Alternative credit scoring: Scouring big data for consumer information

• Author:
• Author name

  Quantumrun Foiresight
• October 10, 2022

Insight summary

More companies are using alternative credit scoring because it benefits consumers and lenders. Artificial intelligence (AI), specifically machine learning (ML), can be used to assess the creditworthiness of people who don’t have access to traditional banking products. This method looks at alternative data sources like financial transactions, web traffic, mobile devices, and public records. By looking at other data points, alternative credit scoring has the potential to increase financial inclusion and drive economic growth.

Alternative credit scoring context

The traditional credit score model is limiting and inaccessible for many people. According to data from the Africa CEO Forum, around 57 percent of Africans are “credit invisible,” which means they lack a bank account or credit score. As a result, they have difficulty securing a loan or obtaining a credit card. Individuals who do not have access to essential financial services such as savings accounts, credit cards, or personal checks are considered unbanked (or underbanked).

According to Forbes, these unbanked people need electronic cash access, a debit card, and the ability to obtain money promptly. However, traditional banking services usually exclude this group. In addition, the complex paperwork and other requirements for conventional bank loans have resulted in vulnerable groups turning to loan sharks and payday creditors that impose high-interest rates.

Alternative credit scoring can help the unbanked population, especially in developing nations, by considering more informal (and often more accurate) means of evaluation. In particular, AI systems can be applied to scan large volumes of information from diverse data sources, such as utility bills, rent payments, insurance records, social media usage, employment history, travel history, e-commerce transactions, and government and property records. Additionally, these automated systems can help identify recurring patterns that translate to credit risk, including the inability to pay bills or hold jobs for too long, or opening too many accounts on e-commerce platforms. These checks focus on a loanee’s behavior and identify data points that traditional methods might have missed. 

Disruptive impact

Emerging technologies are a key factor in accelerating the adoption of alternative credit scoring. One such technology includes blockchain applications due to its ability to let customers control their data while still allowing credit providers to verify the information. This feature could help people feel more in control of how their personal information is stored and shared.

Banks can also use the Internet of Things (IoT) for a more detailed picture of credit risk across devices; this includes collecting real-time metadata from mobile phones. Healthcare providers can contribute various health-related data for scoring purposes, such as data collected from wearables like heart rate, temperature, and any record of pre-existing health issues. While this information does not directly apply to life and health insurance, it may inform bank product choices. For example, a potential COVID-19 infection might signal the need for emergency overdraft assistance or small and medium enterprises having higher risk factors for loan repayment and business disruption. Meanwhile, for car insurance, some companies use telematics data (GPS and sensors) instead of traditional credit scoring to assess which candidates are most likely to be liable. 

One key data point in alternative credit scoring is social media content. These networks hold an impressive amount of data that can be useful in understanding a person’s likelihood to repay debts. This information is often more accurate than what formal channels reveal. For example, checking account statements, online posts, and tweets give insights into someone’s spending habits and economic stability, which can help businesses make better decisions. 

Implications of alternative credit scoring

Wider implications of alternative credit scoring may include: 

• More non-traditional credit lending services fueled by open banking and banking-as-a-service. These services may help the unbanked apply for loans more efficiently.


• The increasing use of IoT and wearables to assess credit risk, particularly health and smart home data.


• Startups using phone metadata services to asses unbanked people to offer credit services.


• Biometrics being increasingly used as an alternative credit score data, particularly in monitoring shopping habits.


• More governments making non-traditional credit more accessible and serviceable. 


• Increasing concerns about potential data privacy violations, particularly for biometric data collection.

Questions to consider

• What are the potential challenges in using alternative credit scoring data?


• What can other potential data points be included in alternative credit scoring?