How do I A/B Test My Forms?

Written by Jeff on May 2, 2013

Posted in Form Optimization, Lead Generation

A/B testing is a fast growing trend in digital marketing, and for very good reasons. Put simply, A/B testing is a statistical method of testing certain variations in a group of subjects and measuring their effects on a given result. So for digital marketing, this can be testing one landing page against another (or 2, or 3, etc.), and then measuring their effectiveness in getting signups. Or another example is to test one email subject line against another (or 2, or 3, etc.) and then measuring which one gets a higher open rate and a better click-through. A/B testing is also the main component of conversion optimization, or the process of improving the rate by which website visitors perform a certain action, like a lead generation or a sale. In other words, A/B testing is a way of statistically measuring what marketing creative or website variation generates more leads or sales. Pretty cool, right?

What does this mean for you?
Well, luckily there are ways to perform A/B tests on your Formstack forms. And why would you want to A/B test your forms? Well, despite your insistence (or your boss’) that your forms are fine the way they are, A/B testing can almost always improve your form’s performance. And improving your form’s performance means more submissions from your visitors. That means more event registrations, more lead generations, and more SALES.

For testing forms, there are generally two approaches. The first is with a redirect – in which you redirect your form visitor’s browser to a new page that contains the new form you’d like to test. Most of the time these redirects are instantaneous as the page loads and hardly noticeable. Then, just simply split the traffic 50% to the new form variation, sit back, wait and watch which form version gets more submissions. NOTE: 100 submissions minimum is recommended for each test variation, so sitting back and waiting might take a while.

The second approach is to dynamically change a form’s appearance with an A/B testing tool. There are a large number of A/B testing programs out there that can do this. The way they work is by dropping in Javascript that can dynamically change a form’s appearance in the browser. Since you’re using the same form during testing and can’t tell which submissions came from what form variation, the A/B testing tool will track and measure these “conversions” for you.

New to A/B Testing? 
I recommend starting out with Google Analytics. It’s free, simple, and most people have it installed on their site already. It uses the redirect method, so you’ll need to create 2 versions of the form you’ll be testing, along with 2 pages to house the forms. I’ll be doing a blog post on exactly how to set one up in the near future, but in the meantime, go ahead and experiment with… uh, Content Experiments. If you have a high traffic volume to your site and are willing to pay for better, more advanced (but simple-to-use) testing platforms, I recommend Optimizely or Visual Website Optimizer.

For ideas on what to test on your forms, please refer to my post on Optimizely’s blog on 3 Essential Tips for Testing Online Forms. It should give you some simple, but effective places to start. Now get out there and get testing! Any questions? Let us know in the comments below!