Do Confirmed Opt-in’s Kill Your Profits?

Questioning Confirmed Opt-Ins

It is common wisdom that a confirmed opt-in (a.k.a. double opt-in) gives you “better qualified” prospects than a single opt-in.  An article by Daniel Levis explores whether this is indeed true.

Definitions

For those who may be new to internet marketing a confirmed opt–in is when a prospect enters her name and email information on a web site (called an opt-in) and then gets an email asking her to confirm her opting in. She must take the extra step of clicking on a confirmation link within the email before her opt-in is complete. Because she has to take this extra step before she can receive whatever benefit comes from the opt-in  (e.g. a free report, or a subscription to a newsletter, etc.), she is considered to be a stronger prospect.

Surely, this extra confirming step loses some prospects.  The confirmation email can get lost in a spam filter, or the prospect may simply forget or not bother with the extra step.  The critical question is, How many of these lost prospects would ultimately have bought something from the site?

An Experiment

In his article, Daniel Levis reports an experiment he did indicating that significant sales are indeed lost. He ran an ad campaign using Google Website Optimizer that tested two identical looking web pages (squeeze pages). One web page had a single opt-in, while the other web page used a confirmed (or double) opt-in.

Google’s Website Optimizer automatically sent half the traffic to the single opt-in and the other half to the confirmed opt-in for a period of about a year.

The Results

Here are the results.

Double vs. Single Opt-in Conversions

Double vs. Single Opt-in Conversions

The confirmed opt-in reduced the number of leads collected by 37% or 1842 leads lost during the year. A commenter to the article reported similar findings with an independent, but similar experiment.

Tune In Next Week

But does this really mean that your profits will be less with confirmed opt-ins?  Whether fewer leads means lower profits isn’t conclusively demonstrated by this experiment. However, Daniel Levis will report next week on another experiment aimed at measuring the actual click-thruogh and sales results for these two groups.

UPDATE: Follow-up Experiment Shows Surprising Turnaround

Dan Levis reported a further experiment where he looked at conversions (people who actually bought something) after sending out an offer to those who remained subscribed after a year.  The table below shows the results.

Double vs. Single Opt-in Profitability

Double vs. Single Opt-in Profitability

When Levis e-mailed the same offer to both lists, the double opt-in list yielded 41% more opens and 46% more clicks.   This suggests that the double opt-ins may be more committed prospects.  Though he did make $773 in sales with the double opt-ins, it appears that this was due to 1 sale (note that there’s only 1 conversion).   One sale is certainly not a statistically significant result.  So, we can’t really conclude anything about profitability from this experiment.   All we can say is that it looks like the double opt-ins, though smaller in number, might be better prospects.

This second set of results was a surprise.  Dan Levis  explains the results as a side effect of the delivery method of the single vs. double opt-in servers.

“Since single opt-in servers send a lot of undeliverable mail (bogus addresses and typos) and spam (subscriptions created by someone other than the recipient), the receiving end servers assign a lower reputation to these sending servers than those that send only deliverable, verified mail (double opt-in servers).”

In other words, many of the single opt-in email offers probably got filtered out as spam.

Do Confirmed Opt-in’s Kill Your Profits?

Nevertheless, Levis concludes that in most cases the double (confirmed) opt-in is preferrable to the single opt-in.  This is especially true if you’re interested in the long term value of each customer.  So, confirmed opt-ins probably don’t kill your profits, but a bigger experiment is needed to be sure.

Leave a Reply