How organizations build email lists is no small issue. Many groups are investing significant resources in staff, consulting, advertising, events and vendor contracts (in particular partnering with Care2, Change.org and similar communities) to increase their list size.
An issue that often comes up as these programs take off is do we use “opt-in” or “opt-out?” What does this mean? Opt-in means that a new subscriber must make a proactive decision to join an email list by clicking a checkbox, filling out a form or, in the case of double opt-in by replying to a confirmation email and essentially telling you twice that they want to be on your email list. Opt-out happens when someone is added to an email list without clear prior acknowledgment and must actively opt-out if they want off the list.
We were intrigued the other day to come across a post on ClickZ looking at the results of over 300 million emails sent to sets of subscribers added to lists via opt-in and opt-out methods. Email professionals will generally discourage organizations from using opt-out methods (though, as discussed below) typical subscription practices aren’t far from opt-out and many organizations use opt-out frequently through email appends. And opt-out is pretty much assumed in political campaign marketing where lists are bought, sold, traded and given away all the time.
Opt-out is cheaper than opt-in. You may put money up-front to rent/buy an email list or run an email append against your mailing list or other house file. But dollar for dollar it costs less than opt-in. Generally, the more subscribers need to do to indicate their interest in subscribing the less likely they are to subscribe.
The payoff is that someone going through the opt-in process is more qualified and likely a better advocacy, engagement and fundraising prospect than someone added to a list without express consent. But if opt-in costs more then is it worth it? Is the ROI of poor-performing opt-out lists still as good or better in the long run as opt-in? And what about assumed opt-in?
In our experience, the most common route taken by organizations is something like “assumed opt-in.” There might be a checkbox below a petition or other advocacy ask (most list growth doesn’t come through straightfoward subscribe forms). The checkbox is pre-checked and has a note next to it saying something like “I want to receive emails from the organization.”
Finding the sweet spot in list building practices is critical for any organization. We were excited to see this ClickZ article looking at the performance of different lists receiving several hundred million messages over more than 18 months in 2010 and 2011. We were, however, a little disappointed that the study didn’t tackle the ROI question. So let’s take a look at some findings of the study:
- Open rates of opt-in messages were nearly double opt-out open rates.
- Click through rates of opt-in messages were two or more times higher than opt-out messages.
- Click to open rates were higher for opt-in messages though the difference narrowed over the course of the study. Click to open measures the click through rate of only messages that were opened.
As mentioned, the study didn’t look at ROI. The data available made it too challenging to get sales data. That’s too bad but it should be possible for most nonprofits.
It’s important that organizations track not just open and click through rates but also key ROI numbers: amount given, actions taken and more. I don’t care about your high open rate if its not driving donations and a low click through rate could be okay if the ask sets a high bar for people. Context and goals matter. The key is collecting the right data and analyzing it in context. Many organizations today use integrated email and fundraising platforms from Salsa Labs, Convio, Blackbaud and others that will, if configured and managed correctly, let you measure end to end ROI.
What do we make of this study? It provides some solid data showing that opt-in subscribers are more likely to open and click through. It doesn’t say too much about whether they act more frequently. In this case the ask was probably an online purchase.
The study also doesn’t address the assumed opt-in question. We encourage organizations to test opt-in methods and measure their ROI over time. Assumed opt-in isn’t a bad thing if you’re running a good welcome series and getting subscribers into the message flow quickly. But don’t go too far with the “assumed” portion of the subscribe process. Include a check box and not just a “you will be subscribed” statement.
What of opt-out? Stay away from it. There are conflicting views on this and one can make a case that opt-out ROI is good over time given the low cost. But ROI analytics rarely measure the cost of upsetting people along the way. What is the value of angering a subscriber added to your list without clear consent? In an ROI calculation that person is just one of many that doesn’t click through. But you’ve likely spoiled any hope for a relationship with that person and their ability to talk about you online through social networks and other channels is only growing.