Popularity, face and voice: Predicting and interpreting livestreamers' retail performance using machine learning techniques

TL;DR

This study uses machine learning to analyze factors influencing livestreaming sales, revealing that popularity and voice attributes significantly impact performance, with explainable AI uncovering new insights into viewer engagement and sales dynamics.

Contribution

It compares multiple ML models for predicting livestream sales and introduces a novel 3D-SHAP diagram for interpreting feature importance and sales stages.

Findings

Popularity of livestream events is crucial for sales.

Voice attributes are more influential than appearance, especially for male hosts.

Sales growth can be segmented into three stages based on engagement metrics.

Abstract

Livestreaming commerce, a hybrid of e-commerce and self-media, has expanded the broad spectrum of traditional sales performance determinants. To investigate the factors that contribute to the success of livestreaming commerce, we construct a longitudinal firm-level database with 19,175 observations, covering an entire livestreaming subsector. By comparing the forecasting accuracy of eight machine learning models, we identify a random forest model that provides the best prediction of gross merchandise volume (GMV). Furthermore, we utilize explainable artificial intelligence to open the black-box of machine learning model, discovering four new facts: 1) variables representing the popularity of livestreaming events are crucial features in predicting GMV. And voice attributes are more important than appearance; 2) popularity is a major determinant of sales for female hosts, while vocal aesthetics is more decisive for their male counterparts; 3) merits and drawbacks of the voice are not equally valued in the livestreaming market; 4) based on changes of comments, page views and likes, sales growth can be divided into three stages. Finally, we innovatively propose a 3D-SHAP diagram that demonstrates the relationship between predicting feature importance, target variable, and its predictors. This diagram identifies bottlenecks for both beginner and top livestreamers, providing insights into ways to optimize their sales performance.