DAS 6A: Creating a Prospect Segmentation Model Using RFM & K-Means Cluster Analysis

Track: Data Analytics Symposium

Session Number: DAS 6A
Date: Thu, Jul 27th, 2017
Time: 10:15 AM - 11:00 AM

Description:

Non-profit fund raisers, who are fortunate enough to have a database of potential donors, need tools to mine their database and a system to classify their prospects. RFM (Recency, Frequency, & Monetary Value) analysis is a tool that can be used to effectively mine the database, analyze prospect behavior and score potential prospects. The theoretical value of RFM analysis is based on the consumer marketing premise that the best predictor of future consumer behavior is past consumer behavior. The presenter will cover the design and implementation of an RFM model to score and rank potential prospects. The RFM data will be used to create a prospect segmentation model by using the K-Means Cluster algorithm. The segmentation model provides a tool to classify the prospect pool for future gift potential based on past behavior. Analysis and interpretation of results will be provided for the RFM model and the K-Means Cluster model.
Sub-Categorization: Start-Up Track
Session Type: Breakout Session (45 minutes)

Primary Competency: DA:Competency 4: Statistical Techniques and Competencies
Secondary Competency: RM:Competency 1: Prospect Pool/Base Analysis
Tertiary Competency: None
Intended Audience Level: Level II
Learning Objective #1: Attendees will learn how to review and interpret the results of a segmentation model.
Learning Objective #2: Attendees will learn how to use the K-Means Cluster algorithm to create a segmentation model.
Prerequisites: Basic understanding of terms generally used in statistical analysis, such as mean, standard deviation and frequency distribution.
Sub-Categorization: Start-Up Track
Session Type: Breakout Session (45 minutes)

Primary Competency: DA:Competency 4: Statistical Techniques and Competencies
Secondary Competency: RM:Competency 1: Prospect Pool/Base Analysis
Tertiary Competency: None
Intended Audience Level: Level II
Learning Objective #1: Attendees will learn how to review and interpret the results of a segmentation model.
Learning Objective #2: Attendees will learn how to use the K-Means Cluster algorithm to create a segmentation model.
Prerequisites: Basic understanding of terms generally used in statistical analysis, such as mean, standard deviation and frequency distribution.