BBE-LSWCM: A Bootstrapped Ensemble of Long and Short Window Clickstream Models
This addresses real-time prediction problems for SaaS product managers, though it appears incremental as it combines existing windowing approaches.
The paper tackles real-time customer event prediction in SaaS products by developing BBE-LSWCM, an ensemble architecture combining long and short window clickstream data, which shows superior performance for subscription cancellation and intended task detection in QBO compared to baselines.
We consider the problem of developing a clickstream modeling framework for real-time customer event prediction problems in SaaS products like QBO. We develop a low-latency, cost-effective, and robust ensemble architecture (BBE-LSWCM), which combines both aggregated user behavior data from a longer historical window (e.g., over the last few weeks) as well as user activities over a short window in recent-past (e.g., in the current session). As compared to other baseline approaches, we demonstrate the superior performance of the proposed method for two important real-time event prediction problems: subscription cancellation and intended task detection for QBO subscribers. Finally, we present details of the live deployment and results from online experiments in QBO.