I recently listened in to a Yahoo workshop with the company’s anti-spam czar, Mark Risher. During the event, Mark advised commercial e-mail senders to periodically reconfirm their lists. As I looked back over past columns, I realized that I wrote about list reconfirmation in my second ClickZ column back in 2004 in the context of list cleanup but haven’t mentioned it again.
While most, if not all, of what I wrote four years ago still holds true, much has since changed in e-mail marketing. In particular, ISPs and marketers are focusing much more heavily on list hygiene. This makes reconfirmation a more important technique for the everyday rather than just an emergency tool for when there’s trouble. Don’t get me wrong. Reconfirmation is an essential technique for list cleanup when there’s trouble. High complaint rates, high bounce rates, spam-trap hits, uncertain list provenance, periods of inactivity, all point to a need for reconfirmation to clean a bad list. However it can also be used day to day as part of the routine maintenance of a list.
Regular, or routine, reconfirmation can sound quite frightening to marketers. It often conjures images of valued and valuable list members leaving because they didn’t recognize they needed to click on a link or put it off and never came back to it. If done correctly, though, that should not be the reality.
Many marketers are embracing the concept of a list exit strategy. That is, developing and managing a process to identify and gracefully remove recipients for whom a list is no longer valuable. Many lists have a long tail of recipients who have been on the list for a very long time but who also have not responded for a very long time. Apart from unconfirmed new registrants, these recipients are the most likely to bounce, complain, or to contain spam traps. Just as confirmation should be used at the beginning of list membership reconfirmation can be used at the end of the lifecycle to separate those recipients who wish to remain from those who have passed their “best before date.”
There are, of course, important rules to getting this right.
Select the recipients. To maintain an effective reconfirmation exit strategy, it’s vital to know when recipients last responded. This should include all potential activity, clicks, opens, profile maintenance, and non-e-mail activity such as purchase behavior. While I would not recommend stopping reconfirmation due to a lack of e-mail activity it may well be worth factoring in to a weighting system.
For example your strategy may be to put recipients into the reconfirmation phase when they have not purchased (off-line) for three months, not opened for six months, and not clicked for a year. This would mean that someone who has not responded for a year but who purchases will have another three months before entering the reconfirmation process, just in case they have renewed interest.
One might wonder why the purchase wouldn’t have a longer time frame. The reason is that since it’s an offline activity it tells us nothing about the current state of the e-mail address even though it tells us a lot about the owner of that address.
Keep options open. This is not an emergency reconfirmation where it is essential to remove problematic addresses immediately. It is a mechanism to gracefully retire addresses from the list once the owners are no longer interested. To this end offer both an opt-in and an opt-out button, a “Yes, I want to remain on the list” and a “No, I’m done”. This gives the option to send a second message to non-respondents before retiring a recipient’s address.
Ensure the creative is clear, simple, and has a strong call to action. If you’re running two messages it’s possible to have a relatively soft sell on the first and a much stronger one on the last chance message.
While there’s no absolute requirement to have a list exit strategy or to perform routing reconfirmation many marketers are adopting such strategies to improve their list hygiene, performance, and ROI. Furthermore, if the rumors that ISPs are starting to look at open and click rates for messages are true that requirement may yet come into existence.
Until next time,