We’ve released our Payroll Prediction Model! Our model detects a user’s payroll transactions, clusters recurring payroll streams, and accurately predicts the next payment date and amount.
Why this matters
People living paycheck to paycheck can find themselves overdrafting their accounts and getting hit with fees due to uncertainty around expected income and spending. Allow your users to see when they will likely get paid next so they can plan accordingly. Identifying the predicted income deposit date can also allow cash advance apps to avoid causing overdraft or NSF fees when pulling a payment from a user’s account. Lastly, loan repayment plans are often not aligned with the paycheck cycle for a user. Pave’s Payroll Prediction allows you to customize debt repayment plans according to the user’s predicted payroll cycle.
Use this model to:
- Understand a user’s payroll frequency
- Predict the payroll’s next_date and next_amount
Use Cases
- Cashflow underwriting
- Optimizing loan repayment schedules
- Predicting overdrafts
- Reducing Cash Advance NSFs
- Build Cashflow forecasting tools
Here are some FAQs to get you started.
How far out does the model predict paychecks?
Currently, we predict the next payroll date and next payroll amount.
How much historical data do I need to upload for a user in order for the model to predict a paycheck?
A minimum of 2 months of transaction history is necessary to detect recurring payroll deposits and to be able to predict the next paycheck deposit.
Where do I access this prediction?
See our recurring_income endpoint here: https://docs.pave.dev/insights/recurring_income and in our attributes endpoint here: https://docs.pave.dev/attributes
Documentation
You can find the documentation here –
https://docs.pave.dev/insights/recurring_income
https://docs.pave.dev/attributes
.png)
