Our Abila experts have researched and crafted Key Fundraising Predictions for 2015 to help you gain a clear understanding of the challenges facing your sector in the coming year, and how you can turn them into real advantages for your organization.
Below is a deep dive into Fundraising Prediction #1 – Conversion Rate Optimization (CRO).
And, because conversion optimization is so important to nonprofits’ success, we’re rolling out a series of blogs on this topic. In the first post, we defined CRO. Last week, we unveiled the 8 Pillars of CRO. This week, we delve into A/B Split Testing.
The basic premise of Conversion Rate Optimization (CRO) is to find out what works and what doesn’t. One of the best and simplest ways to do this is by using the A/B Split Test.
A Split Test is a process of testing two separate variants – an A version and a B version – against each other to determine which one produces better results. It’s basically an arm wrestle to find a winner.
For it to be a true A/B Split Test you need to send an equal amount of traffic to each version during the same time period. This usually requires software to not only manage traffic, but also decide on the frontrunner (more on this in a future post).
If you don’t have the resources to purchase or utilize Split Testing software, there’s also the option to run a “Before and After” Test. For this type of testing, you still have an A and a B version, but you only run one at a time. Use version A for a set amount of time or number of visitors, then measure conversions and actions. Then replace it with version B for the same amount of time/visitors, then measure conversions and actions. While this is not as “scientific” as a true Split Test, it will still give you great information you need to make informed decisions and improve conversion rates.
Here are the Nuts-n-Bolts of Running a Basic A/B Split Test
- Set a Goal – What do you want to increase? Email sign-ups, volunteer registrations, donations? Don’t spend too much time here, simply pick a goal you want to test and get going. Eventually you’ll test all of these things!
- Decide What to Test – Based on your goal, what can you change that might improve conversions? The headline, button color, text on the page? Remember to only change ONE thing at a time for each test (see our previous post on Pillars of CRO for details). Again, don’t worry about picking the perfect item to test … just pick something. You can, and should, come back and test the other elements at a later date.
- Set up Your Test – Create both versions of your test, the A and the B, and decide if you’re going to conduct a true A/B Split Test or a Before and After.
- Walk Away – Now for the hard part. You need to walk away and let the test run for a while. Without interference. This is not the time to jump in and change things if/when you notice one of the versions not doing well. Borrowing from The Beatles, Let It Be.
- Winner, Winner, Chicken Dinner – Once completed it’s now time to measure the results and determine if your test has a clear winner. In many cases this will be easy to determine, while other times it will be a near dead heat.
- Rinse and Repeat – You didn’t think we were done did you? Now that you have a winner you need to keep going. You can choose to run the winner against a new challenger, or move on and test something else entirely. The key is to keep testing and trying to improve your conversion rates over the long haul.