Segment Co-op Database Names and Models

Jun 18, 2007 8:57 PM  By

Most mailers could stand to segment their cooperative database selections to fine-tune prospecting. Here are two suggestions on how to do it.

First, segment the co-operative database names into “previous” and “new” categories and measure the difference in response. Previous means the name was served up in the last mailing by the database, and often the previous names respond better than new ones.

If you find a difference in response between previous and new names, you may want to fine-tune your contact strategy and mail to the better responding segment more often. You may find opportunities to squeeze in more mailings to new names that didn’t get your last catalog mailing.

Also keep track of the typical mix of new vs. previous names because if the mix had been 50-50 and suddenly you see that 85% of the names are new, you may need to confirm that the model was run correctly. On the other hand, if 75% or more of the names are previous names, it may mean that some big contributors to the database haven’t updated their files, and you won’t get many fresh names for your next round of prospecting.

Another tip is to segment your co-op database models into smaller segment sizes.

With smaller segment sizes, like 20,000 rather than 30,000 or 50,000, you will find more marginal names. You’ll have more control over where you draw the line on how deeply you want to prospect.

The result is you’ll have a better handle on the prospecting names you mail. And your average response should increase because you have the ability to rest weaker segments when you work with smaller segment sizes.

Jim Coogan is president of Santa Fe, NM-based consultancy Catalog Marketing Economics.