CHALLENGE
As of two years ago, incentives such as free personal finance reports had stopped increasing The Motley Fool’s membership numbers or improving subscription sales. A large percentage of prospects went halfway to becoming members but bailed during the confirmation email process.

“We weren’t happy that our opt-in rate had plateaued at 60%,” says Greg Martz, Director Email Management for The Motley Fool. “It didn’t satisfy us that we were still unable to market to the 40 out of 100 other possible people. We thought that we at least should investigate the idea of lowering the threshold of becoming a registered member.”

So, the members of Martz’s team got into an intense debate over what effect switching from double opt-in to single opt-in might have. It was a classic reward-vs-risk scenario. After all, according to new data from MarketingSherpa’s Email Marketing Benchmark Guide 2007, only 37% of B-to-C firms use single opt-in as their predominant strategy.

Indeed, Martz was sensitive to best-practices standards and wary of harming the company’s reputation, not to mention creating a huge list that would be soft on ROI. “The decision was tough. It just wasn’t, ‘Let’s A/B it and see what happens.’ And, of course, we wanted to be certain that we were not going to be filling our list with crap.”


CAMPAIGN
Martz and his team set out to discover whether the positives could outperform the negatives when it came to a single opt-in vs double opt-in strategy. To alleviate concerns with list hygiene, they also added better tools to help sort the good emails from the bad ones.

Step #1. A/B test

While philosophy was always near the heart of the debate, they decided that an A/B split test would be the logical place to start. Martz sent a free report as an incentive to two separate files of 35,000 recipients.

Shortly after the initial message, those in the double opt-in file received a confirmation email on which they had to follow up to become a member. The single opt-ins received a simple welcome message before getting a subscription offer 30 days later. Both files were offered the same chance to bow out along the way.

“We needed proof that we could improve business significantly by moving to single opt-in,” Martz says. “[But] at the same time, we wouldn’t want to go down a route where the performance of our advertising partners would decline.”

Step #2. List hygiene

The bigger the pool you build for your audience, the harder it is to keep it clean. And, since some of their email advertising partners were investing six figures a month on the medium, Martz’s team couldn’t afford a glut of bad addresses.

“Of course, it was very predictable that we are not only going to get a higher number of good email addresses -- but also bad ones with single opt-in,” he says. “That’s just something you have to deal with along the way.”

Recognizing the pitfalls of bad list hygiene, they implemented a more sophisticated system to clean invalid addresses, such as chronically bouncing addresses, malformatted addresses, expired addresses, “domain no longer exists” and misconfigured name-server situations.

Step #3. Feedback loop

Another measure Martz took to keep the database clean involved establishing feedback loops with the ISPs/ESPs that offered them. One of the benefits involved getting complaint-rate reports, which showed recipients who were selecting their “This Is Spam” button.

The feedback loops were achieved by filling out a form that required information such as servers used, IP block, types of emails sent, description of the purge process, handling of bounces, etc. (The forms can typically be found at the Help or Postmaster sections for most ISPs.)


RESULTS
OK, at this point, you’ve probably guessed that Martz and his team found single opt-in to their liking. But what’s surprising is the fact that the productivity of the strategy easily turned a potential no-no into a no-brainer.

“Looking back two years now, we wouldn’t have done it any other way,” Martz says. “The A/B test looked very good. We felt confident about the change, and we haven’t looked back. It’s worked wonders for our list growth, and that’s moved out to [sales].”

Since they knew the reports weren’t going to be perfect, Martz’s team stayed focused on revenue generated more than on other DM-centric metrics. Also, in comparison with double opt-in campaigns of the past, paying subscribers are up 45%.

“Conversion rates have gone down since then, but overall sales are up, and that’s the bottom line,” he says. “Cash receipts are everything for a subscription business.”

Meanwhile, the house file has grown by 130% since December 2004.

In addition, the new hygiene system is working wonders in keeping their email program running smoothly. Complaints can be tied back to the source of the address, and the system helps them identify problematic campaigns, containing report information that correlates addresses and bounce/success rates to specific mailings.

“We now have much improved visibility into our logs in terms of finding out who the invalids are and how to get them off the list as quickly as possible,” Martz says. “We are often getting rid of bad addresses within 12 hours.”

Kerrie Sweet, Head Techie at Motley Fool, offers the following nuggets of advice after watching the new hygiene tactics take hold for seven months:

- By using whatever methods or technology at your disposal, identify badly formatted addresses before adding them to your lists.
- Make note of domain name expirations (e.g., home.com) and scrape the addresses associated with them.
- Since domains change or morph (i.e., aol.com - aim.com), it’s tempting to simply switch the domains existing in such addresses in an attempt to improve delivery. “This is not recommended because the recipients didn't consent to that action,” Sweet says.
- Complaint-rate reports are valuable in terms of judging where you may stand with the providers. By participating, you encourage ISPs to whitelist your company. In other words, sign up for those feedback loops.
- Make sure you have a way to process asynchronous bounces (ones that don't happen immediately… they are initially accepted and eventually passed back to your servers).
- Keep a close eye on blacklists before you undeservedly show up on one.


Useful links related to this article

Creative samples from Motley Fool’s test:
http://www.marketingsherpa.com/cs/motleyfool/study.html

Previous Sherpa Case Study: How Motley Fool Doubled Email Registrations With Multivariable Tests (Includes Before-and-After):
http://www.marketingsherpa.com/barrier.cfm?contentID=29
3

StrongMail - the email technology platform The Motley Fool uses for email delivery and management:
http://www.strongmail.com

The Motley Fool:
http://www.fool.com