Lars Perner, Ph.D.
Assistant Professor of Clinical Marketing
Department of Marketing
Marshall School of Business
University of Southern California
Los Angeles, CA 90089-1424, USA
(213) 740-7127

Consumer Psychologist Facebook Forum


Direct Marketing

We consider direct marketing early in the term as a “contrast” situation against which later channels can be compared.  In general, you cannot save money by “eliminating the middleman” because intermediaries specialize in performing certain tasks that they can perform more cheaply than the manufacturer.  Most grocery products are most efficiently sold to the consumer through retail stores that take a modest mark-up—it would not make sense for manufacturers to ship their grocery products in small quantities directly to consumers. 

Intermediaries perform tasks such as

Direct marketers come in a variety of forms, but their categorization is somewhat arbitrary.  The main thing to consider here is each firm’s functions and intentions.  Some firms sell directly to consumers with the express purpose of eliminating retailers that supposedly add cost (e.g., Dell Computer).  Others are in the business not so much to save on costs, but rather to reach groups of consumes that are not easily reached through the stores.   Others—e.g., online travel agents or check printers—provide heavily customized services where the user can perform much of the services.   Telemarketers operate by making the promotion in integral part of the process—you are explained the benefits of the program in an advertisement or infomercial and you then order directly in response to the promotion.  Finally, some firms combine these roles—e.g., Geico is a customizer, but also claims, in principle, to cut out intermediaries.

There are certain circumstances when direct marketing may be more useful—e.g., when absolute margins are very large (e.g., computers) or when a large inventory may be needed (e.g., computer CDs) or when the customer base is widely dispersed (e.g., bee keepers).

Direct marketing offers exceptional opportunities for segmentation because marketers can buy lists of consumer names, addresses, and phone-numbers that indicate their specific interests.  For example, if we want to target auto enthusiasts, we can buy lists of subscribers to auto magazines and people who have bought auto supplies through the mail.  We can also buy lists of people who have particular auto makes registered.

No one list will contain all the consumers we want, and in recent years technology has made it possible, through the “merge-purge” process, to combine lists.  For example, to reach the above-mentioned auto-enthusiasts, we buy lists of subscribers to several different car magazines, lists of buyers from the Hot Wheels and Wiring catalog, and registrations of Porsche automobiles in several states.  We then combine these lists (the merge part).  However, there will obviously be some overlap between the different lists—some people subscribe to more than one magazine, for example.  The purge process, in turn, identifies and takes out as many duplicates as possible.  This is not as simple task as it may sound up front.  For example, the address “123 Main Street, Apartment 45” can be written several ways—e.g., 123 Main St., #123, or 123-45 Main Str.   Similarly, John J. Jones could also be written as J. J. Jones, or it could be misspelled Jon J. Jonnes.  Software thus “standardizes” addresses (e.g., all street addresses would be converted into the format “123 Main St #45” and even uses phonetic analysis to identify a likely alternative spelling of the same name.

Response rates for “good” lists—lists that represent a logical reason why consumer would be interested in a product—are typically quite low, hovering around 2-3%.  Simply picking a consumer out of the phone-book would yield even lower responses—much less than one percent.  Keep in mind that a relevant comparison here is to conventional advertising.  The response rate to an ad placed in the newspaper or on television is usually well below one percent (frequently more like one-tenth of one percent).  (More than one percent of people who see an ad for Coca Cola on TV will buy the product, but most of these people would have bought Coke anyway, so the marginal response is low).