Characterizing Product Lifecycle in Online Marketing: Sales, Trust, Revenue, and Competition Modeling

Santosh K C,Arjun Mukherjee

DOI: https://doi.org/10.48550/arXiv.1704.02993

2017-04-10

Social and Information Networks

Abstract:Recent researches have seen an upsurge in the analysis of consumer reviews. Although, several dimensions have been explored, less is known on the temporal dynamics of events that happen over the lifecycle of online products. What are the dominant sales patterns? How are they affected by review count, rating, helpfulness and sentiment? How is trust characterized and what are its effects on sales and revenue? What happens during a market competition? When does a takeover/recovery happen and by what percentage do sales increase on a takeover? This work aims to answer these fundamental research questions based on a sales time-series analysis of reviews of over 1 million products from Amazon.com. We discover novel temporal patterns of sales and interesting correlations of sales with the ratings. We find that trust and helpfulness are important for higher revenue. Based on the analyses, we propose a model to forecast sales that significantly outperforms other baselines. We then explore the phenomena of market competition. Particularly, we characterize different factors that govern survival/death of a product under competition and a model for competition forecast. Experimental results on large-scale reviews demonstrate the effectiveness of the proposed approaches.