African Mobile Money & Agent Banking Behaviour Dataset
Anonymised transaction flows across mobile money ecosystems — M-Pesa, MTN MoMo, and Airtel Money. Captures agent network activity, P2P transfer patterns, cash-in/cash-out behaviour, and seasonal trends across urban and rural splits in Kenya, Uganda, Ghana, and Côte d'Ivoire.
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.
DS-03 is DataLens Africa's most comprehensive mobile money behavioural dataset, aggregating anonymised transaction flows, agent network statistics, and demand-side usage patterns across four of Africa's most active mobile money markets — Kenya, Uganda, Ghana, and Côte d'Ivoire.
Data is compiled from seven open and licensed sources including GSMA Mobile Money Metrics, the IMF Financial Access Survey, Central Bank of Kenya administrative statistics, World Bank Global Findex 2024, FinScope Consumer Surveys, and PaySim synthetic transaction logs. The dataset spans an 18-year time series (2006–2024) at monthly granularity for supply-side metrics, enriched with triennial household-level demand-side survey data for urban/rural and gender disaggregation.
The unified schema covers 18 canonical fields aligned to the core use cases of agent network optimisation, cash flow forecasting, and mobile money fraud detection. Critical gaps — including individual-level transaction microdata and agent geolocation — are documented with recommended proxy and primary collection strategies.
Use Cases
Data Sources Included
Geographic Coverage
Dataset Schema
Each record represents a single data point within the unified DS-03 schema, aggregated from 7 open sources. 18 canonical fields covering transaction flows, agent network metrics, and demographic context.
| Field Name | Type | Description | Nullable | Example |
|---|---|---|---|---|
| country_code | STRING | ISO 3166-1 alpha-2 country identifier | No | KE |
| report_period | DATE | Reporting period in ISO 8601 (YYYY-MM or YYYY) | No | 2024-03 |
| transaction_type | ENUM | CASH_IN / CASH_OUT / P2P / MERCHANT / BILL / REMITTANCE / BULK | No | CASH_IN |
| transaction_volume | INTEGER | Number of transactions in the reporting period | Yes | 4820000 |
| transaction_value_usd | FLOAT | Monetary value of transactions in USD (ISO 4217) | Yes | 1250000.00 |
| agent_count_registered | INTEGER | Total registered mobile money agent outlets | Yes | 342000 |
| agent_count_active | INTEGER | Active agent outlets in the reporting month | Yes | 198000 |
| agent_density_per_100k | FLOAT | Active agents per 100,000 adults | Yes | 42.3 |
| agent_density_per_sqkm | FLOAT | Active agents per 1,000 km² | Yes | 3.1 |
| registered_accounts | INTEGER | Total registered mobile money accounts | No | 38500000 |
| active_accounts_30d | INTEGER | Accounts active in the past 30 days | Yes | 22400000 |
| urban_rural_flag | BOOLEAN | true = urban, false = rural classification | Yes | false |
| gender | ENUM | M / F / Unknown (demand-side records only) | Yes | F |
| income_quintile | INTEGER | 1 = lowest income quintile, 5 = highest | Yes | 2 |
| mobile_money_provider | STRING | M-PESA, MTN MoMo, Airtel Money, Orange Money | Yes | MTN MoMo |
| is_fraud | BOOLEAN | Ground-truth fraud label (PaySim synthetic records) | Yes | false |
| data_source_id | STRING | Source provenance identifier (SRC-01 through SRC-07) | No | SRC-01 |
| data_vintage_year | INTEGER | Year of original data collection | No | 2024 |
Sample Records
Representative samples demonstrating the unified DS-03 schema across supply-side aggregate data (SRC-01/SRC-05), demand-side survey data (SRC-02/SRC-07), and synthetic transaction records (SRC-06).
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.