E-mailers Flunk the Unsubscribe Test

Do you think your e-mail program is a big success because less than 1 percent opt out every time you send? It’s time to pick a better metric. Your unsubscribe rate could be low because your automated unsubscribe function isn’t working.

How long has it been since you tested to see if the unsubscribe works? Probably never, if you’re like a lot of your fellow marketers.

According to my company’s new survey of unsubscribe policies, practices, and attitudes among commercial e-mail senders, almost half of the marketers who responded said they never test their unsubscribe function once they set it up. One in four tests it only after getting a complaint that it doesn’t work.

Given that 60 percent of marketers rely on their list software to manage the process automatically, this means lots of marketers are skating on the thin ice of noncompliance with CAN-SPAM by not unsubscribing addresses on demand. If you’re one of them, know that government regulations of commercial e-mail in the United States, the European Union, and elsewhere in the world don’t allow do-overs because of automation failure.

The unsubscribe process, then, has to serve multiple needs. It has to be legally sound, trusted by recipients, and deliver value to the company. We used the survey to measure how well marketers are meeting all these challenges by assessing their practices, attitudes, and experiences.

Because of the legal issues involved, your unsubscribe function has to be just as reliable, if not more so, than your opt-in. The law doesn’t care if your subscribe function doesn’t work, but you can be held liable for spamming if the unsubscribe fails repeatedly.

That’s one thing that separates e-mail from all other marketing channels. In the early days, marketers did everything they could to keep addresses on their lists. That, of course, spawned some of the spam problems we’re still dealing with, along with the onus on marketers to make sure those addresses are gone.

So, there’s a lot riding on your unsubscribe. What we found in the survey, among other surprises, is many marketers are leaving the job up to their automated software without going back under the hood from time to time to make sure it’s working.

Most Marketers Rely on E-mail List Software to Encode Unsubscribes

What do you do to make sure all e-mail contains an unsubscribe mechanism?	Response (%)
Encoded into e-mail template	41.11
Manual inspection	31.88
Rely on e-mail software to apply an unsubscribe footer automatically	18.82
Provide multiple options	5.57
None	1.39
List-unsubscribe header	0.17
Flashback testing	0.17
Other (please specify)	0.87
Source: EmailLabs, 2007

This isn’t a negative. In fact, it’s better to automate as many of your list functions as possible to free up your marketing or IT staff and to manage huge lists with thousands or millions of names. We were pleased to see a majority of marketers build the unsubscribe message into their templates so they don’t have to add it manually each time.

However…

One-Third of Marketers Rarely or Never Test Their Unsubscribe Functions

How often do you test your unsubscribe process to make sure it works?	Response (%)
Periodically on a regular testing schedule	27.32
Only after receiving a complaint that the unsubscribe process is not working	22.31
Each time before sending a message	21.80
Rarely or never tested	18.80
Once, during software setup	9.77
Source: EmailLabs, 2007

We were glad to see that just about half of the marketers surveyed said they test their unsubscribe systematically, whether on a regular testing schedule or each time before they send a message.

Still, almost a third of marketers said they haven’t tested their unsubscribe since they launched the software, or rarely or never test it. This is a disturbing finding, given that 60 percent of marketers told us they rely on automated functions to manage their lists.

This finding directly affects your reputation as a sender. If your unsubscribe function fails to remove an address within the time limit your government e-mail regulations dictate — 10 days in the U.S., for example — you can be liable for spamming with the next e-mail you send.

Also, when you keep sending e-mail after subscribers tell you to stop, they’re more likely to click the spam button in their e-mail clients, which generates spam complaints and can end up causing your e-mail to be blocked or shunted to the spam folder.

Take time, at least once in your publication cycle, to test all of your functions, including subscribe and unsubscribe, to make sure not only that the link works but that these functions properly remove test addresses. Test every link in each e-mail message to make sure it leads to the correct page. If the link yields a 404 error (define) or if it leads to a home page or landing page other than the one you expected to find, it’s time to recode.

Marketers Can Get More Out of Goodbye

The picture that emerged from our survey, which we conducted in fall 2007 with responses from more than 400 e-mail marketers, shows that while marketers generally comply with CAN-SPAM requirements, they’re failing to tap the rich veins of information the unsubscribe process can yield.

The unsubscribe is no longer a purely technical function. These days, it can be a key part of your marketing program and even be part of your customer-retention efforts.

Until next time, keep on deliverin’!

Want more e-mail marketing information? ClickZ E-Mail Reference is an archive of all our e-mail columns, organized by topic.