How to A/B Test in Fundraising Emails
Tuesday, March 12, 2019
According to the M+R Benchmarks 2018 Study, email accounted for 28% of online revenue in 2017. Email still matters—but it’s increasingly important to craft those messages strategically. The same study reported that response rate to fundraising emails declined by 6% in 2017.
So how can you start enhancing your fundraising emails? Try A/B testing.
A/B TESTING: A DEFINITION
A/B testing compares two versions of the same variable to determine which is more effective at a stated aim. All other variables are held constant. Here’s an example:
ABC School is planning to send out a fundraising email, and they want to know if changing the call to action (CTA) language will affect the number of gifts received.
They create a fundraising email with a CTA button that reads “Donate Now.” Next, they create an identical email with a CTA button that reads “Support our Students.” Half the School’s subscriber list receives Version A, half receives Version B. Afterwards, the development team compares the number of gifts received from each email.
THE VALUE OF A/B TESTING
Even with best practices at my fingertips, I can’t tell you how your unique audience will respond to a specific email. You’ve carefully built up your email subscriber list. It’s your direct connection to people who have opted in to engage with your organization.
A/B testing gives you the chance to test and re-test small changes, see how that curated list of supporters responds to your emails, and change course for better results.
WHAT VARIABLES TO TEST
Wait! Before you jump to variables, think about your goal. You likely have multiple goals but pick one to focus on. Once you decide this, it will be easier to choose the variable to test. Some common choices include:
Improving open rate:
- Subject line
- Sender name
- Day of week
- Time of day
Increasing click-to-open rate, number of gifts, or average gift size:
- CTA color
- CTA copy
- Location of CTA in email
- Messaging (in body of email)
HOW TO RUN A TEST
First, steer clear of A/B testing during your most important fundraising time periods. It’s best to avoid experimenting during these critical stretches.
Many email marketing services offer a built-in A/B testing feature. If yours doesn’t, you can run an A/B test manually. Follow these steps:
- Create the Version A email
- Change one variable to create Version B
- Randomly select a segment of your email list—this segment should be at least 1,000 subscribers
- Send half Version A and half Version B
- After you’ve compared the results (waiting at least 4 hours), send the “winning” version to the rest of your email list
If the email list in question has fewer than 1,000 subscribers, you can still perform the A/B test without segmentation.
Keep testing: If you think you’ve found something that makes an impact, try it a few more times. If you achieve similar results, incorporate that change into your future fundraising emails. A/B testing can and should be an ongoing process, helping you stay nimble, pivot quickly—and ultimately raise more online.