How to Split-Test Your Email Opt-In Form in aWeber

One of the main benefits of using aWeber is it allows you to split-test different variables on your email opt-in form. This provides great feedback in terms of what is working – and what’s not – when it comes to converting your site visitors into subscribers.

Limit your split-testing to just one variable at a time. This is the best way to get accurate results.

With split-testing, you can test any number of variables – i.e. different templates, color schemes, form types (inline, lightbox, or popover), headlines or incentive offers – to see which ones work the best.

To run a split-test, first you’ll need to create 2 (or more) opt-in forms. (In case you missed it, here is the first post in this series on how to create an email opt-in form in aWeber). While creating these forms, make sure they are identical in every way – except for the one variable you want to test. (i.e. lightbox vs. inline form type).

Setting up a split-test in aWeber is easy:

First, log into your account and select the list you want to create the split test for – (if you have just one list, you can ignore this step). Then, click on “Web Forms:”

Assuming you’ve already created the different versions of your opt-in forms, proceed to “Create Split Test.”

Which will take you to this screen:

This screen may be hard to read on your computer (if so, try hitting “Ctrl” and “+” to zoom in). At any rate, this just shows that I selected 2 opt-in options – one lightbox and one inline – and set them both to display 50% of the time.  So, for every 100 page views, 50 will be of the lightbox option and 50 of the inline version.

Now I need to clarify – this is not an actual test – the sidebar option for this site is still set to display 100% of the time and the only place I have the lightbox set to display is on this page (after 15 seconds). I just created this screenshot to give you a visual of how to do it yourself.

However, here is a cool real life case study showing the results of an actual split-test.

I like this case study because it shows the effectiveness of different form types for a site’s homepage – I don’t know about you, but I found the results kinda surprising… mainly because I find lightboxes a little annoying.

Which goes to show that your own personal preference isn’t always the best measure of what actually works!

3 Things Worth Noting From This Case Study:

1. This was done in 2008. Perhaps preferences have changed since then? (I don’t know – further split-testing in order!)
2. The results were based on roughly 300 page views – or 100 each. Is that a large enough sample to yield statistically significant results? (Again, I don’t know – but do know that the larger sample size you use, the more accurate your results should be).
3. Consider the target audience for the test site. The preferences of that group may – or may not – be different than the one drawn to your niche or individual site. Yet another reason why testing on your own site with your own readership is so valuable!

Would love to hear your thoughts on this – or even better, if you would share some of your own split-test results. Please leave a comment below!

Til Next Time,

PS I am running some live split tests on other sites and will provide an update at a later point in time. If you’d like to get those (and other) blog updates delivered right to your in-box each week – you can sign up using the form below. (You will also receive my 5-part eCourse on the Basics of List Building when you subscribe).