Profitability analysis: Rearview mirrors, windshields and gas pedals
Profitability analysis in banking and financial services delivery is critical, and as the old adage attributed to the scholar Dr. W. Edwards Deming goes, "If you can't measure it, you can't manage it." Is that really what he meant, though? According to the W. Edwards Deming Institute, Deming's often misinterpreted statements on this subject were really meant to say that there are many things of importance to management that cannot yet be measured, but they nonetheless need to be managed. It is crucial to use data and analysis to evaluate whether those decisions are achieving their intended results.
Remember this perceptive advice
Financial services organizations and banks have long used the structured "Plan, Do, Study, Act" approach espoused by Deming to develop and deliver new products and to target particular customer segments. In other words, and admittedly at risk of oversimplifying the process, they start with a goal based on a hypothesis, introduce the product, analyze the impact of the delivery and, based on the analysis, adjust as necessary. Then, they go through the cycle again. Profitability analysis is a key component of the "Study" phase, which then informs the "Act" part of the process.
Product profitability analysis commonly takes into account revenues earned from products and services, as well as factors such as:
- Funds transfer pricing.
- Operational costs assignable to the products used by customers.
- Economic risk of the products.
- Necessary provisions applicable to the products.
- Capital required to support the product.
The aggregate of the profitability of the products and services used by a customer leads to customer profitability information. From there, customer household and customer segment profitability analysis is possible.
The historically focused analysis above has evolved into a predictive approach where financial services organizations use customer lifetime value analysis to predict the profitability of a customer over his or her lifetime of using the bank's products and services. If the provider wanted to know the profitability of a particular customer over a two-year period, it could use a historical approach and collect profitability data over those two years, or it can predict what will be over the two years. Of course, the value of predictive modeling is only as good as the models. Big data and analytics have a clear role here.
Pedal to the metal
Speed is an important element to competitive success. What organizations may not fully appreciate is that the ability to deal with high-velocity data will be an extremely important competency in the very near future, as noted by the Harvard Business Review. Being able to adapt in real time to customer actions and preferences that match or contradict predictions will help businesses serve customers appropriately and tweak predictive models on the fly.
Imagine Deming's model condensed to two stages: Act/Plan and Do/Study. The lines between acting and planning blur when the decision-support systems that are geared toward predicting, doing and studying become one, thereby providing real-time feedback and decision system modification. While it's debatable whether big data and analytics will ever provide the means to measure absolutely everything that is important to management decision making, it can clearly help minimize the amount of unknowns in the "Plan" phase, as well as minimize the wait in the "Study" phase. Using all relevant historical, structured and unstructured data about products and customers and employing the appropriate analytic techniques allows organizations to continually and more accurately predict future outcomes with respect to customer profitability analysis.
Perform complex profitability analysis to identify your most profitable products, customers and sales channels. Visit the IBM Financial Performance Management webpage, and sign up for a free trial today.