Skip to main content
AI Helps You Read a Customer's Pandemic "Story"

My first post in this series posed the essential question, “When uncertainty abounds, how can banks and merchants use artificial intelligence (AI) to successfully reopen, and engage with customers to meet their changing needs?” This blog explains the importance and benefits of constantly scanning transaction data for new insights, as well as related privacy concerns.

Gaining a 360º View into Risk

The pandemic has driven a chaotic scattering on customer behaviors that is unique in modern consumer history, catalyzing a “new normal” driven not simply by financial aspects, but fear, uncertainty and doubt, as well. It’s important to know which customers are:

  • Getting sick
  • Losing their jobs
  • Getting new jobs
  • On mortgage waivers or payment holidays
  • On their way to better or harder times
  • Saving more
  • Suffering mental health issues
  • Having troubles feeding their families

Individual hardships are difficult to see through a typical banking or merchant lens.  Sensitivity and compassion are today’s new guide rails, but new analytics are needed to understand which actions customers are taking, or not taking, that may be risky or prudent. These highly personal decisions won’t be encapsulated in anything other than transactions and engagement with the rest of the payment ecosystem.

In fact, engagement with the payment ecosystem is often the only mode of interaction banks may have with their customers. Integrating credit card transaction data with debit, bill pay, person-to-person (P2P) payments, and mobile and online card-not-present (CNP) payments to gain a 360º customer view has long been a theme in fraud prevention; banks now are also realizing that the more they can connect different transaction silos, the better they can make meaningful offers to specific customers. 

New “Stories” Are Emerging

Connecting information of the customer and their relationships with the bank has never been more important. In my last post I talked about how FICO’s collaborative profiling technologies translate each transaction into an event or “word”, with all transactions past and current forming a figurative “story”. During COVID-19 in particular, as new transactions add new words to customers’ individual stories, banks can personalize their interactions as different customers:

  • Operate in “business as usual” mode, maintaining usual patterns while perhaps increasing discretionary spending on food delivery, for example, a hallmark of the pandemic
  • Show shifts to lower spending levels and to different merchants—for example, from Whole Foods and Costco to regular grocery stores
  • Spend less on travel and entertainment and more on health care and homecare
  • Pay down credit card balances and get their financial house in order while they are able.

Keeping a focus on understanding customers and their journeys, and personalizing interactions based on how they have changed, will help banks to keep positively engaged with customers whether their individual journeys get better or worse through the pandemic. Continuous evaluation, and the way that collaborative profiles use each new transaction to append a new word to each customer’s story, will provide the insight that banks desperately need.

Privacy Considerations Are Essential

Clearly, it’s of critical importance for banks to know their usage rights of customer data. If it’s only credit card transaction for fraud purposes, shifting spending patterns can signal fraudulent transactions that earlier this year would have been considered normal. If access is permitted to all data associated with the customer’s relationship with the bank, this information can be used to determine if mutating archetypes are likely open to continued engagement with products and services.

As banks and businesses try to find their footing during COVID reopening, any and all transaction data can used to help banks assist customers through this difficult time, while gauging customer-specific risk. Broad, AI-based analysis of a wide range of transaction data presents an excellent way to accomplish both objectives.

Keep current with my latest thoughts on AI, machine learning, fraud and more on Twitter @ScottZoldi.

related posts