What typically pulls best? Long or short copy?

What have you found in your experiences with short and long copy in direct marketing communications? Does copy need to shorten to remain effective today?

I run into all kinds of marketers in many industries in my consulting business. The projects include newspaper subscription promotions, B2B lead generation, B2C insurance lead generation, retail store traffic building, B2C catalog sales … the list goes on from there. The question I often get from both experienced and inexperienced direct marketers deals with copy length. Most copy length comments involve direct mail letters that go over one page.

The common statement is: “Well, I would never read all of that.” Then I simply reply that testing in multiple media and various industries show that better results come from saying what you need to say to get the prospect to act.

That is why it is so important to make the copy skimmable with easy to read type, short paragraphs and paragraph headings.

There really is no answer to the question except that the copy length in successful campaigns will exceed what you are comfortable with if you have to ask the question.

Making a sale off the page in ANY medium requires more copy than lead generation copy. When getting a lead, you are building interest in the product, but your primary goal is to sell the appointment. Leave some selling ammunition for the sales person to close the deal.

Selling a product, however, requires anticipating and answering all of the objections. Furthermore, the copy translates the product features into benefits. That approach takes more copy and time than simply positioning the product for branding purposes.

Two-minute spots typically get better ROI than 30 second DRTV spots. Longer emails that must get orders off the page take far more copy than simply trying to generate a click through to your web site. Two and even four-page letters remain controls for years in direct mail.

There are always exceptions, but if you are competing with a selling message that does not answer all of the objections, then go for longer copy to beat your control.

The counter argument I get is that people are too busy today. Buyers don’t read any more. Just stick with the facts before they loose interest. And the list goes on.

The answer to all of these comments is simple.

We can hypothesize why something should or should not work, but in the world of direct response, only the customer’s vote counts. If you test your hypothesis, and it doesn’t work, then the answer is that the hypothesis was wrong. Otherwise, go with what works based on true A/B split testing.

Does this sound counterintuitive to you? Have you found through A/B split testing that shorter copy out pulled your long copy control by a significant margin (that is at least by 15% cost per sale or cost per lead)? What are your thoughts on the subject?