BACKGROUND: American Century Investments (ACI) is a diversified provider of financial services including funds, diversified investment portfolios and services for specialized investment needs. ACI also provides flexible, full-service record keeping for 401(k) plans and managed separate accounts for major corporate and public defined pension plans. More than two million investors work with American Century to help meet their financial goals.
PLATFORMS: SPSS products run on Windows NT and an IBM IntelliStation with dual processors, 25GB of memory and 300 MHz.
PROBLEM SOLVED: ACI had extensive and detailed data about our customers and shareholders. This included transactional data on each account, as well as demographic and behavioral data purchased from data providers. Although we had all this data, we did not really know who our customers were and, therefore, how we could best increase their wealth. Because the financial services industry is so competitive, we knew we had to really understand our customers to keep them as ACI shareholders. The first step was to organize all relevant data in a data warehouse. The next step was to mine that data and get to really know our customers.
PRODUCT FUNCTIONALITY: Before we could begin mining our data we had to address a basic data quality issue: missing data. We used SPSS Missing Value Analysis to impute the missing values in our data warehouse. SPSS Missing Value Analysis analyzes the patterns and internal relationships within large sets of data and estimates those values with a great deal of certainty. For the actual data mining we used a wide range of techniques available in SPSS Base, AnswerTree, Clementine and GoldMiner. We used the Kohonen neural network in Clementine to find some general patterns in the data. We found that there were significant patterns, so we used regression to predict the score for each segment. We also used factor analysis, decision trees and monotonic regression to further refine our model. The result of our model was very detailed customer segments that revealed critical customer information.
PRODUCT STRENGTHS: SPSS truly offers a complete data mining solution. SPSS has everything you need for data mining from basic descriptive statistics to neural networks and decision trees. And, the products are all extremely user friendly and work right out of the box, with no programming required. Also, the report OLAP (on-line analytical processing) environment is excellent and provides us with a comprehensive solution to our business problems. As a result of our data mining with SPSS, we have gained a significant competitive advantage because we now better understand our customers and, thus, can help them meet their financial goals. As a result, they are much less likely to switch to another fund because they are more satisfied.
PRODUCT WEAKNESSES: There is not a client/server version of SPSS Base and modules, although one is planned in 1999.
SELECTION CRITERIA: We needed a product that was user friendly yet analytically powerful and that did not require extensive programming as many tools do. SPSS was really the only vendor that offered a total analytical solution and met all of our needs.
DELIVERABLES: For the first time, ACI has detailed customer profiles that reveal strategic information about our customers. We have identified segments and sub-segments that indicate such things as account balance and transaction history, as well as behavioral and demographic data such as what magazines our customers read, whether they regularly access the Internet, marital status and number of children. The SPSS analysis also produces database scores. We use these scores to target our direct mail campaigns, with the goal of only mailing to customers that would most likely be interested in the product we are promoting.
VENDOR SUPPORT: SPSS sales and technical support personnel are professional and accessible and really understand the business problems that we are facing.
DOCUMENTATION: The documentation, both on-line and printed, is superior and very user friendly.
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