SEO: Search Engine Optimization Bible

Understanding and Using A/B Testing

You’ve heard it so many times by now that you’re probably sick of it, but testing is your best friend

during any SEO efforts that you put forth, and specifically with your PPC efforts. The most common

and easiest-to-use form of testing is A/B Testing. A/B Testing in its simplest form is to test one ad ver-

sion against another.

For a PPC advertisement effort, that means testing two or more ads at the same time. You should

follow some guidelines when testing different ads, however.

The first of these guidelines is to understand what you hope to achieve with the testing. Are you

trying to increase traffic? Are you trying to increase conversions? Or do you want to increase or

change some other element of your PPC performance? Ask yourself these questions and then use

that as a guideline for making changes to test.

When you know what you’re trying to accomplish, you must start your testing with an ad that serves

as your baseline. In order for this to happen, you have to first create at least one PPC ad and then

allow it to perform long enough to have a good baseline of data against which you can compare

future ad designs.

Once you have your baseline established, you can begin comparing that ad against new ads

that you’ve created to test different elements of the PPC ad. Some of the elements you can test

include:

 The headline

 Keywords

 Call to action

 Ad text

 Different landing pages

The key to successful A/B Testing is to change only oneelement in the test ads. And once that ele-

ment has been changed, you have to give the ads time to develop performance data so that you can

tell which of the ads is performing better. If you try to change more than one element at a time, you’ll

find yourself in the position of having to decide what ad changes affected the traffic and conversion.

Stick to one change at a time to make it easier to be certain what changes cause what results.

When you’re A/B Testing your PPC ads, you should have a control ad — that will be the one you

used to create your baseline — and also a test ad. If you don’t have a control ad, you don’t have any-

thing to compare your results to. Just remember that you must run both the control ad and the test

ad at the same time in order for your testing results to be accurate and meaningful.

Using A/B Testing, you can learn what works and what doesn’t in your PPC advertisements. It’s

not the most sophisticated testing method that you can use, but it’s the easiest, and it’s highly

accurate. Testing your ads will help you tweak and fine-tune them until they are as successful

as you can make them.

122

Part II SEO Strategies

75002c07.qxd:Layout 1 11/7/07 9:38 AM Page 122