African FinTech Loan Application & Approval Dataset
500K+ loan application records from digital lending platforms across Nigeria, Kenya, Ghana, and South Africa. Features applicant demographics, employment status, mobile money history, credit bureau scores, loan amounts, and approval/rejection outcomes with reason codes — purpose-built for credit risk modelling, alternative scoring, and financial inclusion analytics.
This is a synthetic dataset generated from high-quality expert-labelled seed data. All records are algorithmically derived — statistical distributions, inter-field correlations, and annotation characteristics faithfully replicate real-world patterns from the source data, while ensuring no real individual, organisation, or transaction can be identified or reconstructed.
The DS-02 African FinTech Loan Application & Approval Dataset is the most comprehensive open-access-derived credit dataset for Sub-Saharan Africa currently available under a commercial licence. Assembled by unifying demand-side survey microdata, competition loan records, and macro financial indicators, it addresses a critical data gap for teams building credit risk and financial inclusion models across four of Africa's largest economies.
The dataset draws on eight primary sources — including EFInA's Access to Finance Nigeria 2023 survey, Kenya's FinAccess 2021 microdata, the World Bank Enterprise Survey, World Bank Findex 2021, and the Zindi African Credit Scoring Challenge — producing a unified schema of 20 canonical fields per applicant record. Macro enrichment from the IMF Financial Access Survey 2024 provides country-year indicators for contextual modelling.
Key features include binary loan approval outcomes, rejection reason codes, mobile money usage flags, employment and income proxies, and a default outcome label derived from repayment records. The unified schema is designed for direct ingestion into credit risk pipelines, with preprocessing steps documented per source.
Use Cases
Data Quality Scores
Geographic Coverage
Source Summary
Dataset Schema
Each record represents a single loan application or survey respondent mapped to the unified canonical schema. Fields use snake_case; country-specific fields are prefixed with ISO 3166 codes where unification is not possible. All currency values are expressed in USD equivalent.
| Field Name | Type | Description | Nullable |
|---|---|---|---|
| applicant_id | STRING | Unique identifier for the loan application or survey respondent | No |
| country_code | ENUM | ISO 3166-1 alpha-2 country code of applicant (NG / KE / GH / ZA) | No |
| application_date | DATE | Date of loan application or survey interview (ISO 8601) | Yes |
| applicant_age | INTEGER | Age of applicant in years at time of application | Yes |
| applicant_gender | ENUM | Gender of applicant: M / F / Other | Yes |
| employment_status | ENUM | employed_formal / employed_informal / self_employed / unemployed | Yes |
| income_monthly_usd | FLOAT | Applicant's monthly income in USD (converted from local currency) | Yes |
| education_level | ENUM | Highest education: none / primary / secondary / tertiary | Yes |
| has_bank_account | BOOLEAN | Whether applicant holds a formal bank account | Yes |
| has_mobile_money_account | BOOLEAN | Whether applicant has an active mobile money account | Yes |
| loan_amount_usd | FLOAT | Requested loan amount in USD | Yes |
| loan_purpose | ENUM | business / household_expense / education / health / agriculture / other | Yes |
| loan_term_days | INTEGER | Duration of loan in days | Yes |
| loan_approved | BOOLEAN | Primary label: whether loan application was approved (1) or rejected (0) | No |
| default_outcome | BOOLEAN | Whether approved loan subsequently defaulted (repayment failure) | Yes |
| rejection_reason_code | ENUM | no_collateral / low_income / no_credit_history / documentation / other | Yes |
| urban_rural | ENUM | Urban or rural location of applicant | Yes |
| lender_type | ENUM | bank / MFI / mobile_lender / SACCO / informal | Yes |
| macro_sme_loans_gdp | FLOAT | Country-year macro: SME loans as % of GDP (from IMF FAS enrichment) | Yes |
| macro_mfi_accounts_per1000 | FLOAT | Country-year macro: MFI loan accounts per 1,000 adults (IMF FAS) | Yes |
Sample Records
The following are representative synthetic samples demonstrating the dataset structure. Each record corresponds to a single loan application unified across the source schema.
Build with Data that reflects Africa
Request access to our full catalog of licensed human-validated African datasets or request custom data tailored to your project.