For decades, the “unbanked” population in Kenya faced a Catch-22: they couldn’t get credit because they had no credit history, and they had no credit history because they couldn’t get credit. However, the rise of mobile money—led by Safaricom’s M-Pesa—has created a massive digital footprint that is now being used to bridge this gap.
By leveraging alternative data, Kenyan fintechs and banks are building sophisticated credit models that look beyond traditional bank statements to assess risk and opportunity.
1. What is Alternative Data?
Traditional credit scoring relies on “hard” data from Credit Reference Bureaus (CRBs), such as previous loan repayments and bank account history. Alternative data, however, consists of non-traditional digital signals that reflect a person’s financial behavior.
Key Data Sources in Kenya:
- Mobile Money Transactions: Frequency, volume, and consistency of M-Pesa or Airtel Money transfers.
- Airtime Top-ups: How often a user buys credit and in what denominations.
- Utility Payments: History of paying electricity (Kenya Power) or water bills via mobile apps.
- Device Metadata: The type of smartphone used, app usage patterns, and even how quickly a user fills out a loan application.
- Social & Network Data: The strength and size of a user’s contact list (used as a proxy for social capital and stability).
2. How the Credit Models Work
Modern credit models in Kenya use Machine Learning (ML) to process thousands of these alternative data points in seconds. Instead of a binary “yes/no” based on a bank balance, these algorithms look for patterns.
- Consistency over Wealth: A micro-entrepreneur who moves small amounts of money daily through their M-Pesa business till is often deemed more creditworthy than a person with a high balance but irregular activity.
- Predictive Behavior: Data shows that users who maintain a consistent airtime balance or pay their “Lipa Later” installments on time are statistically less likely to default on larger loans.
- Real-time Adjustments: Unlike traditional scores that update monthly, mobile money-based scores can update in real-time, allowing for “nano-loans” that grow in limit as the user proves their reliability.
3. Key Players and Impact
Several “Silicon Savannah” innovators have pioneered this approach, significantly moving the needle on financial inclusion in Kenya.
| Provider | Platform/Model | Impact |
| Safaricom & NCBA | M-Shwari | The first major mobile-centric savings and loan product in Kenya. |
| Branch & Tala | Smartphone Apps | Use device metadata and SMS logs (with permission) to issue instant credit. |
| M-Kopa | Pay-As-You-Go | Uses payment history for solar lamps/phones to unlock larger cash loans. |
| Pezesha | Embedded Finance | Provides credit scoring for SMEs by looking at their supply chain transaction data. |
4. Challenges and Ethical Considerations
While alternative data has democratized credit, it is not without risks:
- The “Blacklist” Effect: Many Kenyans have been listed on CRBs for defaulting on tiny “app loans” (sometimes as low as $2), which then prevents them from getting larger mortgages or business loans later.
- Data Privacy: High-tech scoring requires access to personal data (SMS, contacts, location). Ensuring this data isn’t misused or sold is a major regulatory hurdle.
- Algorithmic Bias: If an algorithm is trained on data that excludes certain demographics (e.g., rural women), it may unintentionally reinforce existing financial inequalities.
Conclusion
In Kenya, your phone is your resume. By turning daily transactions into a “financial identity,” alternative data has helped increase formal financial inclusion from 26.7% in 2006 to over 83% today. As these models become more refined, the focus is shifting from simply “giving loans” to “building wealth” through more affordable and personalized credit.
As alternative data reshapes credit decision-making in Kenya,regulatory compliancemust evolve just as fast. Innovative credit models are only as strong as the IFRS 9 frameworks that support them—ensuring accurate Expected Credit Loss (ECL), audit confidence, and sustainable growth.
At FineIT, we help banks, SACCOs, fintechs, and digital lenders in Kenya translate alternative data and digital lending models into robust, CBK-aligned IFRS 9 solutions. From ECL modeling and segmentation to model validation, overlays, and audit support, we bridge innovation with compliance.
Building digital credit? Let FineIT ensure it’s IFRS 9-ready.
Contact us to future-proof your credit risk framework in Kenya.