Always Be Split-Testing

"Always Be Split-Testing" is the motto and creed of some online marketers. Surprisingly, not that many. Certainly, you can find more people looking for some magical "secret" to making tons of money with no effort than you can find marketers following the above creed, any day.

Then again, you can also find more marketers who aren't successful than those who are and in this case, that might not be a coincidence...

Split-Testing: What it is...

Split testing is also know as "A/B Testing" and it simply works like this:

You make two versions of a website and send half of your visitors to version 1, and half of them to version two. Then, you check to see which version performs better.

The performance of the pages is determined by what the conversion goal is. The conversion goal could be something like: getting the visitor to click on your affiliate link or getting the visitor to sign up to a newsletter or generating a sale.

Here's a more concrete example: Let's say you want to build your mailing list, so you create a squeeze page (simple page where you advertise some reason why they should sign up and add an opt-in form). On the squeeze page, you advertise a free guide you've created, which is the incentive for people to opt-in (they opt-in to get the free report). In this case, the conversion goal would be that the visitors sign up.

Now, you create a second version of that page, which is identical to the first one, except that you use a different image to show your free report. I'm sure you've seen these "e-cover" images for ebooks and free reports - they help make the digital product seem more tangible.

Now, you split up all the traffic you send between those two pages. Soon, you'll see that one of them (probably) outperforms the other one, because you'll see that you're getting more people signing up from one of the two pages.

...How to do it...

The good news is that it's not difficult or expensive to set up an A/B test like the one described above. You can use Google's Website Optimizer completely for free and it's a pretty powerful split-testing tool. There are also lot's of instructions you ca find online, on how to set up and use Website Optimizer.

This, as well as any other split testing script, will do the "splitting" of the traffic and the conversion tracking for you. This means, you can send all of the traffic to one address and they'll automatically be split up between the two variations you set up. And the script will also automatically measure which variation converts better. And of course, you can also use more than one variation of a page in a split-test.

Whatever you do, it is advisable to take things slowly. Even if you want to test all kinds of different factors on your page, do them one-by-one. I.e. test two different images against each other, then keep the better one. Then, test two headlines against each other and keep the better one, then do the same with your bullet points etc.

Later on, you can get into multivariate split-testing, if you want to get really serious about all this. To begin with, simple A/B testing is the recommended way to go, though.

...and Why it's Amazing!

Split testing is absolutely amazing. It's one of the things that makes marketing and selling products online so much easier and more effective than selling them offline. Split testing a web-page is like changing the storefront for every other visitor, to see what happens. Unthinkable offline - free, online!

What split-testing does for you is increase your conversion rates and that is one of the best things you can do for your business. If you can increase your conversion rates, it means you make more money without having to drive more traffic, without having to spend more on advertising, without having to create more products, etc. If you have an existing business with existing revenue, split testing is the closest you can come to just magically adding more money to your bottom line.

So, let me know what your stance on all this is: