Point of View

Auto finance analytics: Keeping pace with changing customer behavior

Print This Page

The auto finance landscape is becoming increasingly competitive, with more buyers arriving at showrooms (or used car lots) with pre-approved credit from direct lenders. Dealers find it challenging to present truly competitive offers. Those who do so by leveraging analytics to build risk-based pricing models are able to widen margins while accelerating loan origination.

Data-driven tools also help lenders streamline regulatory compliance, improve fraud detection, reduce process failure rates, enhance customer communications, and allocate capital more strategically. Specialized operating models for such analytics practices enable their cost-effective deployment—at scale.


To succeed in today’s market, auto finance providers must adapt to changing customer behavior and volatile economic conditions. Not only have consumers become more cautious, but they have access to a wider range of financing options than before as lenders from slower-growing consumer credit segments move into the market. A growing proportion of customers use the internet to scout for the best deals, including financing incentives. The competition has become progressively more intense, and with more buyers arriving at the showroom or used car lot with pre-approved credit from direct lenders, dealers are challenged to present competitive offers.

The best-performing auto finance companies meet these challenges through more intelligent business processes at every stage of the product cycle. They apply leading-edge analytics to build risk-based pricing models that widen margins while accelerating loan origination. They create more agile systems for directing account management resources toward high-risk customers. And they use smarter collections and recovery strategies that cut default rates. As a result, the most successful firms find new opportunities in a demanding marketplace while making the most of the portfolios they already have.

Take a copy for yourself

Download PDF

As the market has become less predictable, many auto finance providers are finding that traditional statistical models—based mainly on credit bureau data—are no longer adequate. With more lenders chasing every deal, risk organizations must work quickly to make intelligent acquisition decisions. They also need enhanced vigilance to manage booked portfolios in the face of increased risks from default, early closures, repossession, skips, and fraud. Moreover, they must constantly monitor key performance metrics to ensure that investments in improved processes pay off. Increasingly, they are turning to advanced analytics to meet all these needs. While advanced analytics offer considerable promise for more efficient auto finance operations, off-the-shelf solutions cannot meet the unique needs of individual businesses. Modeling techniques and data sources must be customized for each application.

Aligning pricing with risk—fast and at scale

In today’s aggressively competitive environment, lenders need to rapidly generate accurate risk-based quotations to engage qualified customers while ensuring that margins for other loans offset the greater risk. The first step in optimizing underwriting performance is to stratify the entire portfolio into risk bands for differential pricing. Simulation tools can take this a step further by generating alternative pricing scenarios for each tier while projecting each scenario’s impact on key performance metrics as well as broader business objectives. By integrating this approach into its overall strategy, a US-based bank widened its spread by 6% while improving Return On Equity (ROE) by 35%.

Tapping into additional data for risk management

Credit bureau data is most effective in credit decision models, and while it is also useful in post-purchase account management, other types of data can enhance lenders’ ability to predict customer behavior. Accurate differentiation between customers experiencing transient liquidity issues and customers on the road to default enables greater efficiency in resource allocation while fostering productive relationships with good clients. Effective deployment of these analytical techniques can yield remarkable results: a global auto finance organization reduced annual losses by close to US$1 million and increased forward rollover accounts by 4% by integrating current and historical employment data into its predictive models.

Enhancing risk mitigation by predictive score-based strategies

Advanced predictive modeling transcends traditional Probability of Default (PD) and Loss Given Default (LGD) models, enabling more precise targeting of strategic actions through the auto finance life cycle. The predictive modeling concept can be used to identify high-risk customers, select a welcome call strategy to optimize collections, and implement a data-based extension–renewals– reposition decision to maximize recovery by implementing smart strategies. Sophisticated models can provide customized, evidence-based profiles of bad debt by portfolio segment, allowing more accurate identification of accounts with a high propensity to default following an extension. For example, an auto finance major implemented customized models to enhance its identification and targeting abilities, with incremental improvements ranging from 7% to 10% for various segments— culminating in savings of nearly $7 million for the fiscal year. In another case, an auto finance company refined its depreciation in its collateral model for charge-off to reflect current auction prices, realizing a substantial business impact.

Actionable insights for senior management

The cases described in this analysis demonstrate how business process owners are using increasingly sophisticated, data-driven tools to enhance the performance of their lending functions. Other improvements driven by advanced analytics include streamlined regulatory compliance, improved fraud detection, reduced process failure rates, better customer communications, and more efficient capital allocation strategies. These techniques also support the more effective execution of broader corporate strategies through the delivery of actionable insights to senior decision-makers. For example, an integrated framework would ideally include:

  • Continuous monitoring of key performance metrics
  • Scorecards or dashboards to provide senior management with visibility into operations
  • A model governance framework to ensure the alignment and ongoing re-calibration of data-driven processes across the business

There is no one-size-fits-all solution for prosperity in the current auto lending market. But rigorous reassessment of existing processes and systems to identify weaknesses and to guide reengineering to align with industry best practices is a good start. Advanced analytics provide powerful tools for turning objectives into actions and achieving true business impact. At the same time, to achieve these results cost-effectively, it is important to adopt the right operating models for such processes, often by leveraging shared and scalable analytics resources that are located globally. These strategies will play a growing role in determining tomorrow’s industry leaders.

As a pioneer in the analytics industry, Genpact has a deep understanding of how data management tools can help businesses meet the specialized challenges of the current risk landscape. The article above provides some practical examples drawn from Genpact clients in the auto finance industry.

For more information, contact, analytics.marketing@genpact.com and visit, genpact.com/what-we-do/capabilities/analytics/financial-services-analytics

Continue Reading

Ready to
Explore?