How Alternative Data is Changing Credit Assessment Beyond FICO: Nitesh Khullar's Insights
For millions of Americans, access to credit has long hinged on a single number: their FICO score. But this traditional gateway to loans and mortgages is now facing unprecedented competition. A significant shift is underway, driven by the recognition that a single credit score often fails to capture a person’s true financial responsibility.
With the recent adoption of machine learning and big data and consumers’ ever-expanding digital footprints, traditional credit scores like FICO are no longer the only game in town. Instead, a new paradigm is emerging: alternative data, a sophisticated approach to credit assessment that promises to transform how lenders evaluate potential borrowers.
Nitesh Khullar, an executive with over 14 years of experience in banking and financial services, has been built on understanding risk. His work with Citibank has placed him as one of the few privy to how alternative data is reshaping credit models from the backend.
“The old ways of assessing credit just don’t cut it anymore,” he says. “We’re living in a time where you can’t rely solely on historical data to predict future behavior.”
Alternative data refers to information outside traditional credit reporting sources like FICO. While conventional credit scoring relies heavily on historical loan performance, credit card usage, and formal banking relationships, alternative data encompasses a broader spectrum of financial behaviors and patterns. This could include utility payments, rent history, social media activity, and even online shopping habits.
This expanded view of creditworthiness has particular significance for historically underserved populations. This type of data can be the only type available for many consumers—especially those who are underbanked or have limited credit histories. In fact, according to a 2023 report by the World Bank, around 1.4 billion adults globally remain unbanked. Many individuals may have never taken out a loan or applied for a credit card, but they still pay their bills on time and manage their finances responsibly.
Khullar explains how this view helps lenders make more secure decisions, “We’re no longer looking at just whether someone missed a payment five years ago. We’re asking questions like: How does this person manage their day-to-day expenses? Are they paying their rent on time? Do they have stable employment? These can give us a clearer picture of someone’s financial health.”
Pioneering Change: Nitesh Khullar’s Impact
Khullar’s work in this field has been particularly impactful. As a senior vice president at Citibank, he has led teams responsible for developing risk predictive models that incorporate alternative data into credit decisioning processes. His experience spans risk analytics, fraud prevention, and the development of recession-based scorecards designed to help banks navigate economic downturns.
One of his most noteworthy projects has been his work on recession-based ensembled scorecards—tools that help businesses prepare for economic fluctuations by using alternative data to predict consumer behavior during recessions.
“Every financial institution needs to be prepared for downturns,” Khullar notes. “When using alternative data, we can create models that are far more accurate and resilient when the economy takes a hit.”
The Evolution of Credit Assessment
The rise of alternative data doesn’t mean that FICO scores are going away anytime soon—but it does show a movement towards more comprehensive credit assessments. Khullar envisions a future where credit evaluation becomes increasingly sophisticated and multidimensional, drawing insights from an expanding universe of data sources.
“By 2030,” Khullar predicts, “we’ll see a world where traditional credit scores like FICO are just one part of the equation. Lenders will use everything from social media activity to real-time transaction data to assess someone’s risk profile.”
This transformation is already gaining momentum across the financial sector. Major financial institutions are already investing heavily in these technologies. A 2024 report from McKinsey estimates that ML/AI-driven credit assessments could reduce default rates by up to 30 percent while expanding access to credit for millions of consumers worldwide.
Balancing Innovation with Responsibility
But with great power comes great responsibility. Critics argue that relying too much on alternative data could open the door to new discrimination or privacy concerns. Social media data, for instance, raises important questions about the boundaries between personal and financial information. After all, should your social media posts influence your approval for a mortgage?
Khullar acknowledges these concerns but believes that alternative data can be used ethically and effectively with proper safeguards in place. “We need transparency. Consumers should understand what data is being used and how it affects their ability to get credit. It’s not enough to build better models—we have to ensure they’re fair,” he says.
A More Inclusive Financial Future
Despite the criticism and potential for abuse, Khullar remains optimistic about the future of his field, “It’s an exciting time where technology is helping us rethink old systems and create something better—something fairer.”
The integration of alternative data in credit assessment signifies a shift toward financial inclusion. These evolving systems aim to create a more equitable financial landscape, rewarding responsible behavior and leveling the playing field for millions underserved by traditional financial systems.
Photo Credit by Nitesh Khullar
