When I was growing up, there was only one way to dry clothes: We attached them to clotheslines in the back yard using wooden clothespins.

Then modern science took over, and we got plastic clothespins with metal springs inside. And we thought that was the last word until home electric clothes dryers came along and ruined the clothespin industry.

The same thing is about to happen to the mailing list industry. And the cause of the disruption is prospect databases.

Here’s why. Today there are more than 40,000 lists for rent. A typical mass mailer will rent 300 lists, merge/purge them, and code the output by list. Some lists work well, others not so well.

This system has serious limitations. For example, mailers can learn only two things from their acquisition mailings

When I was growing up, there was only one way to dry clothes: We attached them to clotheslines in the back yard using wooden clothespins.

Then modern science took over, and we got plastic clothespins with metal springs inside. And we thought that was the last word until home electric clothes dryers came along and ruined the clothespin industry.

The same thing is about to happen to the mailing list industry. And the cause of the disruption is prospect databases.

Here’s why. Today there are more than 40,000 lists for rent. A typical mass mailer will rent 300 lists, merge/purge them, and code the output by list. Some lists work well, others not so well.

This system has serious limitations. For example, mailers can learn only two things from their acquisition mailings — namely, which lists and offers work best. They have to wipe their systems clean of the mailed addresses and get to keep only the responders’ names.

A prospect database lets you do much more. The mailer negotiates with list suppliers to let him have their names on his file for an entire year (or a quarter). He pays the list owner whenever one of their names is used.

Since he has the names for an extended period, he can afford to append demographics like age, income, presence of children, home value, dwelling type, own vs. rent, length of residence, mail responsiveness, cluster coding and about 20 other important facts. He could never afford to append this information to a list he rents for a single use.

He also can better keep track of who was mailed to but didn’t respond. His subsequent mailing selections can be made not only by list, but by demographics and behavior. This produces better response rates, according to those who have done it.

The next step is to develop models that predict the type of person that will respond to each offer. This, too, can improve response rates. But that’s not the end of it.

The same model can be used to select consumers from a compiled file. These large databases contain the names of almost every consumer in the country. A well-refined model will select nearly all of the same people from a compiled list that it would from a large group of rental properties.

Compiled lists can be rented for about half the cost of response/behavioral lists. A mailer that’s been paying between $70 and $120 per 1,000 names from response lists finds that he can get complied lists for $45 to $55 per thousand. Even better: many compiled lists are already coded with all the appended data that are needed for a model.

Case in point: An auto insurance mailer hired an outside service to append data to its database and used a model to predict the response. Using this model it did a better job of selecting responders, and its monthly profits doubled. Fig. 1 is a before-and-after chart of the monthly mailings.

Let’s take it a step further. When you factor in the reduced cost of the names from using a compiled list, the profit takes another jump (Fig. 2).

That’s not the end of it. One big cost of monthly mailings is the merge/purge process. Three hundred tapes arrive which have to be reformatted, run through NCOA, have postal hygiene applied, and are deduped to produce a mailable output file. Few merge/purge processes can be run for much less than $12 per thousand, all steps considered. With a prospect database, the monthly merge/purge is considerably simplified. The main file is already deduped and has been run through a quarterly NCOA and data hygiene. The monthly cost will be mainly for renting and adding hotline names, plus postal presort. The savings therefore are substantial (Fig. 3).

The second case study concerns a large mailer that markets products through more than 30 campaigns per year. Working with an outside vendor, this firm decided to reduce list acquisition costs by building a prospect database.

How did this mailer do it? The vendor combined a compiled file with the mailer’s existing customer base. (This included solicitation and response history.) Altogether, the process involved more than 1 billion records. Then the two firms contacted the mailer’s list broker to renegotiate usage arrangements for their rented lists.

The prospect database environment gives the mailer and broker a much more detailed view over time for which lists really drive response at a cost-effective rate. The resulting leverage can be used to negotiate appropriate deals for other list sources that can be incorporated into the prospect environment.

Sometimes this leads to lower nets or net-net arrangements because it more clearly shows the value of the list to the mailer. This, in turn, enables the list owner to avoid potential loss of use through overzealous cost cutting.

What was the overall result? In the first year, the prospect database produced cost savings to the mailer of more than $1 million, while at the same time boosted response rates per piece mailed.

Building a prospect database will work if you’re a high-volume mailer marketing to consumers. You can use a prospect database if you’re renting a variety of lists with multiple campaigns each year, and if you’re willing to negotiate with list managers and owners.

One by one these high-volume mailers will shift to building prospect databases. The list industry will never be the same.

Arthur Middleton Hughes is a vice president and solutions architect at KnowledgeBase Marketing.

FIG. 1

FIG. 2

FIG. 3