Success story.
Predictive Forecast of Contribution and Withdrawal Flows
Pension funds face volatility in contribution amounts and requests for early withdrawals (“10% withdrawal”), which impacts liquidity and investment planning.
We take on any challenge
At Zenta there are no obstacles we cannot overcome.
Challenge
To anticipate with high accuracy the peaks and valleys of contributions and withdrawals in order to optimize cash management, investment allocation and compliance with fund repayment deadlines.
Solution
We trained time series models in Vertex AI Forecast using historical data on contributions, quote rates, and macroeconomic variables (CPI, exchange rate). We published the predictions on BigQuery and designed a dashboard in Looker with automatic alerts for deviations exceeding 10% from the forecast. We also connected Slack for treasury notifications.
Results from our management
94% accuracy in 30-day forecast.
25% reduction in short-term liquidity costs.
Treasury's ability to react in <1 hour to withdrawal peaks.