One of the nice things about direct mail advertising is the fact that it provides a simple means of testing various approaches. Almost any element of a mail program can be tested.
I am often asked a variety of questions about creating a mailing. Do postcards work better than envelope mailings? Does short copy produce better results than long copy? Will one price produce more results than another?
I always answer the question the same way. “You can test and find out”. People are interested in what results others have achieved with the same questions. What works best in one situation may not work the same in another. So it is best to test and find out for yourself.
How do you do that?
To conduct a test, you divide your mailing into two or more segments of equal size and mail one package to one segment and another to the other. Let’s say that you wanted to test price. Your mailing pieces would be exactly the same with the only difference being the price.
You would code your mailing piece so you could measure the response. Let’s say that you asked your recipient to bring in a card and redeem it to get a special price. Price “A” would have a unique code on the coupon, perhaps simply the letter “A”. Likewise, price “B” would also have a unique code. When the promotion was over you would simply count how many responses you received for each code and you’d have the answer as to which price worked best for you.
It is imperative that both mailings are exactly the same except for the price. Why? Well, let’s say that in addition to having a unique price; one card had different copy from the other. Now let’s say that card “A” got the most response. Was it due to the price or due to the copy? You’d have no way of knowing. So be sure you test one thing at a time.
How big a test do you need? Most direct mail experts say that a good sample is 5,000 pieces of each version. For many smaller mailers, a 10,000 piece mailing could be prohibitive both in terms of cost and also possibly because of not having enough names.
If you can’t manage a test of 5,000 of each version, try to mail as many as you can under that number. Perhaps 3,000 of each. But don’t mail too few. A test of several hundred of each version, for example, is not statistically valid. In other words, the results you achieve may not hold up if you tried the test again. Likewise, more than 5,000 is not necessary. You won’t get any greater degree of accuracy.
What can you test? Here are some ideas:
Copy and headlines
Price
Format
Offer
Lists
Mailing frequency
Free gifts
Type of postage (stamps, meter imprint, printed permit)
So, whenever you have a question about what will work best for you and your product or service consider a simple test. No need to guess when it is so easy to find out for sure.
When you’re ready to mail, don’t hesitate to contact Avery Print and Mail Center. Our staff is ready to assist you. You can reach them at 877-Avery PMC (877-283-7976).
Comments