Use Modeling for Expires, Outside Lists

Technological capabilities have advanced almost to where if you can think of it, it can be done. This mentality has led to cutting-edge prospecting tools for many publishing mailers working with brokers who have stayed ahead of the curve.

Publishing mailers have used prospecting models successfully for some years. For example, they would send a group of their core direct mail responders to a company such as Donnelley, Equifax or Experian. Those names would be matched against the modeler’s database to determine strong and weak traits of those individuals, such as age, income and gender. The database then would be narrowed and scored/ranked from most likely to least likely to subscribe.

A publisher could mail to these individuals, sometimes mailing deep within a model. This has been a great tool for acquiring subscribers. But with the knowledge obtained from the acquisition model as described above, a mailer also could optimize both its outside lists and its own expires.

For instance, if you learned that typical responders to your direct mail campaigns are women older than 40 with household income surpassing $50,000, you could overlay your outside lists and expires to find similar individuals – those most likely to respond to your offer. The key ingredients to making this work are:

• A list or group of lists having a universe large enough to which a model can be applied.

• A broker who has the understanding to win over the list managers and list owners.

• A list manager who can understand and relay the broker’s request to the list owner.

• A list owner who understands the long-term benefits on its file and that it would take better pricing to get this to work properly.

• An annual mail volume that shows the list owner that significant potential exists.

• A database large enough to model most of the names you are looking to optimize. For instance, if you input 10,000 names from an outside list, you want the model to find traits (matches) on most of those names (a 10 percent match rate doesn’t cut it).

As always, the biggest issue is pricing. A cost-benefit analysis is needed before using this method of modeling. Some modeling firms require a set-up fee; others just require a good-faith understanding that you will do your best to meet annual volume commitments.

Further, names being bumped up against the model will incur a model scoring fee typically calculated on a per-thousand basis. Needless to say, these costs coupled with typical list costs really add up, probably to where the best model in the world would not offset the costs.

To improve your P&L, the broker must take an active role. The broker must understand the various costs of such a process and apply them to the mail plan. Using experience, the broker must determine the expected lift to both the gross and net levels for each specific model. Using historical response as a benchmark, the broker can determine what list costs are needed in order to attain or surpass the mailer’s goals. List owners would expect payment for names not mailed, and this could be done by paying run charges or working out a net guarantee. Each circumstance will be different, but it’s the broker’s job to ensure the best possible deals, which allow for the best chance of getting the optimization to work.

Related Posts