I am looking forward to tomorrow’s SAS Media Day here on the Cary campus because I get to meet new folks and also reconnect with some familiar faces. One of the speakers tomorrow for the SAS Media Day is Rex Pruitt with PREMIER BankCard who was a speaker at last week’s
Predictive Analytics World.
Pruitt’s presentation last week
The Development of a "Good Customer Score" for Use in Customer Acquisition, Rewards, Retention and Recovery drove home the value of testing new scoring models before production – this can be used as a proof point with senior executives especially when the value, in this case Good Customer Score (GCS), can potentially be debated or management is risk adverse to adopting a new scoring model.
PREMIER Bankcard, based in South Dakota and assets of $900 million, provides traditional and non-traditional banking services.
Recognizing that integration of predictive analytics into the
portfolio management strategy could equate to huge returns in a time of economic flux, Pruitt began looking at what was working in the portfolio management strategy and the opportunities. PREMIER had a good grip on seasonality and
vintage analysis (how particular segment performs over time). However Pruitt saw an opportunity to develop an improved Good Customer Score which would help with several PREMIER objectives (revenue, loyalty, delinquency).
The development of Good Customer Score to apply to retention, acquisition, cross/upsell efforts was key. Pruitt said there were different definitions of “good” customer attributes within PREMIER. From a financial executive perspective, the good customer generates increased revenue with minimum expense. The operations executive defined the good customer as one with good behavior and loyalty. Pruitt interviewed executives and
blended the characteristics to target data attributes that describe a “good” customer.
After thorough data integration efforts with internal and external sources, cleansing and adjusting for outliers, Pruitt and team began the weighting process and developing the GCS. Pruitt found that there was some reluctance to adopting the GCS in certain layers of management but when the opportunity arose to share the project with the executive board the response was “Test it!”.
The willingness to test and proceed to see what the Good Customer Score could do is a cultural best practice.
I was so excited to hear that the response of management was to test! At lunch this
same day at Predictive Analytics World someone asked what could they do to influence management to understand the value of analytics? My response – look at the PREMIER Bankcard example from Pruitt – TEST and show them the revenue they are leaving in the data Start with a project where you can show the lift, improved return – especially if dealing with risk adverse management.
What are Pruitt’s findings in the testing phase for the Good Customer Score? Quite impressive in the first six months! The GCS is projected to generate $15 million in annual revenue. In cross sell program testing using the new GCS identified 2% increase in qualified candidates and a $2.3 million projected lift from the cross sell campaign. Response rates from direct mail? GCS testing is showing response rates over 11% which really kicks the usual 5% response rate if you ask me.
This was one of many sessions at Predictive Analytics World that I’ll be sharing some tidbits from over the next week or so. Now I need to get ready for tomorrow’s SAS Media Day – happy Monday to you all!
As Business Intelligence Product Marketing Manager for SAS, I’m responsible for the Global message direction, strategy and marketing of SAS business intelligence. More on me. About the pot pie, I'm a southern girl, and that's comfort food. But food for thought -- I can explain.