The Rise of Behaviour-Based Scoring Models in Kenya
Read Time:2 Minute, 45 Second
Kenya is experiencing a notable rise of behaviour-based scoring models in credit assessment, reshaping how lenders evaluate risk and expanding financial access for millions of unbanked adults.
While traditional banking reaches just 23.8% of the population, over 86% of Kenyan adults use mobile money, creating an opportunity for alternative credit scoring approaches.
Behaviour-based scoring models in Kenya work by analyzing digital footprints instead of relying on conventional credit histories, which are often unavailable for many consumers.
These models examine mobile money transactions, phone usage patterns, repayment behaviours, and other digital signals to determine creditworthiness.
This approach has become particularly relevant for the 9.9% of adults excluded from formal financial services, including credit, and has helped bridge the gap in financial inclusion.
The pioneer of this model in Kenya was Commercial Bank’s M-Shwari, launched in 2012 through M-Pesa integration. Without access to formal income or bank statements, M-Shwari scored users based on cash flow behaviours, such as regular airtime top-ups and fund transfers.
Within two years, the platform had 4.5 million active customers and achieved a low non-performing loan rate of 2%, outperforming traditional FICO-style credit scores by 26%.
By using real-time spending patterns as predictors of default, M-Shwari democratized access to microloans, contributing to Kenya’s GDP per capita nearly doubling between 2012 and 2017 as credit fueled small business growth and consumption.
The rise of behaviour-based scoring models in Kenya accelerated with fintech innovations. Companies like Tala and Branch use AI to mine mobile data, including SMS frequency, GPS mobility, and wallet transactions, to assess financial reliability.
Tala, based in Nairobi, provides instant loans via app-based behavioural analytics, serving millions without requiring collateral.
M-KOPA, which targets solar panel instalments, scores users based on repayment consistency, building credit histories for over one million households while expanding access to clean energy.
These models increasingly use alternative data sources to improve predictions. Telecom top-up records, social media activity, and device usage can signal reliability or detect potential fraud.
Regulatory support has further bolstered adoption: the Central Bank of Kenya’s 2019 Risk-Based Credit Pricing Model encouraged data-driven lending, and a 2025 consultative review refined guidelines to ensure fairness and transparency.
A landmark 2025 partnership between TransUnion and FICO introduced Kenya-specific scores using CreditVision Variables. These scores draw from over 145 sources and 24 months of payment history, improving risk prediction by 20-30% and increasing loan approvals by 15-20%.
This innovation benefits both SMEs and underserved consumers. TransUnion’s Q2 2024 Consumer Pulse Study for Kenya reported that 36% of consumers felt they had sufficient access to credit, up slightly from 33% the previous year.
The impact of behaviour-based scoring models in Kenya is tangible. Digital loans reached KSh 55.2 billion annually by 2024, spurring entrepreneurship and supporting small businesses in informal sectors.
With 60% of Kenyans anticipating access to new credit by 2026, these models are set to further democratize financial access, turning everyday digital behaviours into gateways for loans and reshaping Kenya’s financial landscape in line with Vision 2030’s inclusive growth objectives.
Jefferson Wachira is a writer at Africa Digest News, specializing in banking and finance trends, and their impact on African economies.
Happy
0 %
Sad
0 %
Excited
0 %
Sleepy
0 %
Angry
0 %
Surprise
0 %
0
0
Average Rating