PLATFORMS: Windows 95 or Windows NT platforms.
BACKGROUND: The Fort Worth Star-Telegram is a morning edition, daily newspaper in the Fort Worth-Arlington area. The newspaper boasts a circulation of 240,000 Monday-Saturday and 340,000 on Sunday.
PROBLEM SOLVED: The Star-Telegram sought to perform a database marketing
campaign to increase its circulation to PIA (Paid in Advance) subscribers. As part of this effort, the company wanted to increase the response rates to its direct marketing programs and also eliminate the waste associated with the mailings. Before using MODEL 1, the company was using segmentation only to send out mass mailings.
PRODUCT FUNCTIONALITY: The Star-Telegram uses the MODEL 1 Response Modeler module to perform various levels of linear and logistic regression that analyze the sensitivity and importance of certain variables. With MODEL 1, the company gains a better understanding of the significant variables as well as a higher level of confidence in the selection process. The end result is a coded file of prospects who are demographically, geographically and psychographically similar to the Star- Telegram's customers and who are, therefore, most likely to respond to the direct mailing and telemarketing efforts. In the future, the Star-Telegram intends to use RFM (Recency, Frequency, Monetary) analysis to analyze historical data.
STRENGTHS: MODEL 1's main strength is its ease of use. While it is good to know the terminology of statistics, the user doesn't need to be a statistician. The wizard functionality walks the user along in a point-and-click manner. The scoring was virtually flawless and is a strong part of the product.
WEAKNESSES: MODEL 1's price is a drawback. Also, the user cannot manipulate reports within MODEL 1 and has to perform several steps to export them directly to a spreadsheet or database. MODEL 1 offers fixed field or comma delimited export options. To gain access to over 70 additional file formats, the user must export the results as a text file and then convert it to the desired file format.
SELECTION CRITERIA: The Star-Telegram selected MODEL 1 on the basis of its ease of use and the need for only a basic understanding of statistical terms. The company believes that a statistical model is important to have in determining a targeted customer mailing. Other competitive products were too cumbersome to use, and now the staff can be more productive.
DELIVERABLES: The Star- Telegram uses MODEL 1's Response Modeler module to perform logistic regression, linear regression and neural network models. The system delivers a scored data file which is used for prospecting. In addition, MODEL 1 creates various analytical reports (sensitivity report, model sensitivity report, variable summary report, campaign report and model performance report.)
VENDOR SUPPORT: Our experience with Group 1 Software has been very positive. We found Group 1 to be easy to work with during the sales cycle and very supportive and accommodating during the pre-implementation phase. Group 1 even ran a model using our actual data and showed us the results.
DOCUMENTATION: The MODEL 1 Response Modeler Users' Guide is self-explanatory. The design is attractive, and the copy is written in laymen's terms. While it may help to have exposure to statistics, a user can definitely run the product using the documentation. Other documentation in the MODEL 1 product line includes Model Cross-Seller Users' Guide, MODEL 1 Customer Valuator Users' Guide and MODEL 1 Segmentor Users' Guide.
Register or login for access to this item and much more
All Information Management content is archived after seven days.
Community members receive:
- All recent and archived articles
- Conference offers and updates
- A full menu of enewsletter options
- Web seminars, white papers, ebooks
Already have an account? Log In
Don't have an account? Register for Free Unlimited Access