An A/B split test case study: How to double your conversion rate

You’ve probably heard about A/B split testing. It’s where you have two (or more) variations of a web page and you send 50% to one page and 50% to the other. In most cases, you’ll see that one page converts better than the other(s).

Recently we helped a client improve the conversion rate of a landing page from an already healthy 11% to 24% by working with their designer to create a simple split test using Google’s free Website Optimizer tool.

Basically, we’ve been running and managing their Google AdWords account for the last few years, and it had pretty much matured in terms of optimising it any further to increase conversions.

The only place left to optimise was the performance of the landing page we were sending traffic to on the website itself.

Melbourne-based web designer Dilshan worked with us (and our shared client) to create two variations of the landing page.

Without giving away who the client is, I’ve tried to illustrate the difference between the landing pages with wireframes (probably should have asked Dilshan to help out!).

The first is the original which was basically quite a long sales page, with a form at the bottom.

The original:

The challenger design from Dilshan removed the navigation (risky, but in this case very effective!) and lifted the form above the fold in another column.

Variation 1 – the challenger:

The results after just a few days surprised everyone.

The challenger was a hands-down winner and a couple of people in our office lost a few dollars on the outcome.

I guess the really important thing to note about this experiment was that there were absolutely no changes to copy, headings, images, etc. The only content change was the removal of the Navigation. This was a simple case of changing the layout of already existing elements.

Normally we like to run some user testing to help formulate a split test, but this was quick ‘one-off’ and it worked brilliantly.

Obviously, the next step for us is to conduct a multivariate test on the winning design.

For our client, it means we can confidently bid more money (like, twice as much) on AdWords knowing that the traffic will convert highly. It also means they can also smash the competition, and confidently hold top AdWords positions at the same time.

And we haven’t even completed a full round of testing yet.

Sometimes it’s a lot easier (and cheaper) to improve website conversion rates than it is to spend more in AdWords (or SEO) to achieve exactly the same result.

Seriously, this is a great industry to be in.

For more Online Sales expert advice, click here.

Chris Thomas heads up Reseo, a search engine optimisation  company which specialises in creating and maintaining Google AdWords campaigns and Search Engine Optimisation campaigns for a range of corporate clients.