Improve Your Mail Order Results Not sure how many catalogs you should be sending? Try doing a test run first.

Q: I own a small company that sells promotional products to automobile dealers. I send catalogs through the mail using bulk mail. This year I'm sending 120,000 catalogs and need some advice. I thought about buying 60,000 names from a mail list company and sending each prospect 2 catalogs. Or I could buy 30,000 names and send 4 catalogs. We'll have the same basic catalog with four different covers on it-our products don't change that much. I've always heard frequency is more important than quantity with regards to how many times you "touch" your prospects. What are your thoughts?

A: Ongoing testing is essential to determine the frequency that will yield the greatest results and to find the best quality lists for your purposes. It's important not to rely exclusively on one list or one list vendor. Since you plan to mail 120,000 pieces over a year, select two or three lists using the same criteria, called "selects," and determine how recently each list has been cleaned to be certain it's no more than three months old. The undeliverable figure for bulk mail is 10 to 15 percent on average, and a recently cleaned list will help reduce your losses. Purchase your lists with multiple usage rights, but take delivery on your labels (in electronic form delivered to your mailing house) just shortly before each scheduled mailing to make sure you're receiving the most recently updated version.

For companies like yours that rely on direct mail, testing is a never-ending process. To test the quality of the lists from different vendors, mail the same piece to approximately 5,000 names from several lists, and monitor the results. Then buy the full quantity of names you need from the vendors whose lists yield the highest returns.

One way to find the optimal mailing frequency is to divide a list of approximately 30,000 names into four equal groups. Mail once to group A, twice to group B, three times to group C and four times to group D-and evaluate the results. Imprint all the mailing pieces with codes so you know exactly which list or group each sale comes from. Carefully track your conversion data, and use it to refine your direct-mail program on an ongoing basis.