You'll never know how much money you're losing until you  measure, optimize and test. There are two schools of thought as to how you should start.

School 1

Start with an "minor" page on your site. Use that page as your "sparring ground" to avoid serious mistakes while learning about  testing. The danger in starting your testing with a "minor" page is that it will take long to see results. You'll grow impatient, and may  abandon the concept of testing before appreciating its benefits.

School 2

Start with a high traffic page so you can see results quickly. We all want results now. We don't want to wait for the answer.

The page parts to test are:

1. Headline

It's the first thing visitors see when landing on your site. It will either rivet their attention, or make them feel they are on the wrong page.

It's easy to test. You can easily write a few benefits packed headlines to test. Craft 3-5 different headlines and designate an A/B test to verify the effectiveness of each.

2. Opening Paragraph

The opening lines and paragraph of your sales letter must enthrall the reader. A copywriter will usually prove useful in this function.

An A/B test can be used to measure the effect of each paragraph. If you are doing it yourself write 3-5 paragraphs, let it ferment for a day or two, look at them again, rewrite them and then run the test.

3. Product Benefits

List the most important benefit first. But your visitor may have a different ranking of the most important benefit.

Vary the order of your benefit statements to test which positioning converts at the highest rate. This could provide clues into rewriting your sales letter for higher conversions. It is valuable experience which you can use in other sales letters or share with a friend.

4. Call to Action

There many ways to tell visitors what you want them to do. And depending on your product, visitor etc, some will be more effective than others.

Rather than simply copying what other marketers are doing, choose several different calls to action.  This is a critical section so don’t leave it to chance by being a copycat. Outline 3-5 different segments and label them A/B. Read them after a day or two, run the test.

5. Visual Elements

There is a song by Bread which goes to say that a picture is worth a thousand words. But the picture has to be right. Garish colors, poor photographs all told bad visuals can bring down your conversion rate.

Good visuals on a  page can make a person’s mood change and induce conversion.

Again you have to test. Here the testing becomes very difficult since there are usually many visual elements on  the page. Choose the first ones to test. Choose the large visuals to change.

Don’t try and change many of the visual elements, you’ll end up with what is called a multivariate test which is a prescription for disaster.

Never rest and assume that your conversion rate has peaked. There will always be room for improvement.