Four Segmentation Models for Increased Sales, Deeper Relationships and Stronger Retention

Personalization is a marketing imperative. But it is virtually impossible without segmentation, or the grouping of members/customers with shared characteristics. Successful segmentation allows an institution’s understanding of members/customers to shine through marketing. It lets members/customers feel special and appreciated and is one of the primary retention drivers. Did you know 56% of consumers feel an increased loyalty to brands who understand and act on their personal preferences, priorities and differences?

Defining Segmentation Models

It can be confusing finding segmentation models that apply to financial institutions. Unlike the retail world, a returned product does not present an opportunity to delight the consumer. A closed checking or savings account could indicate that household will not be interested in future products and services. Financial institutions require more focused segmentation models.

When defining segmentation models, financial institutions need to consider opportunity and risk factors, the ability to cross-sell and the likelihood of account closure or balance diminishment. When you take into consideration the vast amount of data available to financial institutions, segmentation can deliver a deep understanding of your members/customers. The following segmentation models enable financial institutions to create winning campaigns founded on personalization. The right message is delivered to the right household at the right time.

Segmentation Model 1: Value Scoring

Value Scoring is an analytical approach that leverages information such as profitability, balances, tenure and product mix to help identify members/customers that drive value. Value Scoring allows you to rank households based on the value they bring to your institution, then compares and contrasts households based on profit, balances, tenure and number of unique products.

This is extremely helpful since losing one major household requires adding eight new average households to make up for the loss. By determining your most valuable members/customers, you can use the Value Score to guide your marketing strategies and nurture those top relationships.

Segmentation Model 2: Lifestage

To determine Lifestage, this model leverages demographic ingredients to provide further visibility into the member/customer based on their financial lifestage. After all, a college student has different needs than new parents. This model gives you the data you need to create campaigns targeted at members within various lifestages and their propensities for having a baby, taking a vacation or paying for a wedding. It enables greater insight on buying activities and behavior. This, in turn, helps craft solid member/customer profiles to inform and influence marketing campaigns.

Segmentation Model 3: Look-alike

Look-alike segmentation learns from those who engage, and finds those who fit a similar profile as the performers. Once you have your customer/member profile, you can evaluate it to find those who fit a similar pattern. For instance, Mary is a 40-year-old mid-income, married female with two children. One is 16. She recently took out an auto loan and each year she establishes a vacation savings fund. Kay is also a 40-year-old mid-income, married female with two children. But she doesn’t have an auto or personal loan. As Mary’s look-alike, offering her an auto or personal loan makes more sense than a credit card offer.

Profiling consumers based on a household’s relationship, lifestage and demographic data allows you to define target audiences based on those attributes and group them together. Then, your team can create offers that the group has a propensity for and potentially bump them into a higher value group.

Segmentation Model 4: Next Product

This is where art meets science, leveraging many of the aspects of the other segmentation models and is best used for point-of-sale channels. Purchasing patterns exist. What’s a hamburger without fries? Pizza without antacid? Analyze your data to determine who buys a specific product, then determine which products they are likely to buy next. For instance, an auto loan can be easily tied to opening a checking or savings account to expedite monthly loan payments.

Go forth and segment.

Once the data is gathered, it’s time to put it in action. For over 30 years, Marquis has helped financial institutions across the country create winning marketing campaigns with measurable ROIs. We focus on adapting proven marketing strategies to the specialized needs of banks and credit unions and have developed a three-step process for leveraging marketing segmentation.

Assemble: Leverage available data sources to identify which variables best select your target audience.

Analyze: Group data sources into segments to simplify your tactical options.

Act: Leverage automation and repeatable processes to act on the segments identified, creating more granular options based on member/customer personalization, including channel preference, product preference, tailored offers and much more.

Putting it all together.

Financial institutions have access to large quantities of data, and that data needs to be segmented, analyzed and used to create intuitive and relevant marketing messages. When segmented properly, it will elevate overall marketing results, allowing you to retain and upsell your members/customers while maintaining their loyalty.

You’ve got the data. You’ve got the strategy. Let Marquis help you put it into action!

*ABA endorses ExecuTrax and OnTrax data analytics solutions for marketing and business intelligence.

HMDA and Public Access to New Data

How HMDA data and increased transparency can affect fair lending.

HMDA submission season is just around the corner and your institution’s data will be under close scrutiny by more than regulators. Litigators, advocates and the general public can view the data and possibly use it to identify institutions at fair lending risk. But since HMDA data alone is not enough, this can lead to misinterpretation, unwarranted accusations and loss of reputation. To help mitigate these issues, maintaining HMDA data integrity is essential.

The Home Mortgage Disclosure Act (HMDA) was created to enhance the monitoring of lending patterns and to ensure financing needs are met across a diverse field of potential borrowers. Submitting loan origination and application data on borrower demographics and loan features enables enforcement agencies to identify financial institutions who excel at fair lending and those that require further investigation. In order to accommodate that goal, new data points were added in hopes to further keep biases in check and reduce barriers to homeownership for protected classes.

The new data delivers a deeper understanding of institutional borrowing practices. Regulatory agencies can now apply comprehensive data screening, data monitoring and statistical modeling routines across all lenders subject to HMDA reporting requirements. In addition, many of the new HMDA data fields, like age, credit score and debt-to-loan ratio, can be used for more effective identification of institutions with elevated potentials of fair lending risks.

With the release of the new data, 2020 is the first time members of the public will have greater access to some of the key determinants of underwriting and pricing decisions. Be assured, litigators and advocacy groups will be taking a close look for any sign of unfair practices. Since disparities are estimated after a broader range of pricing and underwriting factors are applied, litigators can present more credible fair lending cases that on the surface appear to be true than with previous HMDA data sets. Furthermore, journalists will also have access to the data, possibly increasing marketing and reputational risks.

Peer analysis also benefits from the new data. Because it is accumulated from all covered financial institutions, it is particularly helpful for defining local and national benchmarks. Peer comparisons can be expanded beyond penetration rates in minority census tracts to include APR, total loan costs, product features and so on. A clearer picture is presented, allowing regulators to more accurately compare benchmarks and identify institutions with elevated fair lending risks.

With more public access to HMDA data, regulators advise caution when interpreting this data, especially if it leads to accusations or conclusions of discrimination. According to a FFIEC Press Release, “HMDA data alone cannot be used to determine whether a lender is complying with fair lending laws. The data do not include some legitimate credit risk considerations for loan approval and loan pricing decisions. Therefore, when regulators conduct fair lending examinations, they analyze additional information before reaching a determination about an institution’s compliance with fair lending laws.”

In today’s world, businesses rise and fall on the whims of public perception. An unsubstantiated claim of discriminatory lending practices based on misinterpreted data could have far-reaching consequences. What can financial institutions do to protect themselves? Understand your data, especially when underwriting and pricing decisions can create and identify disparities. Realize how your data can be interpreted by public regulators, advocacy groups, journalists and litigators. And then be prepared to tell your story and/or present the corrective and preventive actions taken.

The only way to minimize or eliminate risk is to consistently monitor and analyze your own data for pricing, underwriting and redlining risk. Keeping data clean and relevant is essential for accurate interpretation. In addition, separate assessments should be conducted to identify possible anomalies generated by the expanded data fields. This can be an intensive undertaking. Automated compliance software for HMDA reporting will help ensure data accuracy. At the same time, it will help identify fair lending risk points in the application and origination process. When combined with analysis and interpretation, you should be able to identify any additional risk factors.

Marquis can provide a turnkey solution when combining industry-leading tools like CenTrax NEXT compliance software with the experienced and intuitive skills of the Marquis Compliance Professional Services experts. These services can make a great difference in your HMDA reporting process by regularly monitoring and cleaning your data and then helping you understand the HMDA Integrity Analysis. With cleaner data and a deeper understanding of how it can be interpreted, your institution will be better able to respond when your HMDA data is used by regulators and the public to evaluate fair lending risks.