A List Selection Model

In constructing a test mailing from list data sheets, I use a letter-number system for prioritizing lists. Letters — A, B, C, etc. — designate how closely the list fits my customer profile. Since an exact fit would likely be available only from a direct competitor, and since they probably won’t rent to you, your “A” designation will have to be approximate, but as close as you can get with the list selections you have.

For example, if I’m promoting a newsletter on the subject of recruiting, I probably will have to live without subscribers to a hypothetical “Recruiting” magazine (unless I can work out a swap, which is always worth a phone call). But a compiled list of “corporate recruiters” would rate an “A.”

A list of human resources managers might also rate an “A,” but a list of human resource directors or VPs wouldn’t. Why? Because human resource managers often do the actual recruiting work and they need the help. HR directors and vice presidents are somewhat removed from the day-to-day recruiting process (been there, done that) and probably wouldn’t think they need hands-on advice on recruiting.

Watch Out for Fuzzy Logic

OK, you might say, but wouldn’t HR directors want their people to know as much as possible about recruiting and pass the mailing piece along to them? Or, might not some of these HR VPs make hiring decisions? Sure, some might. But that’s not a primary and normally anticipated behavior for this group. Be careful of that kind of fuzzy logic, tempting as it can be. It will have you selecting marginal lists, which will tend to drag down your aggregate response. If the promotion or offer isn’t directly related to the person receiving it, your odds of a sale drop dramatically.

The numbers are assigned according to the total size of the list, relative to the lists I have to choose from. The largest lists rate (1), next largest (2), etc.  Smaller lists (3) can always be added to a continuation or rollout later at low incremental cost, rather than claiming a spot in the initial test. You don’t want to mail too many small lists, since even if you get a good hit on some of them, there won’t be many names to go back to.

I then construct my test list A-1s first, A-2s, B-1s, B-2s, and so on. If you find yourself getting down into “C” lists rather quickly, you may have a list problem, or need to look further.

Some additional points to remember when renting and using lists:

  • Be sure all lists are coded by test panel (or order forms are coded if the record is not on the return piece) so you can record results by list.
  • Make list selections by “Nth” name to spread your test evenly across the list.
  • Always eyeball the list before you mail. You might even pull some names at random and call them to verify that they are who you think they are.
  • Remember that the list owner will require a sample mailing piece before approving the list rental. Plan ahead.

#            #            #

Adapted from Streetwise Direct Marketing by George Duncan, Adams Media, Inc.
January 2001.copyright 2000-2001, George Duncan, all rights reserved.

Leave a Comment