How To Start Split-Testing Your Web Pages

In working with a client recently, it soon became apparent that we each had different ideas on what would work best with his integrated marketing campaign. I suggested we do some split-testing and roll out the “winning” combination in our full-scale program.
 
My client was initially concerned that the time and expense of split-testing would never pay off. If these are the same concerns that are holding you back from split-testing your own web marketing, allow me to relieve your fears.
 
First, it’s quite understandable that many marketers think split-testing is both time-consuming and expensive. In fact, Future Now, Inc. has identified over 1,100 pieces that contribute to a successful conversion funnel. And if you take into consideration your various prospect personas and different emotional motivators, factor in your numerous offers and multiple all that by the number of your landing pages… yikes, I’m already overwhelmed just thinking about it!
 
But if we break the testing process down into actionable steps, it becomes much easier to look at web site split-testing as just another necessary cog in the internet marketing wheel.
 
The first step in split-testing your web page actually happens before your page is ever created – and believe it or not, this is a step that many marketers miss altogether. The first step is deciding what you want your page to do.
 
It might sound silly, but in our web site critique process we often run across web pages that do not have a clearly defined goal. Before you can improve conversions, it is necessary to define exactly what a conversion is.
 
Once you’ve decided the overall goal of the page, you must determine how it will be measured. Are you looking for an overall number? Percentage of visitors? Action within a specified timeframe?
 
Next, settle on the platforms that will be used for testing and measuring. Will you be using your standard web logs to measure traffic? Your shopping cart program to measure conversions? How will you feed different versions of the page to your visitors? Will you be using timed tests, replacing your page with a new version after a pre-determined time? Or will you simultaneously display different page versions to unique visitors? Will you be using the free Google Website Optimizer or the robust Offermatica or Optimost platforms? Or your own software and database program?
 
If you’re not sure where you would like to start testing, why not identify your least effective web page. Simply review your traffic logs to determine which page is receiving the lowest conversion rate based on the number of visitors it gets.
 
And finally, it’s time to identify the elements that can influence the conversion process. Will you be testing graphics? Text and copy? Placement of different elements? A combination of the above?
 
Some of the elements you should consider testing include:
 

• Calls to Action
• Security Indicators
• Product Narratives
• Buy Button Text

While it might seem a little intimidating, the key to testing is to just get started. If you “make a mistake” and land on an element that decreases conversion, it’s easy to go back to square one and test something else.
 
If you’re nervous about getting started, I invite you to download a copy of my own web site split-testing schedule. It’s a great way to begin your testing and log your results, and you can grab a free copy at my web site…
 
 
Don’t have a web site yet? Want to get it on line fast? Check out our new Rapid Development Web Site>>>

You might also be interested in:

• An Easy Website Split-Testing Schedule You Can Use
• Increase Landing Page Conversion By Testing These 5 Elements
• Does Short Web Copy Outperform Long Copy?
• Increase Open Rates: Be Direct With Busy Readers
• 6 Steps For Managing Your Search Marketing Campaign

Tags: conversions, sales conversions, web site split-testing, website split-testing

This entry was posted on Thursday, May 6th, 2010 at 12:18 pm and is filed under Affordable Website Design, Small Business Marketing. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.