A False Dichotomy
We know not all address collection processes are equally effective or equally error-prone, so why are we treating single and confirmed opt-in as the only options?
Shortly before Christmas, Spamhaus added several national retailers to the Spamhaus Block List (SBL), creating a major controversy. The cause was mailing to unconfirmed email addresses and the action reignited the long-running debate about single opt-in vs. confirmed op-tin. Battle lines have been drawn with luminaries on both sides of the argument. It started out on mailing lists and Twitter, but has expanded to blog posts with back-and-forth, pro and con, argument, rebuttal, and counter argument.
While there are many great points being made, I think the single vs. confirmed opt-in choice is a false dichotomy. It doesn’t have to be one or the other and you’re either a spammer or a Spamhaus suck-up depending on which side you sit.
It’s widely agreed that marketers’ goal is to get as much email to the right people as easily as possible and the minimum to the wrong people. Unfortunately there is no entirely foolproof way to do this. Even though confirmed opt-in (aka double opt-in or COI) helps to minimize erroneous mailing, it does not entirely prevent it and it also creates friction for new subscribers, causing a substantial drop-off in subscription rates.
Emotionally where people fail, I believe, depends primarily on how they weigh the delivery of unwanted email against the inability to deliver email people asked for. Those who see the former as paramount lean more heavily toward COI. Those who consider the latter to be an essential business imperative lean more toward single opt-in (SOI).
But this isn’t a one-size-fits-all situation. We know not all address collection processes are equally effective or equally error-prone. So why are we treating single and confirmed opt-in as the only options?
Most email platforms have a tri-state for permission: subscribed, unsubscribed, and unconfirmed. Traditional SOI takes every new address and places it immediately in subscribed state until a behavior moves it to unsubscribed. COI, on the other hand, starts in unconfirmed and never moves to subscribed until a specific behavior moves it there. Furthermore, most systems prevent any mailing to addresses that are not subscribed.
So instead of completely trusting or distrusting any new address, let’s add a fourth state, call it “provisional.” New subscribers go into this state when they’re added and start to receive welcome communications. Subscribers who exhibit any negative behaviors move into unsubscribed. Those who exhibit positive move into subscribed and those who show no response at all are sent a COI-style opt-in confirmation request and are moved to unconfirmed.
What’s the point of all this? Rather than presenting a simple black-and-white barrier to entry – “You must click on this link in this email or you’ll get nothing further” – or presuming that every new address is equally valid – “If you don’t like it or it wasn’t for you or we got your address wrong, then unsubscribe” – we’re recognizing that our address collection processes are not perfect, and are treating new addresses as having provisional permission until we have corroborating evidence one way or the other. It’s similar to what many companies are already doing for long-term inactive subscribers. This is the same principle applied during the collection process. So even without specific platform support, this approach is achievable with today’s technology platforms.
The benefit of course is that a person who is receiving messages in error but who takes no action will not end up deluged with email forever. She’ll receive a few messages and then nothing more. The same applies to a spam trap address and most spam trap operators offer some leeway. It’s the continued deluge that triggers block listings.
If what you’re doing today is working for you and your subscribers and you’re not having list hygiene or deliverability issues, then stick with it. But if you’re using SOI and experiencing hygiene or delivery problems and you’ve done all you can to improve your data collection processes, then rather than sweating over a move to COI, or railing against the unfair ISPs or block list operators, perhaps you should consider a middle ground.
Until next time.
Image on home page via Shutterstock.
Editor’s note: This column was originally published in May, 2013. We hope you find it relevant as we head into the holiday shopping season.
