The biggest buzzword in e-mail marketing today seems to be “relevance.” I’ve always argued that relevance and timeliness are key to successful e-mail campaigns, but achieving them has long been a significant challenge.
In recent years, e-mail marketing systems have become more sophisticated, allowing marketers to easily deliver timely, relevant messages through trigger, cadence, and personalization capabilities. So much so that “it’s no longer about batch and blast” has become a mantra in some circles and the boundary between transactional and marketing messaging is blurring.
Unfortunately, in many cases such messaging growth seems to have outstripped quality-control developments. The growth, then, has been accompanied by a commensurate increase in errors. Those errors also affect some of the more complex, marketing-oriented transactional messages.
Sophisticated campaigns run into new quality-assurance challenges on these three fronts.
In most organizations, the quality-assurance process for messaging was developed for one-time mailings. So quality assurance is typically a one-time process performed immediately prior to sending. This is perfectly fine for traditional campaigns. Little is likely to change after the send, and as long as servers and services are appropriately monitored, there’s no problem.
Triggered messages, on the other hand, may continue for months or even years. As such, any external resource is subject to change. A hosting change or site update could remove graphics required by a still-operational triggered message. Any link destination or externally referenced content could be altered or removed at any time and without notice.
Deliverability can also change over time. The smaller numbers of individual recipients can make delivery easier, often reducing the likelihood of blocklisting. But monitoring is harder. It’s much more difficult to obtain representative samples, and clusters of complaints can more easily skew results. Adding seed addresses for delivery monitoring can have a disproportionate effect on overall outcome. These challenges require close scrutiny from deliverability-monitoring personnel.
Triggered, cadenced, and dynamic e-mail campaigns are inherently more complex than their traditional counterparts. I’ve written in the past about the testing challenges presented by dynamic messages. The recipient-selection business rules for triggered and cadenced campaigns present similar issues.
Recipient profiles are more complex; selection criteria are similarly more complex and may have both behavioral and temporal components. This complexity requires a more sophisticated approach to testing, with the use of carefully selected test cases and test coverage analysis rather than the traditional viewing of every message variant during a test phase.
Sanity checks, such as recipient limits, can be added to business rules to mitigate the problem. However, even when carefully constructed these can be both a blessing and a curse. On the one hand, they can prevent an accidental mailing to the wrong group in the event of a bad data load or erroneous query. On the other, a false positive can prevent an important message going out when there’s an unexpectedly large, yet correct, number of recipients.
Errors Go Unnoticed
This issue compounds the first two. Due to these communications’ automated nature, it’s not uncommon for errors to go unnoticed. Often organizations find out that messages are broken only when recipients tell them, which, of course, is far too late.
Constant, structured monitoring is essential to resolve the situation. Triggered and cadenced campaigns can’t be treated as “fire and forget.” The net result is that triggered, cadenced, and dynamic campaigns put more load on the quality-assurance team, not less, and it’s important to plan and budget accordingly.
Recipient counts, response rates, and delivery numbers should be monitored on an ongoing basis. Each triggered campaign should have a regular quality-assurance review on a fixed schedule to ensure that everything is still functioning today as it was when the campaign was initiated.
As we have seen over and again, sending the messages is the easy part of e-mail marketing. Getting the right message to the right person at the right time, every time, is the hard part. Without these improvements to our quality-assurance procedures, our ability to generate sophisticated marketing campaigns will continue to outstrip our ability to successfully and accurately implement those campaigns.
Want more e-mail marketing information? ClickZ E-Mail Reference is an archive of all our e-mail columns, organized by topic.