This article was originally published in 2024 and has been updated to reflect developments in alternative data, open banking, AI-driven credit assessment, and findings from the World Bank’s report, The Use of Alternative Data in Credit Risk Assessment: Opportunities, Risks, and Challenges.

Understanding Alternative Data

The creditworthiness evaluation landscape is evolving rapidly, driven by the growth in usage of alternative data in financial analysis.

Alternative data refers to information not traditionally used by financial institutions for creditworthiness evaluation. This can include bank transaction data, utility and rental payment histories, telecom payment records, buy now pay later (BNPL) activity, digital wallet transactions, accounting software data, e-commerce sales data, supply-chain information, as well as broader data sources such as social media activity, online customer reviews, and environmental, social, and governance (ESG) practices. These novel inputs offer fresh perspectives on an organization’s financial health and potential risks, complementing traditional data sources.

Open Banking and Alternative Data

The growth of open banking and open finance frameworks has significantly expanded access to customer-permissioned transaction and cash-flow data, allowing lenders and analysts to gain a more detailed and timely understanding of financial behavior. Examples could include stable cash-flow patterns visible through transaction data, consistent utility and rental payments, strong supplier relationships, recurring e-commerce sales, or robust ESG practices that may indicate lower operational and reputational risk.

Benefits of Alternative Data in Credit Analysis

Research from the World Bank’s report on alternative data in credit risk assessment suggests that alternative data delivers the greatest value when used alongside traditional credit information rather than as a replacement. It can supplement traditional metrics like debt ratios and cash flow analysis, providing a more comprehensive view of credit risk.

The use of alternative data in financial analysis can unlock credit opportunities for organizations that might be overlooked by traditional credit scoring models, particularly small businesses or startups. This is particularly valuable for SMEs and startups, where transaction, payment, and operational data may provide additional insight when traditional credit histories are limited.

It can also enhance predictive accuracy, thereby reducing default risk.

Risks and Limitations

However, the use of alternative data isn’t without challenges. Concerns around data privacy, the quality and reliability of non-traditional data sources, and the potential for discriminatory practices arise. As the use of alternative data expands, it is expected that guidelines on what constitutes acceptable data for credit evaluation will follow to ensure ethical and fair practices.

Additional challenges include ensuring transparency and explainability in AI-driven credit models, obtaining meaningful customer consent for data usage, and managing the risks associated with increasingly complex automated decision-making systems.

Key Considerations for Credit Analysts

When using alternative data to assess a company’s creditworthiness, an analyst should consider the following factors:

• Data relevance: Ensure that the alternative data is relevant to the specific analysis. For example, transaction, payment, or supply-chain data may be more useful for assessing business credit risk, while social media activity may provide supplementary context for consumer-facing organizations.
• Data quality and reliability: The data should be accurate, consistent, and reliable. Since alternative data can come from various sources, checking its credibility is crucial.
• Predictive value: Analysts should assess whether a particular alternative data source improves predictive accuracy beyond traditional financial and credit information.
• Privacy and compliance: Analysts must ensure they’re in compliance with all relevant data privacy laws and regulations when collecting and using alternative data.
• Data interpretation: Understanding how to interpret alternative data correctly is essential. Misinterpretation can lead to incorrect conclusions about a company’s creditworthiness.
• Integration with traditional data: Analysts should consider how alternative data complements traditional financial data. The goal is to use alternative data to enhance the insights gained from traditional data, not replace it.
• Bias and fairness: It’s important to ensure that the use of alternative data doesn’t result in unfair or biased assessments. For instance, using social media activity might disadvantage companies that do not actively use these platforms.
• Model explainability: Analysts should be able to explain how alternative data contributes to a credit assessment and ensure decisions can be justified to stakeholders, regulators, and borrowers.
• Data timeliness: The timeliness of alternative data is also important as outdated information may not accurately reflect the current financial health of a company.
• Scalability: Analysts need to consider whether the alternative data source can be scaled across multiple companies or industries for comparative analysis.

Frequently Asked Questions

• What is alternative data in credit analysis?

Alternative data refers to information not traditionally included in credit reports or financial statements but that can provide additional insight into creditworthiness. Examples include bank transaction data, utility and rental payment histories, telecom payment records, and accounting software data.

• How is alternative data used in credit risk assessment?

Alternative data is used alongside traditional credit information to provide a more complete view of a borrower’s financial behavior and credit risk. It can help identify risks and opportunities that may not be visible through financial statements or credit bureau records alone.

• What are examples of alternative data?

Common examples include transaction and cash-flow data, utility and rental payment histories, telecom payment records, BNPL activity, accounting software data, e-commerce sales data, and supply-chain information.

• How does open banking support credit analysis?

Open banking enables customers to securely share financial information with approved providers. This can give lenders access to transaction histories and cash-flow data, helping them make more informed credit decisions.

• Can alternative data improve credit access for SMEs?

Yes. Alternative data can help lenders assess SMEs with limited credit histories by providing additional insight into cash flow, business performance, and payment behavior.

• What are the risks of using alternative data?

Key risks include data quality issues, privacy concerns, potential bias, lack of model explainability, and cybersecurity challenges. Organizations should ensure alternative data is used responsibly and in compliance with relevant regulations.

• Is social media data used in credit scoring?

Social media data has been explored as a source of alternative data, but its use remains limited. Many lenders place greater emphasis on verified transactions, payment, and cash-flow data because these sources are generally more reliable and easier to validate.

• Does alternative data replace traditional credit analysis?

No. Alternative data is generally most effective when used alongside traditional credit information rather than as a replacement, helping analysts build a more comprehensive view of creditworthiness.

• How does AI use alternative data in credit analysis?

AI can help lenders analyze large volumes of alternative data, identify patterns, and improve credit risk assessments. However, organizations must ensure AI-driven decisions remain transparent, explainable, and compliant with relevant regulations.