Centrality and community detection: a co-marketing multilayer network Andreia Fernandes Faculty of Economics, University of Porto, Porto, Portugal Patrícia C.T. Gonçalves Laboratory of Artificial Intelligence and Decision Support of INESC TEC, Porto, Portugal, and Pedro Campos and Catarina Delgado Laboratory of Artificial Intelligence and Decision Support of INESC TEC, Porto, Portugal, and Faculty of Economics, Porto, Portugal Abstract Purpose – Based on the data obtained from a questionnaire of 595 people, the authors explore the relative importance of consumers, checking whether socioeconomic variables influence their centrality, detecting the communities within the network to which they belong, identifying consumption patterns and checking whether there is any relationship between co-marketing and consumer choices. Design/methodology/approach – A multilayer network is created from data collected through a consumer survey to identify customers’ choices in seven different markets. The authors focus the analysis on a smaller kinship and cohabitation network and apply the LART network community detection algorithm. To verify the association between consumers’ centrality and variables related to their respective socioeconomic profile, the authors develop an econometric model to measure their impact on consumer’s degree centrality. Findings – Based on 595 responses analysing individual consumers, the authors find out which consumers invest and which variables influence consumers’ centrality. Using a smaller sample of 70 consumers for whom they know kinship and cohabitation relationships, the authors detect communities with the same consumption patterns and verify that this may be an adequate way to establish co-marketing strategies. Originality/value – Network analysis has become a widely used technique in the extraction of knowledge on consumers. This paper’s main (and novel) contribution lies in providing a greater understanding on how multilayer networks represent hidden databases with potential knowledge to be considered in business decisions. Centrality and community detection are crucial measures in network science which enable customers with the highest potential value to be identified in a network. Customers are increasingly seen as multidimensional, considering their preferences in various markets. Keywords Centrality, Co-marketing, Community detection, Multilayer networks Paper type Research paper 1. Introduction Globalization and the speed at which technology has evolved have changed the way society lives, communicates and works (Hand, 2010). As a consequence, the competitiveness between companies is greater and, in the competition to win customers, any comparative advantage, however small, makes the difference between winning and losing a business (Chintagunta et al., 2016). We live in a new world; the power structure is undergoing dramatic changes due to the Internet that has brought connectivity and transparency to our lives (Kotler et al., 2016). The difficulty in attracting and retaining customers generated in companies the need to adopt relationship strategies based on information systems (Huang and Hsueh, 2010). Knowing the profile and behaviour of a customer represents a competitive advantage, since it can be a decisive factor for doing business. The competitiveness of firms may no longer be determined by their size, country or past advantages. Smaller, younger and local companies have the opportunity to compete with larger, more experienced and global companies. A company can be more competitive by connecting with consumer communities or cooperative partners, or by competing in a mix of collaboration and competition (Kotler et al., 2016). We live in a world interconnected by networks in the most diverse forms, be they communication networks, transport networks, or social networks, among others (Hand, 2010). Every day we are all confronted with countless networks in different areas, from the simplest tasks to the most complex ones. Very soon, practically everyone in the world will be interconnected (Kotler et al., 2016). There are many advantages in the use of networks by companies for either networks between businesses (B2B) or between clients (C2C) or both (B2C). Companies such The current issue and full text archive of this journal is available on Emerald Insight at: www.emeraldinsight.com/0885-8624.htm Journal of Business & Industrial Marketing 34/8 (2019) 1749–1762 © Emerald Publishing Limited [ISSN 0885-8624] [DOI 10.1108/JBIM-11-2017-0266] The support of ERDF – European Regional Development Fund through the Operational Programme for Competitiveness and Internationalisation – COMPETE 2020 Programme and the Portuguese funding agency, FCT – Fundação para a Ciência e a Tecnologia within project POCI-01-0145- FEDER-031821 is acknowledged. Received 5 November 2017 Revised 15 April 2018 30 July 2018 5 November 2018 13 March 2019 28 August 2019 Accepted 30 August 2019 1749