Testing to Find Effective Approaches

The article Testing gives details on how and why to test, but sometimes setting up a quick and dirty test can yield quick results and insight.

There is a concept known as A/B Testing which explains the principle. With an A/B Test, you run two different approaches alongside each other then compare the results of each to see which was the most effective.

Online applications such as Google Adwords or email subject lines lend themselves well to it.

With a Google Adwords test you would create two adverts to run simultaneously and get results in virtually real-time. After leaving the test to run long enough to get some reliable results, the Google Adwords statistics will tell you which performed best in terms of the click-through rate, i.e. how often it was shown and how many people responded. You would then pick the winner and use that as your ongoing advert.

Testing email subject lines needs email despatch software. You would set up two emails with different subject lines and send each to a small section of the database, chosen at random. Then, after a while, see which of the two has the best open and click-through rates and send that one to the rest of your customers. You can read more about creating effective email subject lines in this blog posting Creating Effective Email Subject Lines.

Both of these approaches are quick and easy ways to tell you the types of message your customers are most likely to respond to. You could set them up in the space of an hour or so and within a day know what approaches work best, providing sufficient volumes are considered.

A more sophisticated form of the principle would be to test landing pages. A landing page is the first page a visitor lands on when they visit your website. Usually that will be your home page but if your site optimises well it could be anywhere on the site. If you use Google Adwords or directories then you could create different landing pages for the same advert and see which one performs best, using information from Google Analytics such as bounce rates or time on the page.

You have to be sure that you can rely on the results of any tests and ways of doing this are explored in the Trusting Test Results article. If you are unsure on how best to use Google Analytics you could also read Using Google Analytics.