We’re excited to announce our Spending Patterns 2.0 API, which uses machine learning to analyze your users’ spending habits. This helps bring insight into what users’ spend money on, and the frequency of their expenses.
The Spending Patterns API combines two machine learning models to classify transactions into spending categories and payment methods.
Spending categories show what your users spend on. There are 23 categories in total, ranging from: housing, utilities, insurance, entertainment, and more.
Payment methods help you determine how your users spent their money, such as through ATM, web or PoS, mobile or web application, or via USSD.
The API also returns a confidence score to indicate an estimated level of accuracy in classifying transaction data.
Use Spending Patterns 2.0 to:
Automate lending decisions based on fixed expenses in certain spending categories
Provide personalized recommendations for personal finance optimizations
Determine average fixed expenses vs. discretionary expenses
Rank expense categories as a percentage of total income
Analyze inflow and outflow patterns over time
Check out the docs here, API reference here, or talk to an expert to get started.