With personalization being the buzzword of the year, we’ve all glossed over one important point: a personalized experience doesn’t necessarily mean a wholly unique experience – it’s a byproduct of sophisticated segmentation.
“If you and I buy the same books off of Amazon, we will both get targeted with the same book recommendation the next time we login. This recommendation is equally relevant for both of us, but in no way exclusive to either of us. It’s a personalized experience, even if that experience is the same for 100 other people,” explains Vic Moschitto, Customer Insight and Predictive Modelling expert.
Even with Amazon, personalization wouldn’t be possible without segmentation. But what Amazon does better than everyone else, is refining customer segments into tiny, or micro, clusters that share such unique commonalities, that they can be targeted with an offer that is relevant to every person in said segment. In doing so, the company is able to create personalized experiences at scale, build high-value relationships with customers, and ultimately earn long-term loyalty.
The need for better microsegmentation in banking
Most companies—including banks—segment to some extent, targeting different customers with different products via different channels. But while most industries are moving towards grouping customers based on deeper behavioral information, banks are still segmenting based on demographic factors, financial parameters, income, and geographic location.
Experts at EY point out that, “while important, such basic information usually has a weak correlation with clients’ needs and results in simplistic segmentation. Basic segmentation is therefore inadequate for banks to develop the nuanced insights necessary to understand customers.”1
The goal of segmentation is to decide how to relate to customers in order to maximize the value of each customer. But if segments are large and broad, companies end up relating to customers in broad and imprecise ways.
What’s more, historically, the bank’s approach to segmentation has been designed to support a person-based sales force, either branch staff or contact centre staff.
“When you’re training a salesforce of 1000 people to deliver a consistent customer experience, you can’t expect them to remember hundreds of different segments. So traditionally, banks bucketed customers together into broad segments that were easier for their salesforce to remember. The personalization bit came from the human interaction with the client,” explains Moschitto.
But relying on human interactions to deliver a personalized experience isn’t scalable, or relevant for the modern consumer who prefers a digital experience. As more of the bank’s services are being delivered via digital channels, it’s becoming increasingly critical that banks shift their approach to segmentation so that it complements a mobile-first experience. Luckily for them, technology is far better than a human salesforce at remembering hundreds of segments.
Simply put, moving from broad-based segments to microsegments is essential for banks in their quest for digitally personalized experiences. Since demographic data is too generic, banks need to focus on acquiring more dynamic and accurate data. One approach would be to focus on understanding customer psychographics – lifestyles, preferences, and traits.
Segmentation based on spending behavior
“Probably the greatest potential of data monetization comes from merging cardholder data with data from the merchant side to gain an end-to-end view on transactions that can unlock additional value.” (McKinsey, 2017)
What people buy and how they pay for things is directly tied to personality traits, lifestyles, attitudes, and expectations, and could be used with accuracy to predict and fulfill customer needs.
Based on an ongoing US Consumer Financial Life Survey, McKinsey published the report “New Frontiers in Credit Card Segmentation” to validate how information about credit card spend and POS payment preferences are comprehensive for creating customer typologies.
For example, traits like “Sticks to a budget”, “Card hops or transfers balances to keep rates & fees low”, and “Believes issuers want customers to accumulate debt” can all be gleaned from these payment behaviors. Notably, these financial traits correspond to underlying needs that issuers can address – whether that means needing interaction with human staff to assuage them about credit card debt, or offering cards with interest fees and rewards to match their spend.
With insight into merely a fraction of payment activity—that is, credit card purchase volumes and most-used instruments for POS payments—McKinsey discovered nuanced segments, and proposed a way for card issuers to fine-tune offers to meet distinct customer needs.
Though McKinsey makes a good case for the power of payment behavior, their survey only captures a component of the customer buying journey. To create refined clusters, or microsegments, institutions need to understand the entire customer buying journey—from motivations to buy, down to the items purchased.
Smarter insights with receipt data
Accurate insight into customers’ financial needs is only made possible with insight into what existing customers are actually buying product-wise, and an understanding of how these purchases change over time.
Understanding payment down to the SKU-level would also help banks better understand customers’ profiles beyond generic profitability indicators. For example, a customer’s SKU-level information might reveal a large proportion of items purchased on sale, suggesting price sensitivity. This could come as a surprise if profitability indicators, such as high purchase volumes on credit cards, originally suggested otherwise.
The McKinsey report also reminds us that financial institutions rely on third-parties to acquire qualitative and quantitative data about their customers, which serve as good benchmarks of customer behavior, but can’t be attributed to any specific customer at the bank. Banks must therefore consider the implications of relying on externally sourced data to understand customer needs and personalize banking experiences.
Financial institutions can capture the greatest value and insight by holding their data internally. Once banks recognize this, deriving the purchase insights they need is a two-part process. Banks must first acquire sources of purchase data, either by institutionalizing receipts in their banking apps, or by acquiring receipt data from merchant partners and/or POS providers. They must then recruit existing customers and combine qualitative research to uncover how lifestyle, needs, and attitudes are mapped to specific purchases. This will inform how they design their segmentation strategy, and provide the details they need to create nuanced microsegments that deviate from traditionally generic segments.
Purchase data on receipts are tied to a specific individual, not an average, or a summary, from a market research study. This means the insights from receipts are highly actionable. Armed with this–and only then–will financial institutions be able to understand true spending behavior and map it to psychographics like lifestyles, values and risk aversion. After all, people are more than a dollar sum of their purchases.
- McKinsey. “Monetizing data: A new source of value in payments”. https://www.mckinsey.com/industries/financial-services/our-insights/monetizing-data-a-new-source-of-value-in-payments. September 2017.
- EY. “How well do you know your customers?”. http://www.ey.com/Publication/vwLUAssets/EY-understanding-the-key-to-customer-loyalty-report/$FILE/EY-understanding-the-key-to-customer-loyalty-report.pdf. 2016.
- McKinsey. “New Frontiers in Credit Card Segmentation: Tapping Unmet Consumer Needs”. 2014.