Do It Yourself A-B Testing

I’ve decided to really step up my tracking and testing game with my Sception landing pages. For projects in the past, it’s always been an afterthought, and I wanted to start this off the right way. There are many different ways to A/B test your landing pages. Some free. Some not. I chose a free method. Once Sception creates revenue, I may decide to use one of the great, paid methods, like Optimizely for example, but for now this is how I’m testing my landing pages.

As of this writing I have 4 separate versions of the landing page. The original. The original’s variation. One slightly more designer focused. And my favorite one, the designers variation.

If you look at them all, you’ll notice that they’re all very similar with just a few different things changed on each.

Different headlines. Added an extra email signup form. Changed some of the copy for the designer focused pages.

You don’t want to change a ton of different items on each, because you won’t really be able to tell which of those changes got you better results.

In addition to that, keeping the changes to a minimum makes it super easy for you to quickly create each landing page. You’re not reinventing the wheel each time. You just copy and paste, make the small changes, save, and upload.

After you’ve made your separate pages head over to Google Analytics to set up your goals. At the time of this writing, the “Goals” section is on the bottom left menu bar AFTER clicking on “Conversions”. Set up a new goal if you don’t have one already.

My goal right now is to have interested people sign up for my email list for Sception updates.

What I did was tell Google Analytics to track when a user arrives at my “thank you” page. I use Mailchimp for my email lists, and have it set so that a user is redirected to a custom “thank you” page after entering their email on any Sception landing page. Since I’m not advertising the direct link to that page anywhere, it’s safe to say that anyone who does end up there, did so because they signed up for my email updates. When Google tracks that a visit has been made there, it means that my goal has been met successfully.

Once you have your goals squared away, you can move on to the “Experiments” area of Google Analytics. At the time of this writing “Experiments” can be found AFTER clicking on the link that says, “Behavior”. Click on “create new experiment” and you should be able to figure out how to set it up with their instructions. Basically you name each experiment so that you can better tell them apart from each other. You can choose how much of your traffic is diverted to your variation pages. You specify which page is your original and which is the variation page that users will be diverted to. You pick what goal you want the experiment to achieve. (This is the goal you set in the previous step.) Then they’ll give you some code to place at the head of each ORIGINAL page. And that’s really it! They take care of the rest for you. I think they let each experiment run for up to three weeks. During this time you can monitor which page(s) are giving you the best results.

I’ve stopped my experiments for now because I have gone in a completely different direction with the look of Sception. I’m going to put together some new versions here shortly, but I was very surprised to see that my first, original landing page gave me the best results.