer depends on the accuracy and completeness of the data that goes into it. And that’s where your bounce management system comes into play as the source of the data as well as your tool for accessing it.

In making sense of the inconsistent, sometimes misleading bounce data received from the ISPs, direct marketers should approach bounce management like they would any other database challenge. The same rules apply. There are five factors to consider in assessing your bounce management system:

Capture all data streams. Make sure that your system collects all data streams – both synchronous and asynchronous bounces. A synchronous bounce occurs at the time of the SMTP transaction and takes the form of a code and text message indicating the reason for non-delivery. An asynchronous bounce occurs after the SMTP transaction and takes the form of a “bounce e-mail” that’s received along with other inbound e-mails you get as a result of your mailing. This second data stream can represent a sizable portion of your bounces, particularly if mailing to corporate domains or smaller ISPs. Unfortunately, this is the data stream that’s often overlooked or incompletely captured by most bounce management systems.

Correctly interpret data. The right system will be able to process the incoming bounce data across the data streams and accurately interpret the myriad of inconsistent messages received from ISPs and other domains. This capability doesn’t come out of a recipe book: It’s the result of interrogating and comparing both the codes and text messages and applying confidence codes to different ISPs based on the reliability of the data received. To ensure correct interpretation, the system should accommodate continual testing and tuning.

Organize (normalize) data. No one can cope with hundreds of different bounces messages, so the next step is to normalize the data and organize it into logical categories, such as hard (permanent) bounce, soft (transient) bounce, block, and technical failure. A good system will map these delivery failures into subcategories, such as “unknown user” under hard bounce. Identifying the reasons for failure is crucial to your ability to diagnose the underlying causes and take the corrective actions that will improve your bottom-line results.

Make data actionable. Once organized, the system should generate reports that make the data actionable in addressing the causes of failure. The reporting should be user-friendly and available near real-time so you can easily and quickly take action to hygiene your list, adjust your targeting, and modify other practices to improve the deliverability of your e-mail. Ideally, the reports will provide definitions of the bounce categories and subcategories with prompts on the likely causes and corrective actions you should consider. Your reporting system should also feature intuitive drilldowns that allow you to use it as a diagnostic tool in examining failures by the categories or the domains that are important to you. You should then be able to quickly cross-reference your bounce data with other key metrics on the same campaign – open, click, unsubscribe and complaint rates. Finally, your system should enable you to pull up sample records to validate the data mapping and the conclusions you’ve drawn from it.

Keep data current. Even the best conceived bounce management system will degrade rapidly in our ever-changing e-mail environment. So this last criteria is an important one – it must contain a “future proofing” feature to stay abreast of changing ISP bounce codes and messages and to preserve the system’s data integrity.

With a bounce management system that meets these requirements, you’ll be in a position to properly evaluate your performance, manage your list and improve your practices – all of which translate into better bottom-line results. And you’ll be well on your way toward mastering e-mail deliverability.

Dave Lewis is vice president of market and product strategy for StrongMail Systems.