The Importance Of Split Testing

I get a lot of questions from subscribers and website visitors asking me what the most important part of running your own business is. Usually, they guess answers like finding the right product to sell, finding the right market to sell to, good customer service, automating tasks, writing good sales copy. Well, all of those things are very important, and picking one single most important part is extremely difficult when so many are vital.

But if I had to settle on one, my answer would be testing.

And I mean testing everything. You have to test constantly to know for sure what works best. If you don’t test different marketing strategies then how will you know what marketing strategies work? If you don’t test different products, then how will you know what products sell? If you don’t test products with different customers, then how will you know which is the right market for your product? If you don’t test different offers and benefits, then how will you know how to offer the best customer service? If you don’t test your sales copy, then how will you know if you have good sales copy.

I’ve actually met people who didn’t feel like they needed to test different aspects of their business, because they already felt like they’d gotten it right. For example, anyone who tells me that they’ve already written the perfect sales letter gets a response from me telling them they had already failed in their business. You might have written a sales letter that works pretty well for you, but how could it possibly be perfect if you’ve never tested it against any other sales letter. Unless you’re just that lucky.

Plus, if you’re working online, testing is so easy to do, why not at least try it?

Which brings me to split testing. Split testing is essentially trying out two or more marketing strategies at the same time to see which one yields the best results. So you can try two different ads, two different sales letters, etc. and then keep an eye on your stats and you’ll know which one works best. But it’s important that you only test one section of these things at a time. Which means that if you want to try two different sales letters, use the same sales letter but with two different headlines. When you’ve settled on a headline that works well, try two sales letters with the same headline, and two different sub-headlines. Then keep doing this with every part of your sales letter. If you test more that one part at a time, then you won’t be able to accurately tell what’s working and what’s not.

If you use Google Adwords for your pay-per-click advertising, then you’re in luck because Google has made split testing about as easy as it can be. They will let you run two or more ads for every keyword that you bid on, and they will send traffic to both of them for you. The default setting has Google send more traffic to the ad that converts better, but you can set it to an even 50-50 split if you like.

So try this with your Google Adwords campaign. Set up two ads for any keyword list you have. Use the same ad, but with two different headlines. You only have 25 characters for the headline so don’t go crazy, but it’s still plenty of room to do this right. Google will provide you with every type of statistic you will need to measure your results, including number of clicks, click through rate, how much you’re spending and so on.

Let your ads run for a while, and after you’ve figured out which headline works best, delete the other one and replace it with another headline. Keep doing this until you’ve found the best possible headline for your ad. Then move on to the the body of the ads and do the same. Sometimes simple changes like capitalization or rearranging two words will make a difference. So basically, test everything.

But you’re not done there. Go back and do it again. Test more headlines and more ad copy. There’s no reason why you should ever be satisfied with your results. There’s usually a very good chance that you can continue to improve your ad’s performance.

While all that is happening, you are going to want to test your landing page. Your landing page is the web page that visitors go directly to after clicking on your ad. This should be your sales letter, or opt-in page, or whatever page leads the visitor directly to the action you want them to take. Google will allow you to split test different landing pages. In fact, they have a whole Website Optimizer area that will help you with it. It's completely free to use, so get in there and try out different headlines, sub-headlines, move different sections around the sales page, try it with and without pictures, try it with and without audio, try different “buy now” text, everything.

Google also has a piece of code that you can add to your website that will track conversions for you, so you will know how much you’re spending per sale, or per lead, or whatever you like. I highly recommend using this feature as well so you have every possible stat you can get. It’s important to stay on top of these things. And, again, it's free to use.

Google has created an excellent system for you to get real, measurable results for your pay-per-click campaigns. And since it’s working so well, some of the other pay-per-click search engines are getting in on the act, too. Most likely, they all will eventually if they want to keep up. If you don’t have a Google Adwords account, you can get one by following this link: https://adwords.google.com/select/Login

And even if you’re not using pay-per-click, you can still test every part of your website and your sales process. There is no good reason not to.

If your website isn’t performing as well as you like, then keep testing. If your website is perfoming as well as you like, keep testing. It’s the best way to squeeze every cent of profit out of your business. And almost any questions you have about your business can be answered by testing. Keep that in mind.

And if you put in into practice, then you can almost guarantee your own success.