If you have experience in the field of marketing, then you probably know what a split test is. IF NOT…read this article. Now. You Need It. So let me tell you about it.

Here is the deal with split testing. The same message delivered two different ways can have very different results. And it’s not just big things. It can be something very small. It can be one word. If you can figure out what that word is…you can get very rich.

So split testing is conducting tests to figure out the best way to say or present a marketing message.

In advertising terms, it’s running two or more pieces of ad copy at the same time to the same market to test the difference. That “same market” part is important. For split testing to work, you have to have an “all else equal” kind of approach. Everything except the actual ad needs to be the same.

A split test can be looked at as having stages. In stage 1, when first beginning a test, you might use two completely different ads. You would run those ads against one another to see which one performs best. This requires some kind of way to know which ad is performing.

The number one way is to compare how much revenue each ad generated. Or maybe it’s how many leads you get from each ad. Split testing really doesn’t work without some way to compare results.

There are different ways to accomplish that. I’m an Internet marketing guy, so for me it would involve using tracking parameters. I won’t go into great detail about that in this article, but tracking parameters are just little pieces of information passed into your website analytics through a link URL.

However you do it, you compare the results of the two different ads and see which one worked better. If neither one made any money, then maybe you just need to start over.

Assuming one of the ads did reasonably well, then you would then create two different versions of that ad. Maybe you try a different picture in one of them. Or maybe you just change the title. Or it could even be the offer itself. This process can continue in finer and finer detail. It becomes a process of continual, gradual improvement.

The concept of split testing is not new. Marketing professionals have been doing it for years. Traditional split testing is time consuming and expensive. It involves running ads in different types of media and manually comparing results. This could be magazines, radio, TV, sales letters, post cards  – all expensive forms of advertising.

That’s a big advantage of Internet marketing. It makes this process remarkably quicker and easier, and you can start with a lower budget.

Google built split testing into its Ads platform from the very start. You can set up multiple ads for the same set of keywords and Google will automatically rotate the ads. Since the ads are used for the same set of search queries (or display network locations, or wherever the ads are being shown), it will by default represent a fair split test.

Facebook also offers an automated split test option when running ads on Facebook. All it requires is for you to take the time to set up two different ads.

The results of your Internet split testing can be applied to your other marketing. In fact, you can look at Internet split testing as a low cost way to conduct split testing for all of your other advertising. Maybe you discover a headline or slogan that works very well. You may find that it also works very well with your offline ads.

It can take a while to be able to judge the performance of your ads. Your budget plays a bit part in that. The faster you can afford to spend money and generate data, the sooner you will know what works. Split testing isn’t a magic formula. But if you can figure out how to do it for your business, it can help you greatly improve how well your advertising works for you.

So now go out into the world, young businessperson…and split test away!