This is a practical, no-fluff session designed for analytics professionals who want to move beyond dashboards and deliver forecasts that the business can trust and act on.
Support case volumes are noisy, seasonal, and business-critical. Getting them wrong impacts staffing, SLAs, and customer experience. In this masterclass, we’ll walk through how to build, evaluate, and productionize time series forecasts for case volumes using Python, Facebook Prophet, and exponential smoothing—plus when (and when not) to use deep learning models.Using real-world patterns from customer support and customer success environments, we’ll cover the full journey: connecting and transforming data, getting it “forecast-ready,” choosing the right model for short- and long-term horizons, tuning Prophet with custom holidays and events, validating accuracy against actuals, and finally, embedding forecasts into business workflows and automation.
Learning Outcomes
By the end of this masterclass, participants will be able to:
- Explain the landscape of forecasting approaches Differentiate between classical statistical methods, machine learning, and deep learning for time series forecasting—and understand when deep learning is not the right answer.
- Prepare data that is truly “forecast-ready” Design the data connection and transformation layer, handle seasonality, trends, missing data, and create a readiness checklist for forecasting projects.
- Choose the right model for the use case Compare exponential smoothing, FB Prophet, and deep learning models for short-term vs long-term case volume forecasting, and make informed trade-offs between complexity, interpretability, and effort.
- Tune, evaluate, and trust your forecasts Use FB Prophet with custom holidays and hyperparameter tuning, define appropriate performance metrics, run backtesting, and design a validation framework against actual case volumes.
- Operationalize forecasts for real business impact Learn how to productionize forecasting pipelines, automate retraining, and integrate forecasts into business decision-making (capacity planning, SLA management, and resource allocation), including how to communicate accuracy and impact to stakeholders.