Network Structure and the Long Tail of ECommerce Demand

We report on a research project that studies how network structures affect demand in electronic commerce, using daily data about the graph structure of Amazon.com’s co-purchase network for over 250,000 products, gathered over this year. We describe how the presence of such network structures alters demand patterns by changing the distribution of traffic between ecommerce Web pages. When this traffic distribution generated by the presence of the network is less skewed than the intrinsic or “real world” traffic distribution, such network structures will even out demand across products, leading to a demand distribution with a longer tail. We estimate an econometric model to validate this theory, and report on preliminary confirmation by contrasting the demand distributions of products in over 200 distinct categories on Amazon.com. We measure the overall extent to which a product influences the network by adapting Google’s PageRank algorithm, applying it to a weighted composite of graphs over four distinct seven-day periods, and we characterize the demand distribution of each category using its Gini coefficient. Our results establish that categories whose products are influenced more by the network structure have significantly flatter demand distributions, which provides an additional explanation for the widely documented phenomenon of the long tail of ecommerce demand. Our research in progress aims to extend these results by using a model of consumer behavior more consistent with the economics of electronic commerce, one we term the “strategic surfer” model.

[1]  Chris Volinsky,et al.  Network-Based Marketing: Identifying Likely Adopters Via Consumer Networks , 2006, math/0606278.

[2]  Arun Sundararajan,et al.  Local Network Effects and Complex Network Structure , 2006 .

[3]  Duncan J. Watts,et al.  Collective dynamics of ‘small-world’ networks , 1998, Nature.

[4]  Erik Brynjolfsson,et al.  Consumer Surplus in the Digital Economy: Estimating the Value of Increased Product Variety at Online Booksellers , 2003, Manag. Sci..

[5]  Rajeev Motwani,et al.  The PageRank Citation Ranking : Bringing Order to the Web , 1999, WWW 1999.

[6]  Gary J. Russell,et al.  Modeling Multiple Category Brand with Household Basket Data Preference , 1997 .

[7]  Gary J. Russell,et al.  Modeling Multiple Category Brand Preference with Household Basket Data , 1997 .

[8]  Tülin Erdem,et al.  Econometric modeling of competition: A multi-category choice-based mapping approach , 1998 .

[9]  H. Bauer,et al.  Structuring Product-Markets: An Approach Based on Customer Value , 2007 .

[10]  Wayne S. DeSarbo,et al.  Sculptre: A New Methodology for Deriving and Analyzing Hierarchical Product-Market Structures from Panel Data , 1990 .

[11]  Sergey Brin,et al.  The Anatomy of a Large-Scale Hypertextual Web Search Engine , 1998, Comput. Networks.

[12]  S ButlerBrian.  Membership Size, Communication Activity, and Sustainability , 2001 .

[13]  Jennifer Preece,et al.  Online communities: a longitudinal analysis of communication activities , 2003, 36th Annual Hawaii International Conference on System Sciences, 2003. Proceedings of the.

[14]  Carl D. Meyer,et al.  Deeper Inside PageRank , 2004, Internet Math..

[15]  J. Avery,et al.  The long tail. , 1995, Journal of the Tennessee Medical Association.

[16]  Gerald C. Kane,et al.  Casting the Net : towards a theory of multimodal knowledge networks , 2006 .

[17]  Yu Jeffrey Hu,et al.  From Niches to Riches: Anatomy of the Long Tail , 2006 .