Confronted with shrinking list universes and reduced response rates, more catalogers are turning 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.
According to the CATALOG AGE Benchmark Report on Database Marketing, 81% of respondents maintain a database. Determining promotion recipients is the most common use of a database, with nearly two-third of respondents using it for that purpose. Fifty-eight percent use database marketing to customize offers, while 42% use it to locate cross-selling opportunities, 38% to generate list rental revenue, and 29% to find upselling opportunities.
Modeling is the interaction between historical data and variables and which provides a scoring mechanism to identify the best names to mail and/or the worst names to suppress. Simply put, models are used to predict future events using information from past events.
So why build a model? If a cataloger does not have a lot of weak performing names, there is probably not a need for a model. Or, when a cataloger is mailing more than 90% of its buyer file, there is no need to build a model.
The Webinar will cover the modeling options available, and the pros and cons of each. It will also cover how to determine which database modeling strategy is best for you and what you’ll need in terms of expertise, systems, and other resources to reap the greatest return on investment.
Panelists include Michael Grant, principal of Michael Grant Direct and Elisa Krause, Ph. D and vice president, analytical services for New York-based DoubleClick.
To learn more about database modeling beyond RFM, log onto http://www.bulldogsolutions.net/primedia/frmRegistration.aspx?EventGuid=8F42F51F-820E-4D87-A779-C2B6D8985824
The Webinar begins at 2:00pm Eastern