Temporal Segmentation in Boosting Algorithms

Imagine you have a dataset of Netflix users collected over a decade. You're tasked with training a boosting algorithm to predict whether users will trust the platform, as shown by them entering their credit card details for a trial subscription. Would it be beneficial to categorize users based on the year they joined Netflix when training your model? Explain your reasoning.