Why you should love A/B split testing

Website Experimenting Part 1

I’ll be honest. I’ve had a love/hate relationship with A/B split testing. The thing is, I’ve worked closely with customer support teams for close to 10 years now. Our greatest fear? It wasn’t upset customers or broken computers or long conversations with your crazy uncle Irwin, who managed to call in and find you and talk about his latest creative gig playing banjo for a jam band in Costa Rica.

Our greatest fear? Looking stupid. And sometimes A/B split testing can make you look stupid. A customer calls in and says, “I see a giant orange button on your home page,” and you say, “Uh, I don’t see that at all.”

See? Stupid.

But, it’s totally worth the risk. It’s 100 percent worth it. Simple A/B split testing is one of the easiest ways to figure out what’s working and what might work better on your website.

Like anything on the Internet, split testing can be really, really complicated and involve a lot of math or it can be really simple—especially if you’re just getting started. For now, we’re gonna keep things simple.

What is A/B split testing?

Here’s how it works: you have two versions of your website live at the same time. Half of your visitors see your normal website and half of them go to your new one (or 30/70 or 60/40 or 90/10). After a little while, you compare the results between the two versions and see if the changes you made to your new site get you more visitors, leads, customers, whatever is important to you.

It can be a really subtle change: changing the color of the hyperlinks on your home page. Or something a little more bold: changing your call-to-action from a big round orange button to a big square green button. Either way, you make one or two changes at a time and test the results.

Why should you do it?

Why should you do it? Simple. Because people are unpredictable. If you’re like me, I think I’m pretty smart when it comes to words and design and marketing stuff. I know what they want! They just don’t know what they want! But, it doesn’t always work that way.

For example, you might think that the easiest way for your customers to contact you is a phone call. That’s the call to action on your home page. It’s gigantic. It’s bold text nested in a bright yellow box. You’re getting hundreds of visitors to your website, sometimes thousands, but **no** phone calls. What gives?

A/B split testing is a simple way for you to experiment and see if something different will help. Maybe if you change the color of the box, you’ll get phone calls? Maybe if you use a Web form for people to contact you? Maybe they need to learn a little more about you before they give you a call? Maybe the call to action leads them to examples of the cool stuff you’ve been doing?

A/B Testing

Split testing is simple way to try out something new, measure the results, and get a good idea of what works and what doesn’t work for people who visit your website.

How do you do it?

Split testing requires you to make some technical modifications to your website and how it shows up for customers. In general, the process looks something like this:

  1. Make sure you have a way to measure interactions on your website (I use Google Analytics™).
  2. Design a new Web page with the new element you want to test.
  3. Save it with a new, unique name on your Web server.
  4. Program your site to show one page to some visitors and another page to other visitors.
  5. Measure your results and update your pages as necessary.

Now, in this article, I’m not going to go into a lot of technical details about how to set up Google Analytics, develop a new Web page, or pull this off with JavaScript or DNS goodness. There are a number of ways to do it and it depends on what you (or your developer) used to make your website, how you’re measuring traffic and visitors, and whether or not you decide to pay for an A/B split testing tool or put something together using your own code or something for free.

I will, however, point you in the right direction—at least to get you started.

What’s next?

If you’re itching to get started, first make sure you have a Google Analytics account hooked up and working. After that, read through this guide, open up your analytics account, and start your first experiment. Make sure you have an alternative home page already set up. You’ll need that when you create your first A/B split test, or experiment, in Google Analytics. Again, start with something simple:

  • Change the color of your call to action.
  • Try a different call to action (call me, contact me, sign up now).
  • Switch up your headline.
  • Swap out the photo of your business with a photo of yourself.

If you’re still a little freaked out about the technical work you need to do to make this happen, don’t worry. This article is the first part of three-part series. In part two, I’ll walk you through an A/B split test using Google Analytics Experiments and a simple WordPress® website. In part three, we’ll talk about some of the other tools out there that make split testing a snap. Until then, think about what you might want to test. It’s gonna be fun.

Shawn Pfunder
Shawn's been working with freelancers, entrepreneurs, and business owners for more than 20 years. He's consulted companies large and small on communication, social media, and marketing strategies. At heart he's a small-business superfan. He believes that working for yourself is one of the most courageous and creative moves anyone can make. Currently, he's the Editor in Chief for The Garage. When he's not hanging out with solopreneurs, Shawn loves to write, run, and travel. He's passionate about teaching and he's convinced that a good story is the best way to do it — especially if it involves El Caminos, potato fields, and really loud music.