Contents lists available at ScienceDirect Technovation journal homepage: www.elsevier.com/locate/technovation Improvisation for innovation: The contingent role of resource and structural factors in explaining innovation capability Yang Liu a , Diwei Lv b , Ying Ying c , Felix Arndt d,e , Jiang Wei a, a School of Management, Zhejiang University, Hangzhou 310058, China b School of Business Administration, South China University of Technology, Guangzhou 510640, China c School of Business Administration, Zhejiang University of Finance & Economics, Hangzhou 310018, China d De Montfort University, Leicester LE1 9BH, United Kingdom e University of Agder, Postboks 422, 4604 Kristiansand, Norway ARTICLE INFO Keywords: Improvisation Innovation Innovation capabilities Centralization Formalization Resource slack China ABSTRACT This paper focuses on resource and structural factors to explore the relationship between organizational im- provisation and innovation capability. Although the role of improvisation has attracted increasing academic attention in fast-changing environments, little is known about the conditions under which rms benet from improvisation. This paper addresses this gap using an organizational learning perspective that explains the role of a rm's organizational structure and organizational resources for improvisation and innovation. A large-scale survey in China nds that rms vary in their levels of (I) centralization and formalization of decision making and (II) resource slack and investment irreversibility and that these factors moderate the relationship between im- provisation and innovation capability in distinct ways. Consistent with our theorizing, improvisation enhances innovation capability when rms have a decentralized but formalized structure or pursue the dual goals of maximizing resource slack and minimizing investment irreversibility. 1. Introduction In an accelerated competitive environment, rms constantly create new products and processes often in an improvised manner (Adomako et al., 2018; Leskovar-Spacapan and Bastic, 2007; Wang et al., 2008; Seo et al., 2017). Extemporaneous organizational action thus is in- evitable, and has increasingly triggered scholarly interest due to its potential value in building innovative capability (Miner et al., 2001; Hadida et al., 2015). Understanding organizational improvisation, de- ned as the degree to which composition and execution converge in time (Moorman and Miner, 1998a: 698), has become a crucial element for research on dynamism, emergence, and innovation (Linstone, 2011; Hadida et al., 2015). However, prior research has found mixed results regarding the ef- fects of improvisation on innovation (Flach, 2014; Vera and Crossan, 2004, 2005; Vera et al., 2016; also see Hadida et al., 2015, for a re- view). Two views dominate the debate. According to the reactive view, organizational improvisation may occur within what Moorman and Miner (1998b): 5) called the logic of responsiveness,thus rendering prior plans irrelevant or incomplete when facing unexpected jolts. In their study, they nd that improvisation can reduce new product ef- fectiveness. In contrast, scholars propagating the logic of activeness argue that organizational improvisation enables managers to con- tinuously and creatively adjust to change and to consistently move products and services out the door(Brown and Eisenhardt, 1998: 33). For instance, Akgun et al. (2007) nd that team improvisation posi- tively aects new product success by utilizing/implementing new knowledge. To resolve the tension in the theoretical approaches and empirical ndings regarding the eects of improvisation, this paper argues that the logic of responsiveness and that of activeness pertain to dierent facets of organizational improvisation: the logic of responsiveness fo- cuses on the outcomes of extemporaneous action, while the logic of activeness refers to the selective retention of such outcomes, so that they are intertwined in explaining the eect of organizational im- provisation on innovation capability. Whereas the actions that emerge or are recognized during improvised activity vary in their degree of novelty, coherence, and speed (Moorman and Miner, 1998a), selective retention of the outcomes of these actions will create awareness that reduces inertia and stipulates learning from improvisation with con- sequences for a rm's innovation capability. Following the logic of the two lenses and the associated learning mechanisms, this paper focuses on two factors. First, improvisation requires resources that are readily available. One the one hand, slack https://doi.org/10.1016/j.technovation.2018.02.010 Received 2 April 2016; Received in revised form 30 December 2017; Accepted 23 February 2018 Corresponding author. E-mail addresses: liuyang.zju@gmail.com (Y. Liu), lvdiwei@163.com (D. Lv), yingying@zufe.edu.cn (Y. Ying), arndt@gmail.com (F. Arndt), weijiang@zju.edu.cn (J. Wei). Technovation xxx (xxxx) xxx–xxx 0166-4972/ © 2018 Elsevier Ltd. All rights reserved. Please cite this article as: Liu, Y., Technovation (2018), https://doi.org/10.1016/j.technovation.2018.02.010