Split testing is something that most people know that they should be doing but for some reason, never do.
And it’s not just the one man band. We’re also talking about 6-,7-, and even 8- figure businesses that rely on a large extent on their website for business. Even though they know that split testing can 10x their revenues, it usually drops down to somewhere at the bottom of their to do list.
If you have substantial traffic with a goal in place (e.g. to increase opt ins, conversions, sales etc.), start split testing! It will bring results. Check our Kyle Rush’s presentation at MozCon 2013 (better if you watch the video which you can purchase from Moz.com) where he talks about how he increased donations to Obama’s campaign by over a hundred million dollars.
I run split tests on the home page of my website. It’s not really the most essential because I drive most of my paid traffic to dedicated landing pages (where I run a lot more experiments) which brings better results. But I do split test the home page also since I get some organic traffic coming in their through search engine optimization.
Call to action (CTA) buttons are an easy item to test and they can easily boost your conversion rate so it’s something that I generally recommend people to earlier in their testing rather than later. You should always go for the easy wins first including:
- Headlines (very important)
People can sign up for a trial of my video course. Using the new button increased the sign up rate for my video course by about 9.1% (after improvements in headlines, images, and so on). I tried a few buttons and the results of the A/B test demonstrated with significant probability that pretty much all of the nice image buttons beat the plain “Start Free Trial” that I have been using for some time which was to be expected.
The one that I settled with is shown below (“learn mandarin”) which in some opt in pages resulted in a 34.2% opt in rate (based on about 400 conversions at the time of writing).
You can purchase some a pack of high converting buttons on places like Graphic River for just $4 (affiliate link). An absolute bargain if you ask me.
The sign up rate on the home page is about 11.2% without filtering out existing members (so the actual rate would be higher) and for fairly wide ranging terms via Google.
When you’re looking for things to split test on your website, it’s a good idea to look at your heat map and see where people are clicking. If they tend to click on a certain image, make the image clickable. If they tend to spend more time in a certain area of the page, experiment by putting the opt in form there (if it’s not there already).
You can also experiment with size. In some cases, super-sizing the play button on a video can increase the play rate of a video by several folds.