Database Marketing • Six Basic Assumptions q Past is best predictor of future q Life Time Value (LTV) more important than any single transaction q Customers more important than noncustomers q All customers are not created equal q Customers can be grouped (segmented) based on their characteristics q Good prospects are like good current customers
Database Design and Maintenance There are two main types with different objectives and purposes.
Hierarchical vs Relational • Single record • Mulitiple records in contains ALL info • Excellent for high volume processing • Limitations on data detail and revision • Best used for operations data “Linked” tables • Improved analytical capabilities • “Infinite” detail data accumulation • Best used for marketing data See Exhibit 3 -2, page 52
Match Codes Allow Merge/Purge • Purpose: Faster processing to ID duplicates on two or more LARGE files • Example Address: James E. Coleman 1840 Wentworth Drive Montgomery, AL 36106 • Match Code: AL 6106 CLE 1840 J 75 • Explanation: AL: state 6106: last 4 zip code CLE: 1, 3 & 4 positions from last name 1840: first 4 street num. J: first initial 7: character count of last name 5: character count of first name
What To Do With Duplicates? • Eliminate from new lists so you don’t pay for names you already have • Eliminate from existing lists to prevent duplicate mailings • ID “multi-buyers”: people who buy from numerous DM companies are more likely to be good prospects
Database Maintenance Duties • Nixie removal q Undeliverable or “no forwarding address” • NCOA file comparison q National Change of Address from USPS • Routine updating of record status q Old data is expensive and misleading • Security q Would your competitors like to have this info?
Database Maintenance Duties • DMA mail & telephone preference list q DM companies should always drop these names from their lists. q Telephone Preference Service, or Direct Mail Preference Service Direct Marketing Association P. O. Box 9014 Farmingdale, NY 11735 -9014 q Email Preference Service: www. e-mps. org q FTC ‘Do Not Call’ list.
Customer Profiling • The better you ID current good customers, the better your prospecting will be. • Beware of the “equal effort over time dilemma” q Certain types of customers appear better because your efforts concentrated on them in the past.
Customer Profiling • Internal: if you already have RFM, demographic & psychographic data • External: SMRB reports & profiles • Enhancements: running your file against external demo/psycho DB’s • Environmental: census based Zip. Code info for consumers or SIC code data for businesses
Скачать презентацию Database Marketing Six Basic Assumptions q Past
7bef35c8bb42e45909d43db9949cdf81.ppt