Abstract
Electronic commerce, especially in the business-to-consumer (B2C) context, has for years been a popular research topic in information systems (IS). However, the prior research on the topic has traditionally been dominated by the consumer focus instead of the business focus of online stores. For example, whereas various segmentations exist for online consumers based on their purchase behaviour, no such segmentations have been developed for online stores based on their sales patterns. In this study, our objective is to address this gap in prior research by identifying the most typical sales patterns of online stores operating in the B2C context. By using self-organising maps (SOM) to analyse the monthly sales time series collected from 399 online stores between January 2016 and December 2017, we are able to identify four approximately equally sized segments, each with its characteristic sales pattern. More specifically, two of the segments are characterised by a clear upward or downward trend in the sales, whereas the other two are characterised by strong seasonal sales variation. We also investigate the differences between the segments in terms of several key business and technical parameters of the stores as well as discuss more broadly the applicability of SOM to IS.
Recommended Citation
Makkonen, Markus and Frank, Lauri, "Identifying the Sales Patterns of Online Stores with Self-organising Maps on Time Series Data" (2018). MCIS 2018 Proceedings. 11.
https://aisel.aisnet.org/mcis2018/11