Home / Banking Strategies / Customer profitability analysis for better decisions

Customer profitability analysis for better decisions

Share

Is your financial institution capable of answering questions related to the profitability of its customer base? For example, is the current service model for a specific customer segment too expensive when compared to its revenue-generating potential? Which specific types of loans should you stop originating given their low interest margin? When considering the full relationship of a customer, should the pricing terms of a personal loan be changed?

Customer base profitability needs to be fully understood to answer any of these strategic questions and it is within reach for all organizations. Decision making at all levels is significantly enhanced by having a robust customer profitability model, similar to using a GPS when driving a car. It represents a competitive advantage with meaningful short- and long-term returns.

Building an organization-wide customer profitability model is a complex project that may take several years to complete, particularly if the client data are scattered in multiple core systems and databases; there is disparity in the level of detail across products or segments; and the financial institution is not customer-centric. However, that should not prevent you from progressively deploying a flexible and practical company-wide model. To create this “navigation system” you may start by defining a calendar with realistic deployment expectations. To that end, we suggest:

Divide and conquer. Start with a specific product or business line (credit cards, installment loans, auto loans, etc.), preferably one that does not involve working with massive amounts of data. Add business lines as you go, stopping each time to assess progress.

Keep it simple. If a profitability software package is not within reach, know that you can use an application like Microsoft PowerPivot to build a very robust and flexible product profitability model at the account level, especially if the bank has solid account and transactional databases.

Fine tune. As soon as the first model prototype is built, slice and dice the information, fine tune the expense allocations as results are analyzed, incorporating relevant attributes like age and FICO scores. Then, start taking advantage of the many expected and unexpected profitability results.

Once the initial models are deployed, ensure an organization-wide visibility of results, particularly among business leaders. In our experience, the most successful deployments are supported by a progressive drive towards accountability at all levels of the financial institution. Start small to build buy-in. Once the organization understands the value of account level profitability in each product, the potential value of having the “360 degree” customer view will be obvious. The winning combination is the mix of analytic insights obtained from the customer profitability model with existing market knowledge and business sense.

Our work shows that once the models are built, common wisdom based on “traditional” views is sometimes confirmed but frequently dispelled. Here are a few case examples:

Wealthy customer segment. There is a general belief that the customers with the largest deposits or investment accounts are the largest net income generators. The first thing we need to mention in this regard is that customer wealth should be measured based on all the deposits and credit relationships with your competitors as well as with your organization, assuming this information is available. Even under a correct identification, these customers may or may not be the most profitable. In reality, they often show a low or even negative return on equity compared to other less “sexy” customer segments because they consume large resources through a personalized service model. In addition, their margins are lower as they receive the best interest rates in both credit and deposit products, and fees related to their credit cards and deposit accounts are often waived.

College-student deposit accounts. These accounts are usually offered with the objective of creating loyalty among customers that have the potential of generating material net income in the future. We have found that sometimes a good portion of these accounts are owned by customers who finished school a long time ago, generating a thin or even negative net income margin. These customers are good candidates for migrating to higher premium and more profitable deposit products.

Personal loans. In this business there is often a minimum loan amount below which profitability is negative, even marginally (the amount is different for each institution). In those cases, management may establish a minimum loan amount in their underwriting policy, charge a higher price or higher underwriting fees, or create an expedited and less expensive underwriting process.

Auto loans. In the analysis of a particular geographical market, we found a large concentration of net charge offs in loans collateralized by specific car models within each credit score range. Those car models seemed to be associated with a higher probability of default and loss given an event of default. The net losses generated by those customers could be avoided by incorporating tougher underwriting standards in loans collateralized by the identified car models.

Number of existing relationships. Our research shows that having additional bank relationships is a variable that explains higher profitability for some businesses but not for others. Understanding which relationships are more powerful in explaining customer profitability is key for deciding when to offer bundled products or apply additional discounts or special underwriting treatment.

As the effort of building the customer profitability model is deployed throughout the organization, it is important to have a consistent approach when going through the different businesses. Here are five key steps to follow:

  • Start with each bank’s product/business line statement of income and balance sheet.
  • Identify which revenue and expense items are available at the customer account level in the bank’s databases or operational systems. Direct those revenue and expense items to the accounts that generate them and ensure that variances against the financial reports are small or have a reasonable explanation. This process also has analytical, organizational and cultural impact.
  • Classify the expense structure in business activities and map the cost center/expense line combinations to the identified business activities.
  • Select the most appropriate cost allocation driver for each business activity in order to distribute the activity expense to product families and customer accounts.
  • Organize accounts at the customer level, run their income statement and incorporate the attributes that will help you explain the profitability.

When going through these steps, you will need to make numerous modeling decisions and overcome unanticipated roadblocks. To facilitate the journey and save some valuable time consider the following:

Determine the use you will give to the customer profitability model. For example, if you want to know if a specific customer segment adds value, you may need to include only variable expenses; while the segment may be unprofitable when you factor in overhead and other fixed expenses the organization would generate less net income without that customer segment. Similarly, the financial institution is better off with a new account that is only marginally profitable if market dynamics do not allow for a higher price.

On the other hand, you may need to include all expenses, variable and fixed, when looking at how the pricing strategy and credit policy of the different business lines impact overall profitability. If the current pricing does not allow the organization to reach its target profitability, the answer may be to: (a) increase price in segments with low price elasticity, (b) discontinue a product or business, (c) consider not serving a particular customer segment, or, (d) look for other efficiencies.

Require that leadership provides you with the methodology to allocate cost of money expenses across businesses. The model will require this information to provide a comprehensive view of the organization and its customers. However, attempting to define or reconstruct the methodology will be time-consuming and politically costly. By obtaining this information as a model input you will guarantee that this allocation is acceptable to the highest stakeholders.

Ensure flexibility in your model construction. Incorporate all revenue and expenses, even if some are not distributed to the accounts as part of the customer profitability (for example, ATM surcharge fees to non-customers). In this way, you will always be able to reconcile with the business lines’ financial reports. Also, for reserve-type expenses, build the model with two views to allow for the user to alternatively choose a provision expense view or an actual charge offs view at any point in time since one reconciles with accounting, and the other one shows the actual losses and product/customer behavior.

Focus on the details, spending the necessary time to reach more granularity in areas where expenses are material. For example, branch infrastructure costs usually represent a significant portion of the expenses associated with deposit products and customers and call centers might be particularly important for mortgage servicing, payment processing and collections efforts. In the absence of transactional data, you may need to acquire benchmark data or perform time studies in a statistically representative sample of branches to understand the activities performed and how they impact the profitability of specific deposit products and customers.

Mr. Divi is a senior banking practice expert at V2A, a strategic management consulting firm with offices in San Juan, Puerto Rico and Miami. He can be reached at [email protected].