FOR MANY CATALOGERS, the right customers may be just a modeling application away. Confronted with shrinking list universes, reduced response rates, and overmailed “best customer” segments, more catalogers may turn to modeling to find new or better targets from both house files and prospects.
But while an estimated 85% of catalogers conduct basic RFM (recency/frequency/monetary value) modeling, according to industry figures only 5%-10% engage in the more-sophisticated types of modeling, usually because they find the techniques difficult to understand, not to mention expensive.
Costs are declining, however. “List owners are a little more willing to negotiate modeling prices,” says Stephen Tamke, senior vice president/director of list brokerage at Hackensack, NJ-based Mokrynski & Associates. Several years ago, he says, a typical base price for catalog list rental was $100/M, with modeling charges adding another $40/M-$60/M to the cost. Today, Tamke says, brokers often secure a base price in the $75/M-$80/M range with modeling costs of $30/M-$50/M, bringing the cost of a modeled list to the same level as that of a traditionally segmented list.
As for the complexity of the modeling process, more and more list firms are offering inhouse modeling services or are outsourcing such services to third-party providers. “I work with my list broker, I’ll tell him what I want and how many names I want, and he’ll coordinate the specifics with the modeling company,” says Jason Scheets, chief operating officer of Bethel, CT-based licensed sports merchandise cataloger Star Struck. The list firm of Richmond, VA-based Children’s Wear Digest “helps determine specific lists that might be worth modeling,” says Tracy Schneider, marketing manager of the children’s apparel catalog.
SPOILED FOR CHOICE
The cataloger’s marketing goals will determine which database model to use. “For a model to be effective the cataloger needs to be involved from the inception,” says Joseph McCluskey, senior vice president of Bethel, CT-based List Services Corp.
And mailers need to ask the right questions to define the right objectives. “For a food cataloger, the question may not be ‘Why do men order chocolates and flowers in February,’ but ‘How can we get more men to order in January or March?’” says Geoff Batrouney, executive vice president of New Rochelle, NY-based Estee Marketing.
Once goals are defined, basic RFM analysis is a good place to start. “RFM usually precedes the statistical model. It’s a little easier because it doesn’t necessarily involve statistical elements,” says Peter Vlahakos, vice president of analytical services for Woodcliff Lake, NJ-based Donnelley Marketing, parent company of list services company Walter Karl.
Moving beyond RFM depends on a cataloger’s willingness to share proprietary information with its broker. “If we’re building a database, we need to receive the response information or transaction information,” says Tito DiFilippo, president of Pearl River, NY-based Database Direct, a sister company of list firm Edith Roman Associates. This information often includes inbound activities purchase data such as key codes, purchase dates, items purchased, and dollars spent; customer requests and inquiries; Website traffic data; and data about promotional offers and other outbound activities. In addition, for consumer catalogers, capturing age, gender, presence of children, and household demographics might be useful. For business-to-business marketers, company demographics such as employee size, SIC, and industry information might apply. “The model can only be as strong as the data on the database,” says DiFilippo.
And when it comes to modeling, size matters. Catalogers with sales of more than $50 million can often benefit from segmenting their database to target prospects for a campaign by product category. For example, an apparel mailer may find that customers who have purchased sweaters are likely candidates for dress offers.
Conversely, “if I’ve got 100,000 total customers, I’m not getting a lot of valuable data out of slicing and dicing my database 20 ways, because segment sizes get so small,” says Steve Trollinger, senior vice president of marketing for Shawnee Mission, KS-based consultancy J. Schmid & Associates.
A DATABASE OF ONE’S OWN
A number of list firms offer proprietary databases. The Millard Group, for instance, has RFM Plus, in which about 50 catalog companies make their most recent buyers available. “It’s an open database. You know who is participating,” says Ben Perez, president of the Peterborough, NH-based list firm. “The names are shared among the participants and made available for noncompetitive offers on a rental basis.”
The Datamatch Consumer Direct Database, from Bethel, CT-based List Services Corp., is updated monthly, says senior vice president Joseph McCluskey. In January 2004, the 12-month hotline had more than 39 million names, and the monthly update was more than 4 million names.
On the business-to-business front, Greenwich, CT-based Direct Media’s Data Warehouse, launched in 1979, has 92 million business names from more than 900 business mailing lists. “It is selectable by so many criteria, including individual lists, that it’s used by everybody, consumers as well as b-to-b mailers,” says executive vice president Linda Huntoon. And MeritBase, from White Plains, NY-based MeritDirect, has 150 million names from more than 100 business catalogers.
Not to be outdone, Pearl River, NY-based list company Edith Roman Associates launched the Roman Alliance Database (RAD) this spring. “The RAD contains approximately 50 million business executives and professionals in the U.S. and Canada,” says John Ganis, president, list management. “These executives are from responsive files within Edith Roman’s alliance of list owners.
The RAD offers mailers postal, e-mail addresses and phone numbers, and can be selected by job function, SIC, employee size, sales volume, SOHO [small office/home office buyers], multibuyers, and executive’s specified buying authority.” According to Ganis, every business site in the country is represented in the database.
And several list firms, including Lake Group Media and Direct Media, will build private databases for clients. This is a particularly good option for large b-to-b catalogers that mail to more than 15 million-20 million names a year, says Ryan Lake, CEO of Rye, NY-based Lake Group.
But consumer catalog marketers can benefit from private databases too, adds Direct Media’s Huntoon. The owner of a private database can rent names, then merge/purge, score, and mail them according to their needs. The database owner pays the list owner only for the names it mails. “In a consumer world, the great majority of mailers have their names in a co-op database,” Huntoon notes. “Why wouldn’t they build one of their own?”
12% of catalogers don’t know how many purchases a customer must make before they are making a profit from him.
Source: Catalog Age 2003 Benchmark Report on Marketing