Capturing Conversion Rate Fluctuation during Sales Promotions: A Novel Historical Data Reuse Approach

TL;DR

This paper introduces HDR, a novel approach that reuses historical promotion data to improve conversion rate prediction during sales promotions, addressing distribution shift issues and enhancing model performance.

Contribution

The paper proposes HDR, a new method combining data retrieval, shift correction, and quick fine-tuning to adapt CVR models to promotional contexts effectively.

Findings

HDR significantly improves ranking and calibration metrics.

Deployment during Alibaba's Double 11 increased RPM by 9% and CVR by 16%.

The approach effectively handles distribution shifts in promotional data.

Abstract

Conversion rate (CVR) prediction is one of the core components in online recommender systems, and various approaches have been proposed to obtain accurate and well-calibrated CVR estimation. However, we observe that a well-trained CVR prediction model often performs sub-optimally during sales promotions. This can be largely ascribed to the problem of the data distribution shift, in which the conventional methods no longer work. To this end, we seek to develop alternative modeling techniques for CVR prediction. Observing similar purchase patterns across different promotions, we propose reusing the historical promotion data to capture the promotional conversion patterns. Herein, we propose a novel \textbf{H}istorical \textbf{D}ata \textbf{R}euse (\textbf{HDR}) approach that first retrieves historically similar promotion data and then fine-tunes the CVR prediction model with the acquired data for better adaptation to the promotion mode. HDR consists of three components: an automated data retrieval module that seeks similar data from historical promotions, a distribution shift correction module that re-weights the retrieved data for better aligning with the target promotion, and a TransBlock module that quickly fine-tunes the original model for better adaptation to the promotion mode. Experiments conducted with real-world data demonstrate the effectiveness of HDR, as it improves both ranking and calibration metrics to a large extent. HDR has also been deployed on the display advertising system in Alibaba, bringing a lift of $9\%$ RPM and $16\%$ CVR during Double 11 Sales in 2022.