One of the most sensible strategies used by direct marketers is what’s known as a split A/B test. A split test offers a way to compare two different messages, or two different offers, and to determine which will produce a greater rate of return.

Suppose you’re thinking of sending an offer to a list of 100,000 prospects. You aren’t sure whether those prospects would prefer a discount or some kind of bonus offer in return for their business. So what you do is take a small percentage of that list… say 5 percent… and divide that portion in half.

Half of that smaller list receives a marketing message that’s built around the discount. The other half receives a marketing message that gives some kind of bonus when they take the action you desire. Once the results are in, you can determine which approach produced a greater response. With all other things being equal, the approach that did better is the approach you should roll out to the entire list.

There is another benefit to this approach. You can usually project the results of your test across the entire mailing. For example if your chosen approach generated a response or sale from 5 percent of the group that you mailed it to, you can assume that roughly 5 percent of the entire audience of 100,000 will respond to your offer. That makes it much easier to project your ROI and to determine other actions you need to take related to the marketing effort, such as ordering premium items.