Predicted Value Score instead of a tier that only grows

The old player segmentation was cumulative: a tier could only grow and said nothing about future value. I replaced it with a forecast: the model estimates each player's value for the next 90 days from payments, play and recency, and places everyone on a nine-level scale. It retrains quarterly and re-segments monthly, with a quality check before every release and a monthly fact-vs-forecast alert. At equal list size the new scale captures 48–49% of quarterly revenue versus 38–39% for the old one. VIP managers and bonus limits work on it, and the model has been handed over to the ML department.

Input

19 features on payments, withdrawals, play and recency over 30–365 days, across the whole player base.

Model

Gradient boosting predicts 90-day value and maps it to a V0–V8 scale.

Action

Re-segments every player monthly; managers and bonus limits run on the new scale.

Guards

Out-of-time backtest on 4 cuts, a quality gate before each upload, monthly fact-vs-forecast alert.

Impact

At equal list size it captures 48–49% of quarterly revenue vs 38–39%. Handed to the ML department.