Ultimate Customer Value Predictor: CxC Algorithm

posted Jun 3, 2010, 7:51 AM by Michael Hoffman

The Ultimate Customer Value Predictor: CxC Matrix Algorithm

 

CxC Customer Algorithm Details at Bottom 


Modeling and predictive analytics can answer any question. As a salesperson and a marketing strategist (that loves to help companies grow sales: create jobs, grow the economy…) I have always been in search of “What question has the greatest return on investment (ROI)?  

 

What question has the greatest ROI?

 

The Single Most Important Question in Business: Customer?

 

Predictive Analytics and Modeling Tools are becoming more main stream but with all these tools (and any technology), the value is derived from the scale of the problem to be solved, therefore the question is the most important part of the decision..

 

hat is the key variable in every business? Customer. Without customers there is no business. Every business expenditure should be measureable by its impact per customer.

 

What is most important question about customers? Will customer buy?

 

“Buy” or “not buy” is the type of binary question where modeling works best. But of course customers are more complex and cannot be effectively modeled or segmented based on the single data point, “buy”. So after analyzing customers for tens of business across financial services, packaged goods, publishing, communications, travel, retail, web, direct marketing and other industries it became clear that the customer process need to be modeled in order to manage a business to optimum profitability.

 



The CxC Matrix describes a customer’s movement across any company providing a means for collecting data and predicting customer movement from customer stage to customer stage using the CxC Algorithm

 

The Business Case

Most analysts and companies do not model their customer’s life cycle because of the perceived complexity, their inability to articulate the ROI and their uncertainty on how to use the findings.

 

The CxC Matrix Framework Matches Customers to Departments, Systems, Managers

The CxC Matrix provides the framework for visualizing, monitoring success and deploying the results of the CxC Algorithm. The algorithm helps companies predict how many customers will move from stage to stage and channel to channel across the customer life cycle.

 

 

Next Level Customer Predictive Analytics

The base CxC Matrix Algorithm provides a foundation where companies can score individual customers or customer segments for their likelihood to buy, defect, use customer service resources, total life time value, etc.   For companies using tools from SAS, IBM, Oracle, SAP, Tibco the CxC Algorithm and CxC Matrix provide not just tremendous business and strategic insight but dramatic business results measured by revenue growth, reduction in advertising, marketing and customer service expense.  

Integration Points

The CxC Matrix provides technology companies and business analyst a schematic of all the customer data integration points and can be used to plot business process management deployments as well as script, monitor and revise embedded analytics which will become more prevalent over the next 2-5 years.



 

An example or imbedded analytics in a contact is Amazon.com’s web page serving over 80 messages and functions based on a combination of customer information, historical performance information, resources and predicted outcomes as decribed in Customer Worthy, Why and How Everyone in Your Organization Must Think Like a Customer.

 

 


CxC Base Customer Algorithm Details



The above formulas give mathematical representation to the relationship of variables driving the Client X Client Matrix.  The above makes the assumption that all progress measured is forward progress, and no progress from later life cycle steps or lateral movement within a life cycle step is measured.

 

Formula (1) shows that the change in the number of customers in a given point of contact – the intersection of a channel (I) and a life cycle step (J) – is a function of the change in the numbers of customers (ΔN’s) in all the contact points (from 1 to the total number, a) in the prior life cycle step (J-1).  This relationship is presented in the form of a linear model that might result from a multiple regression analysis where the known quantities, the ΔN’s, are modified by coefficients (β’s) which need to be determined and are unique to each client and operation.  A single β may be large or small indicating the importance of its corresponding ΔN.  For this representation, it is assumed that the model is linear, i.e that none of ΔN’s are exponential, though this may not be the case in a particular situation.

 

Formula (2) is a simplification of formula (1), replacing the series of βΔN’s with the summation symbol.

 

Formula (3) shows the change in the number of customers at all points of contact within a given life cycle step (J).  This formula is simply the summation of formula2 (1, 2) calculated for each I from first (1) to the last (a).  Formula (4) is a simplification of formula (3), replacing the series of ΔN’s with the summation symbol.

The CxC Matrix and how to use it is explained thoroughly in Customer Worthy, How and Why Everyone in Your Organization Must Think Like a Customer, Michael R. Hoffman, Paramount market Publishing, February 2010 and available at Amazon.com.

Client X Client would like to speak to any companies interested in discussing applications, tools, deployment, consulting, and partnerships using the CxC Algorithm and the CxC Matrix. Please contact Michael R Hoffman at mrhoffman@clientxclient.com

Comments