Small Businesses and Digital Financial Services – Predictive Modelling and Segmentation for Market Sizing and Product Design

2019 
Micro, Small, and Medium-Sized Enterprises are the backbone of vibrant and dynamic economies. But they are sometimes hard for financial institutions to identify because of the methods they use to conduct their transactions. As a result, many MSMEs do not get access to financing and financial products that are designed specifically to support businesses. Identifying these MSMEs and addressing their needs can be very advantageous for digital financial service providers. This report discusses predictive data models to help a mobile network operator in Sub-Saharan Africa, identify MSMEs in its market and better understand how to serve them. The MNO has a large market share in the country and tens of millions of transactions pass its digital financial services channel each month. This report examines those transactions to determine how many are made by individual consumers and how many are made by entrepreneurs and business owners who use personal accounts to conduct business. The report postulates that a significant number of MSME owners conduct commercial transactions through their personal accounts and are therefore not being identified as business customers and are not being afforded the benefits of business customers. The research shows that MSMEs can be accurately identified with a high degree of statistical confidence. Moreover, the analytic method can be used to segment those MSMEs into more granular business profiles. The segmentation algorithm is driven by patterns of how MSMEs use mobile money. The emerging segments differ in their business characteristics and their financial needs. Multiple research components generated comprehensive insights into the MSME segment in the study country. Apart from analyzing mobile money usage patterns, the team also conducted a survey with 1,275 MSMEs. The survey data was used to inform the development of an MSME identification model and to study and profile businesses.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    0
    References
    0
    Citations
    NaN
    KQI
    []