Electronic commerce has been expanding rapidly in the last 15 years and is now present in almost all sectors and in a majority of developed countries’ markets. As a distributed environment, e-commerce involves a large number of market participants: customers, traders, intermediaries and other service providers who communicate, trade and collaborate among themselves using ICT-based applications. In the beginning of the e-commerce era, adoption of e-business applications provided companies with significant competitive advantage. It now may not be the case. A large number of companies are able to develop their e-commerce infrastructure relatively quickly and offer their services and products via the Internet. In order for e-commerce business to be successful, it is necessary to develop good business strategies and offer additional services to customers (Rosaci and Sarne, 2012). Examining the behaviours of stakeholders in e-commerce has been the subject of numerous research studies. The academic literature often uses regression analysis as the most common approach for recognising the impact of key success factors of a considered e-commerce model (Kim et al, 2008). Besides, neural network-based models are increasingly developed (Pao-Hua et al, 2010). To improve the existing solutions and explore new means to support better business decisions, recent research has increasingly implemented agent-based models in the analysis of e-commerce models. Among the first to report such a model are Janssen and Jager, who investigated processes that lead to ‘‘lock-in’’ in the consumer market (Janssen and Jager, 1999). One of the best-known models used in practice was developed by North and Macal for the needs of Procter & Gamble (North and Macal, 2010). Tao and David Zhang used the agent-based simulation model to present the effect of introducing a new product on the market to serve as decoy (Zhang and Zhang, 2010). Although the authors confined themselves to only explaining the application of the mentioned effect, the model itself is far more comprehensive and deals with psychological mechanisms that govern customers in choosing a particular product. Okada and Yamamoto used the agent-based simulation model to investigate the impact of the eWOM effect upon the habits of customers purchasing on B2C websites (Okada and Yamamoto, 2009). Special attention is paid to the exchange of knowledge and information on the product between the buyers. Furthermore, the literature describes a large number of agent-based simulation models used in customer behaviour studies. An interesting example is the CUBES simulator (customer behaviour simulator), which studies mechanisms of customer interactions and their effect on different economic phenomena (Said and Drogoul, 2001). Liu et al used the agent-based simulation model to investigate the nowadays common continual price reductions on online markets (Liu et al, 2013). In recent years, this methodology has been successfully used in simulating customer behaviour on social networks and research into the effect of social networks on viral marketing (Hummel et al, 2012; Zutshi et al, 2014).