Revenue ForecastingADA Asia2022 – 2024Production
Booking Prediction Model
Record-level ML system predicting booking likelihood and value, contributing to pipeline and booking projection.
Weekend
MVP Build Time
The Problem
Revenue teams needed per-record booking predictions for bottom-up forecasting, separate from the aggregate macro forecast.
What Was Built
Built using the Generic Regressor Framework — XGBoost Regressor and Classifier combination predicting booking likelihood and value at the record level. Evolved into the Product Prediction Model (P14) and established the reusable framework (P22).
Business Impact
Provided per-record booking contribution estimates for the Pipeline Projection Engine and established the reusable modeling framework.
Tech Stack
PythonXGBoostscikit-learnBigQuery
Domain Tags
Booking PredictionRevenue ForecastingRecord-Level ML
Details
- Role
- Primary Owner
- Status
- Production
- Tier
- Tier 1
- Period
- 2022 – 2024
- Employment
- ADA Asia