A dive into the process of delivering forecasting projects. Often data scientists, analysts, and management get stuck wanting to get the best accuracy out of their forecasts, which leads to weeks stuck in the model development and feature engineering loop. Meanwhile the business operates without your forecast, and end-users are not building up their data literacy to make important decisions with them.
What to prioritize when preparing, modeling, and deploying a forecast?
How to identify a forecast minimum-viable product (MVP)?
Opportunities to iterate and improve on forecasts