Earnix Blog > Pricing
A New Approach to Risk-Based Pricing: The Time is Now
Earnix Team
February 14, 2022
- Pricing
By applying risk-based pricing with precision, banks can broaden their customer bases, improve their profitability, and offer an enhanced degree of personalization to drive higher customer satisfaction and a better overall customer experience (CX). All of which results in market advantage in an increasingly competitive world.
The analysts at Bain might have said it best: “Smarter pricing helps to optimize yields, manage the cost of funding, gain market share, do right by the customer and manage risks.
The problems with this approach have become more and more apparent over time. In fact, time itself is the problem. Credit scoring relies on parameters that have notoriously long time lags – changes in income, repayment performance on other loan products, and changes in demographics, to name just a few.
Credit scoring is inherently about history attempting to predict the future, and without more modern tools it is a blunt instrument at best. In particular, credit scoring falls short of the personalization and timely offer-making that customers have come to expect from other businesses, such as online retail.
When decisions seem arbitrary, consumers who have legitimate reasons to feel they deserve better terms, must plead their cases to loan officers in inefficient and costly exception processes. The bank also risks making offers to riskier borrowers whose credit histories may not have caught up with their changing circumstances, subjecting the bank to unnecessary risk.
Similarly, statistical modeling attempts to take behaviors exhibited by large swaths of the population and apply that to decisions about a specific borrower. This rules-based approach cannot detect nuances in individual circumstances, suffers many of the same time lags as credit scoring, and cannot adequately account for large and unexpected economic shifts, such as recessions and other global disruptions.
Rules-based systems “run out of gas” when the decision-making becomes complex and/or there are large volumes of data to be considered. In other words, in situations such as risk-based pricing in banking, where it’s difficult to define the rules in advance, when decision-making needs to change continuously, and when banks may have millions of records to work with.
With all these shortcomings, it’s no wonder that banks are looking for a better way of pricing their offerings.
Quantifiable improvements in bank pricing software are being driven by artificial intelligence (AI) and machine learning (ML).
Systems based on AI and ML are essentially infinitely expandable, and their accuracy grows with time and with the growth of applicable data sets. In fact, the combination allows decisioning systems to “learn”, becoming more accurate over time.
Even with large data sets, AI and ML make it possible to identify patterns and apply judgment to decision-making, leading to greater personalization and quicker turnaround time, both critical to the customer experience. Customers and prospects feel that offers are crafted just for them, while the bank avoids the costs and risks inherent in earlier decisioning methodologies.
Many may have even attempted to implement risk-based pricing in the past, only to be disappointed by less than stellar results, high development costs, and a long tail of ongoing expenses for model refreshes, software maintenance and updating, and dealing with unpredictable and changing interactions with other bank systems.
Banking Software-as-a-Service (SaaS) eliminates these problems. SaaS banking software offers an unmatched combination of speed and flexibility. Implementations can be completed in weeks or months, rather than the years often required of in-house/on-premise development and maintenance. Updates occur continuously, and scaling these systems is automatic and seamless.
Choosing the right SaaS banking software solution can also pay dividends in other areas.
Support for application programming interfaces (APIs) will allow your new software to peacefully coexist with legacy systems, and avoid a “rip and replace” implementation. As the wave of banking mergers and acquisitions (M&A) continues, the speed and efficiency of bringing together expanded bank organizations won’t be limited by technology choices.
With this approach, it’s also possible to employ proven risk models augmented by new AI- and ML-based predictive capabilities.
A leading Australian auto finance company utilizes this hybrid approach. By combining their well-established financial risk modeling with the ML-based predictions from its Earnix solution, this market leader is able to predict consumer acceptance of various price points and loan terms, to stay ahead of the competition.
Financial performance has also shown dramatic increases: this organization’s return on assets (ROA) has risen over 10% through this Earnix-driven approach, and is expected to generate millions of dollars in new revenue annually.
Summary
Banking is going to continue to become more competitive. Risk-based pricing is a key component to win and retain customers and to keep them expanding their business with the bank. The modern approach, utilizing AI and ML, and deployed as a SaaS/cloud-based solution, is critically important to realizing the full potential of risk-based pricing.
From eBooks and blogs to videos and more, discover insightful Banking content right here.
The analysts at Bain might have said it best: “Smarter pricing helps to optimize yields, manage the cost of funding, gain market share, do right by the customer and manage risks.
Risk-Based Pricing is Nothing New Under the Sun
Risk-based pricing has been in wide use in banking for over 30 years. In this historical paradigm, lending decisions have been driven by two factors – credit scoring for the individual borrower, and statistical models based on the behavior of large numbers of consumers over time.The problems with this approach have become more and more apparent over time. In fact, time itself is the problem. Credit scoring relies on parameters that have notoriously long time lags – changes in income, repayment performance on other loan products, and changes in demographics, to name just a few.
Credit scoring is inherently about history attempting to predict the future, and without more modern tools it is a blunt instrument at best. In particular, credit scoring falls short of the personalization and timely offer-making that customers have come to expect from other businesses, such as online retail.
When decisions seem arbitrary, consumers who have legitimate reasons to feel they deserve better terms, must plead their cases to loan officers in inefficient and costly exception processes. The bank also risks making offers to riskier borrowers whose credit histories may not have caught up with their changing circumstances, subjecting the bank to unnecessary risk.
Similarly, statistical modeling attempts to take behaviors exhibited by large swaths of the population and apply that to decisions about a specific borrower. This rules-based approach cannot detect nuances in individual circumstances, suffers many of the same time lags as credit scoring, and cannot adequately account for large and unexpected economic shifts, such as recessions and other global disruptions.
Rules-based systems “run out of gas” when the decision-making becomes complex and/or there are large volumes of data to be considered. In other words, in situations such as risk-based pricing in banking, where it’s difficult to define the rules in advance, when decision-making needs to change continuously, and when banks may have millions of records to work with.
A 21st Century Technology Approach to Risk-Based Pricing
With all these shortcomings, it’s no wonder that banks are looking for a better way of pricing their offerings.
Quantifiable improvements in bank pricing software are being driven by artificial intelligence (AI) and machine learning (ML).
Systems based on AI and ML are essentially infinitely expandable, and their accuracy grows with time and with the growth of applicable data sets. In fact, the combination allows decisioning systems to “learn”, becoming more accurate over time.
Even with large data sets, AI and ML make it possible to identify patterns and apply judgment to decision-making, leading to greater personalization and quicker turnaround time, both critical to the customer experience. Customers and prospects feel that offers are crafted just for them, while the bank avoids the costs and risks inherent in earlier decisioning methodologies.
Rapid Implementation and Flexibility are Key
The first reaction among many banking executives, pricing teams, underwriters, and product managers is that implementing what sounds like very sophisticated risk-based pricing software must be a long, expensive, and time-consuming process.Many may have even attempted to implement risk-based pricing in the past, only to be disappointed by less than stellar results, high development costs, and a long tail of ongoing expenses for model refreshes, software maintenance and updating, and dealing with unpredictable and changing interactions with other bank systems.
Banking Software-as-a-Service (SaaS) eliminates these problems. SaaS banking software offers an unmatched combination of speed and flexibility. Implementations can be completed in weeks or months, rather than the years often required of in-house/on-premise development and maintenance. Updates occur continuously, and scaling these systems is automatic and seamless.
Other Benefits of Risk-Based Pricing
Choosing the right SaaS banking software solution can also pay dividends in other areas.
Support for application programming interfaces (APIs) will allow your new software to peacefully coexist with legacy systems, and avoid a “rip and replace” implementation. As the wave of banking mergers and acquisitions (M&A) continues, the speed and efficiency of bringing together expanded bank organizations won’t be limited by technology choices.
With this approach, it’s also possible to employ proven risk models augmented by new AI- and ML-based predictive capabilities.
A leading Australian auto finance company utilizes this hybrid approach. By combining their well-established financial risk modeling with the ML-based predictions from its Earnix solution, this market leader is able to predict consumer acceptance of various price points and loan terms, to stay ahead of the competition.
Financial performance has also shown dramatic increases: this organization’s return on assets (ROA) has risen over 10% through this Earnix-driven approach, and is expected to generate millions of dollars in new revenue annually.
Summary
Banking is going to continue to become more competitive. Risk-based pricing is a key component to win and retain customers and to keep them expanding their business with the bank. The modern approach, utilizing AI and ML, and deployed as a SaaS/cloud-based solution, is critically important to realizing the full potential of risk-based pricing.
From eBooks and blogs to videos and more, discover insightful Banking content right here.
Share article: