Master’s Thesis
The Impact of Enterprise Social Media Platforms’ Use on Exploratory and
Exploitative innovation. The Moderating Effects of Feedback and Top
Management Engagement
Student name: Monica Vicol Student number: 10527435 MSc: Business Administration Track: Digital Business
Faculty of Economics and Business Supervisor: Dr. M. de Haas
Statement of Originality
This document is written by student Monica Vicol, who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Abstract
Organizations are increasingly implementing Enterprise Social Media Platforms (ESMPs) internally to support collaboration and connectedness among employees. While there is research investigating both the benefits and pitfalls of using these technologies, this paper represents a quantitative study that examined whether the use of ESMPs brings added value to organizations by having a positive effect on employees’ extent of exploratory and exploitative innovation. Moreover, by using the Feedback Intervention Theory developed by Kluger & DeNisi (1996), this study examined whether peer task-motivation and task-learning feedback amplify/reduce the effect of ESMPs’ use on exploratory and exploitative innovation. Additionally, the moderating effect of top management engagement was tested as well. The data was collected through a survey and the sample for which no geographical boundary has been preset included N=100 employees using ESMPs at work. To test the direct effects of ESMPs’ use on employees’ exploratory and exploitative innovation a MANCOVA followed by Scheffe’s post hoc test were performed. The moderating effects were tested by conducting two MANCOVAs for task-learning and task-motivation feedback, and one factorial ANOVA for the effect of top management engagement. The findings of the study revealed that a frequency of ESMPs’ use of at least six hours per week is positively related to a higher level of exploratory innovation, while using them at least 1-5 hours per week is positively related to a higher level of exploitative innovation. No support has been found for the moderating effects of task-learning feedback, task-motivation feedback and top management engagement.
Keywords: Enterprise Social Media Platforms, exploratory and exploitative innovation, Feedback Intervention Theory, top management engagement
Table of contents
Table of contents ... 4
1. Introduction ... 5
2. Literature review ... 9
2.1. Enterprise Social Media Platforms (ESMPs) ... 9
2.2. Exploratory and Exploitative Innovation ... 11
2.3. Feedback ... 14
2.3.1. Task-motivation feedback ... 16
2.3.2. Task-learning feedback ... 16
2.4. Top Management Engagement ... 18
2.5. Conceptual framework ... 20 3. Methodology ... 21 3.1. Research design ... 21 3.2. Sample ... 21 Table 1 ... 23 3.3. Measurement of variables ... 24 3.3.1. Independent Variable ... 24 3.3.2. Dependent variables ... 24 3.3.3. Moderating variables... 25 3.3.4. Control variables ... 26 3.4. Statistical procedure ... 28 4. Results ... 30 4.1. Correlation analysis ... 30 4.2. Pre-analyzing data ... 31 4.2.1. MANCOVA ... 31 4.3. Direct effects... 34 4.4. Moderation effects ... 35 4.3.1. Task-motivation feedback ... 35 4.3.2. Task-learning feedback ... 37
4.3.3. Top Management engagement ... 39
5. Discussion ... 42
5.1. Theoretical implications and future research ... 42
5.2. Managerial implications ... 46 5.3. Limitations ... 46 Bibliography ... 48 Appendix A ... 55 Appendix B ... 61 Appendix C ... 64 Appendix D ... 65
1. Introduction
Web 2.0 technologies are disrupting industries, markets and organizations (Sultan, 2013). One of the main characteristics describing this type of technologies is the empowering of Internet users with the possibility to create and exchange content using Internet-based platforms. One such platform type that has been high on the agenda lately for both practitioners and academics are the Enterprise Social Media Platforms (ESMPs). ESMPs are web platforms implemented internally in organizations to support collaboration and connectedness among employees. According to the latest reports of several global consulting companies, the leading ESMPs at the moment are: Yammer (by Microsoft), Slack (by Slack Technologies), Chatter (by Salesforce), Jive (by Jive Software), (Alimam, Bertin & Crepi, 2015). The global market of ESMPs is forecasted to reach USD 4.8 billion by 2020 (Global Industry Analysts, 2015) due to the multiple advantages they are believed to bring.
One of the major benefits resulting from the implementation and use of ESMPs is innovation fostering. By making information content and communication equally visible to all users of the platform, ESMPs help reduce redundancy, enhance meta-knowledge (knowledge about what and whom other people in the organization know) and open space for idea sharing and development, which in turn boosts employees’ innovative performance (Leonardi, 2014; Recker, Malsbender & Kohlborn, 2016). Other benefits discussed in previous research are knowledge management facilitation, fostering of collaboration and knowledge access (von Krogh, 2012; Kuegler, Smolnik & Kane, 2015), improvement of customer relationship management, employee engagement, employee training (Andriole, 2010), among others. Achieving these benefits is especially relevant in today’s economy, increasingly characterized by project work, flex-work and cross-functional cooperation, hence the positioning of ESMPs as a strategic component in companies’ IT portfolio (Karoui,
Still, in practice, these long-term benefits from implementing and using such a platform are not always perceived to outweigh the immediately perceptible disadvantages. The main reasons why companies can be reluctant to adopt ESMPs are: data security issues, lack of employee engagement on these platforms and hence an insufficient rate of their use, and uncertainty about their added value even when used actively (McAfee, 2009). This is why this paper will build upon the findings of previous studies on the benefits of using ESMPs and extend the knowledge about the added value ESMPs can bring to organizations, particularly in contributing to innovation.
Despite the abundant literature studying the impact of ESMPs’ use on various firm outcomes, there is relatively scarce research investigating its implications on organizations’ innovation. Andriole (2010) explored the general business impacts of ESMPs, including innovation; Recker et al. (2016) described the progress of innovation ideas on ESMPs across five stages of innovation and finally; Kuegler et al. (2015) studied the relationship between ESMPs’ use and employee innovative performance by examining the moderating effect of task equivocality in both an inter- and intra-team context. These studies have adopted a relatively broad perspective on the concept of innovation, without accounting for the possible different impacts ESMPs’ use can have on the development of radical vs. incremental knowledge outcomes, which Benner & Tushman (2003) classified as exploratory and respectively, exploitative innovations. Exploratory innovation is built upon fundamentally new knowledge, targeted towards new market and consumer segments whereas exploitative innovation is built upon existing knowledge and is targeted towards existing market and consumer segments. As engaging in both types of innovation is considered to be essential for companies striving to compete in a dynamic environment, this paper will study the extent to which the use of ESMPs can impact their development (Jansen, van den Bosch & Volberda, 2006).
Moreover, current studies have called for the need of further research to expand existing findings and analyze other moderating factors affecting the extent of innovation stimulation via ESMPs’ use. The first moderator to be analyzed is feedback (under the form of comments). Previous research has found that (1) feedback focused on the quality of a user’s idea in an ESMP context (Wooten et al., 2017) and (2) feedback posted on ESMPs under the form of comments has positive impact on subsequent user participation and idea sharing (Brzozowski, Sandholm & Hogg, 2009). This paper will further extend these findings by building on the Feedback Intervention Theory (FIT) developed by Kluger et al. (1996) which differentiates among several types of feedback. Namely, it will analyze the impact of task-motivation and task-learning feedback on the relationship between ESMPs’ use and the two types of innovation.
Furthermore, another variable found to have a positive impact on employees’ participation on ESMPs and on perceived value of ESMPs’ use is the engagement of top management on the platforms (Brzozowski et al. 2009; Chin, Evans & Choo, 2015). Additionally, as previous research has shown that top management can affect the advancement of innovation outcomes through promoting their strategic orientations, it is expected that they can use ESMPs to endorse these strategic orientations and foster both types of innovation through directly participating on the platforms. Hence, this paper will investigate the effect of top management engagement on the relationship between ESMPs’ use and exploratory and exploitative innovation.
To sum up, the central question investigated in this study is to what extent ESMPs’ use has a positive impact on both exploratory and exploitative innovation in an organization. Consequently, the following three sub-questions will be answered: What is the effect of (1) task-motivation feedback, (2) task-learning feedback and (2) top management engagement on
ESMPs, on the relationship between ESMPs’ use and exploratory and exploitative innovation.
This study is expected to add value to the academic literature by extending the current studies on ESMPs and comparing their impact on the two types of innovation, while simultaneously measuring their effect under conditions of feedback and top management engagement. It is also expected to have managerial implications for organizations and managers seeking to boost employee innovation efforts via ESMPs. It will do so by offering organizations an insight into which type of innovation is more likely to be promoted by ESMPs. Further, results of this study may guide managers in choosing strategies regarding feedback type and extent of top management engagement to be encouraged on ESMPs in order to derive more value.
Regarding the structure of the paper, the next section will further review the relevant literature on ESMPs, explorative and exploratory innovation, task-motivation and task learning feedback, and finally, top management engagement. Consequently, this section will present the eight hypotheses analyzed in the study, which will further be depicted in a conceptual framework. Next, the methodology part will elaborate on the research design, sample description and measurement of variables. The fourth section will present the analysis and main results with regard to the eight hypotheses, which will further be interpreted and concluded in the discussion section. The theoretical and practical implications of the findings along with some recommendations for future research will be pointed out and finally the limitations of the study will be mentioned.
2. Literature review
2.1. Enterprise Social Media Platforms (ESMPs)
Enterprise Social Media Platforms also referred to as enterprise social software, enterprise social system and Enterprise 2.0 systems are web-based platforms that allow the communication and interaction among internal and external stakeholders of a company, the explicit indication or implicit revealing of particular coworkers as communication partners, by means of posting, editing, sorting and viewing of messages, connections, text and files at any chosen point in time (McAfee, 2009; Leonardi et al., 2013). The content of these media platforms can be categorized in formal work-related discussions and informal non-work-related discussions (Mantymaki et al., 2016). The development of such platforms may follow one of the three paths: public sites (e.g. Facebook, Google+), private systems, which are either open source (e.g. TWiki, StatusNet) or proprietary software (e.g. Yammer, Webex Social etc.), and in-house developed solutions which are typically prototyped by software vendors to consequently be sold commercially. This study focuses on the second and third types of platforms, namely on private and internally developed systems, as these are the most frequently adopted ESMPs amongst the large corporations mainly due to security and privacy concerns (Leonardi et al., 2013).
There are two main characteristics underlying the difference between ESMPs and other communication technologies used by enterprises (e.g. e-mail, Q&A forums). Firstly, they offer members visibility of the communicative actions of any other member of the platform and of the network each member has with the other users, thus increasing transparency and generating meta-knowledge (knowledge about what and whom other people in the organization know) (Leonardi, 2014). Second, they offer persistence over time by recording and storing all communicative actions, hence offering access to this information in
its original format even after the presenter is disengaged (Leonardi et al., 2013). Additionally, using an affordance-based approach, Treem and Leonardi (2012) suggest two additional opportunities for the use of ESMPs. The first one is association, which similar to meta-knowledge, allows for the visualization of social connections among platform members, and of the connection between the content and its source. The second one is editability and it allows the content creator to edit or rectify his/her content over time, which in turn has positive implications for the continuous improvement of the content quality on the platform. These ESMPs’ characteristics can generate new possibilities for knowledge management through increasing the social capital and thus enhancing organizational knowledge sharing in such activities as rapid application development, customer relationship management, communication, collaboration, social learning, training (Andriole, 2010; Ellison et al., 2015). Consequently, this leads to an increased level of idea generation whereby any platform member can offer his/her input, hence enlarging the pool of knowledge and ideas of the enterprise. This concept of leveraging on the ‘wisdom of the crowd’ to create and make decisions is known as collective intelligence, and it has been considered as one of the main benefits brought by Enterprise 2.0 technology (McAfee, 2009).
Besides these inherent characteristics, Mantymaki et al. (2016) have found in their study that employees view ESMPs as self-organizing platforms for sharing ideas and information rather than as tools used for accomplishing predefined tasks. This idea highlights the difference between ESMPs and traditionally used applications such as ERP (Enterprise Resource Planning) or CRM software, which are based on structured information and involve planning (Mantymake et al., 2016; McAfee, 2009).
Therefore, due to ESMPs’ characteristics of visibility, persistence, accountability, editability and self-organizing which open up opportunities for all platform members to accumulate and generate knowledge and ideas, the implementation and usage of ESMPs can
have a positive impact on employee innovation. This idea has been supported in the study of Kuegler et al. (2012) where they found a significant positive relationship between ESMPs’ use and innovative performance (p < .001). However, as explained in the introduction, this study will account for the possible difference the impact ESMPs’ use can have on two types of innovation: exploratory and exploitative. The following sections will further elaborate on the impact of ESMPs on these two types of innovation and the effect of feedback type and top management engagement on the relationship between ESMPs’ use and employee innovation.
2.2. Exploratory and Exploitative Innovation
Innovations can be classified in two types depending on (1) their proximity to existing technologies, products, services and (2) their proximity to existing customers or market segments. Exploratory innovations are characterized as radical by offering new designs, creating new markets, new channels of distribution and are designed towards fulfilling the needs of new customers or markets. Conversely, exploitative innovations are incremental, focusing on improvement, refinement, efficiency and implementation (Schamberger, Cleven & Brettel, 2013). Exploitative innovations use as a starting point existing knowledge, skills, processes and structures and are designed to fulfill the needs of current customers or markets, by expanding existing products, designs, distribution channels (Abernathy et al., 1985).
There are two opposing views regarding exploratory and exploitative innovations in an organizational context. On the one hand, they are viewed as conflicting activities situated on a continuum, each requiring different systems, structures, processes, strategies, capabilities and cultures which can consequently, lead to different performance outcomes (e.g., Gupta et al., 2006; Lavie et al., 2010). On the other hand, some authors view the two
types of innovations as complementary and occurring simultaneously through the adoption of an ambidextrous perspective. Jansen et al. (2006) have found that an effective coordination mechanism enhancing organizational ambidexterity are the strong social relations among workers. This idea is supported by Subramanian et al. (2005) as well, who claim that a firm’s social capital enables its capability to develop both incremental and radical innovations. The institutionalization of these social relations, measured in density, can bring several advantages.
Firstly, they enhance connectedness among employees, which has been proven to have a positive effect on both exploratory and exploitative innovation. The implementation of ESMPs can stimulate interaction among employees across all organizational layers regardless of their function or status, thus building a sense of connectedness among them. This in turn creates positive emotions and feelings of attachment and reciprocity for the employees (Chin et al., 2015).
Secondly, enhanced social relations in an organization allow for informal coordination mechanisms, opposed to centralization and formalization, which have also been found to have a positive impact on both types of innovation. Based on the qualitative study performed by Chin et al. (2015), employees using an ESMP reported perceiving an enhanced feeling of free speech which increased their engagement and amount of content shared on the platform. Moreover, they reported that due to the platforms’ characteristics, employees are able to engage in more casual and informal communication compared to other communication channels. Therefore, as the use of ESMPs has been shown to both (1) build a sense of connectedness among workers and (2) strengthen the informal collaboration among them, it is expected that their use would also have a positive impact on both exploratory and exploitative innovation.
Moreover, the theory of communication visibility developed by Leonardi (2014) claims that by making invisible communication between employees of an organization visible to the rest of the employees, improves their meta-knowledge. Translated to the context of ESMPs, this implies that by switching communication from private channels (e.g. e-mail, phone, text message) to an ESMP, employees can visualize both the messages and the social network of the rest of employees, thus expanding their meta-knowledge. Consequently, this leads to work streamlining across the organization, avoidance of task duplication, while at the same time it allows employees to combine their existing knowledge with the knowledge of other platform members, hence, fostering their innovativeness (Leonardi, 2014). Similar to exploitative innovation, this type of innovation described in the theory of communication visibility is based on existing knowledge and on knowledge recombination to generate product or process innovation. Therefore, this points out another reason for expecting the use of ESMPs to have a positive impact on exploitative innovation. Furthermore, this highlights the need to examine whether the impact of using ESMPs is equally significant for fostering exploitative and exploratory innovation or whether it differs for the two types.
All in all, it is expected that the use of ESMPs can have a positive impact on both exploratory and exploitative innovation, with a higher level of significance for exploitative innovation. Consequently, the two hypotheses follow:
Hypothesis 1: ESMPs’ use has a positive impact on exploitative innovation. Hypothesis 2: ESMPs’ use has a positive impact on exploratory innovation.
This section has introduced the concepts of exploratory and exploitative innovation. Also, it has noted that according to prior research ESMPs’ use might have a positive impact on these two types of innovation. In the next two sections, the paper will discuss the role of
feedback and top management engagement on the relationship between ESMPs’ use and exploratory and exploitative innovation.
2.3. Feedback
Feedback is defined as the “action taken by (an) external agent(s) to provide information regarding some aspect(s) of one’s task performance” (Kluger & DeNisi, 1996, p. 255). In an organizational context, feedback evaluates an employee’s behavior and reflects its extent of desirability by the organization (Rosen, Hall & Levy, 2006). Throughout the literature, feedback has been found to have inconsistent effects on performance. For instance, in analyzing the impact of feedback on idea generation capacity in the context of innovation tournaments, Wooten et al. (2017) have found that different types of feedback have different effects on the idea generation process. The Feedback Intervention Theory (FIT) was developed by Kluger et al. (1996) to explain these contradicting effects by differentiating among several types of feedback. The five interdependent arguments underlying the FIT are: a) behavior is regulated by comparisons of feedback to goals or standards, (b) goals or standards are organized hierarchically, (c) behavior is regulated only when the feedback-standard gaps receive attention from the feedback receiver actively, (d) attention is normally directed to a moderate level of the hierarchy under which goals are organized, and finally the last and most important argument (e) feedback changes the locus of attention of the feedback receiver and therefore affects his/her behavior (Kluger & DeNisi, 1996). The changes in locus of attention can take place at three hierarchically organized levels of control. Each level towards which feedback is targeted, can result in a different performance outcome. These three levels of control are described below.
First, meta-task feedback is located at the top of the hierarchy and involves one’s self (e.g. self-goal, self-esteem). Second, task-motivation feedback involves the focal task and is focused towards increasing one’s effort towards eliminating the gap between the current performance level and a pre-determined standard. Third, task-learning feedback involves providing specific task details of the focal task to enhance performance. Feedback at meta-task level should generally be avoided as focusing on the actions and behavior of a particular individual may reduce the cognitive resources and attention from the focal task, hence resulting in a decreased level of performance.
In the context of ESMPs, it has been found that the type of feedback is indeed associated with differences in the idea generation process (Wooten et al., 2017). They show that directed feedback, or in other words feedback which is focused on the quality of an idea, results in the highest level of participation, and random feedback, which is described as being inaccurate, induces more participation than no feedback. The current study will extend these findings further by differentiating among the types of directed feedback. Directed feedback is focused on the quality of the final output, and as mentioned in the previous paragraph, according to the FIT there are two types of feedback which are focused on the output: task-motivation and task-learning. Therefore, this study will analyze the impact of these two types of feedback on exploratory and exploitative innovation in an ESMP environment.
Furthermore, in analyzing the effects of feedback on contribution to enterprise social media, Brzozowski et al. (2009) have found that feedback in the form of posted comments is highly correlated with a user’s subsequent participation. Additionally, they have found that visual feedback, in the form of comments, has stronger overall impact on ESMPs’ use than hidden feedback, in the form of clicks. Hence, this study will examine the effect of the two types of feedback (task-motivation and task-learning) in the form of posted comments under
a user’s post, on the propensity of the user to engage in explorative and exploratory innovation.
2.3.1. Task-motivation feedback
As task-motivation feedback is described to be directing towards increasing the effort of accomplishing a task, it is expected that receiving this type of feedback via ESMPs can have a positive impact on the generation, promotion and realization of innovative ideas. Moreover, as this type of feedback results in the (dis)encouragement and (dis)engagement of the individual receiving the feedback, depending on the level of criticism or praise of the idea, it is expected that it can have a positive impact on the generation and realization of both radical ideas which are entirely novel to the organization and refining ideas that characterize incremental innovation. Therefore, it is hypothesized:
Hypothesis 3: Task-motivation feedback positively moderates the relationship between ESMPs’ use and exploratory innovation.
Hypothesis 4: Task-motivation feedback positively moderates the relationship between ESMPs’ use and exploitative innovation.
2.3.2. Task-learning feedback
At the same time, task-learning feedback triggers cognitive processes that restructure one’s understanding. Learning occurs when individuals assimilate information, (e.g. add new information to their existing cognitive schema), or accommodate their prior knowledge to new information (e.g. re-arranging, re-organizing, or redefining existing knowledge). In order
to stimulate learning, feedback must add new perspectives (Seeber et al., 2017). As previous studies have shown that the use of ESMPs has a positive impact on extensive knowledge sharing and learning, it is expected that task-learning feedback occurs in an ESMP environment as well. This type of feedback brings new perspectives to individuals sharing their ideas and is therefore expected to have a positive impact on the generation, promotion and realization of innovative ideas. However, this type of feedback is rather focused on suggesting corrective and refining actions from the feedback giver towards the feedback receiver. Therefore, it is expected that this type of feedback can have a positive impact on exploitative innovation and a negative impact on exploratory innovation, as the latter type of innovation is characterized by radical ideas.
Furthermore, Hildebrand, Häubl, Herrmann & Landwehr (2013) have found in their study that in a social media context customers receiving community feedback directed towards the generated ideas, were less likely to come up with extreme ideas and more likely to move towards the intermediate level of self-expression. These findings can be translated to an ESMP context as in both cases the focus is on peer-to-peer interaction (consumer-to-consumer and employee-to-employee), and on the effect of receiving community feedback on the level of radicalism of ideas. Therefore, it is hypothesized:
Hypothesis 5: Task-learning feedback negatively moderates the relationship between ESMPs’ use and exploratory innovation.
Hypothesis 6: Task-learning feedback positively moderates the relationship between ESMPs’ use and exploitative innovation.
2.4. Top Management Engagement
Another variable that is expected to play a moderating role on the relationship between ESMPs’ use and innovation is top management engagement. The next paragraph will explain the findings in the current literature underlying this idea.
The level of TMS (Top Management Support) has been found to affect knowledge management practices, such as knowledge sharing within organizations through the creation of conditions encouraging employees to exercise their knowledge sharing skills and expanding the organization’s pool of knowledge (Crawford, 2005). Polities (2002) suggested that the role of leaders and top managers is increasingly changing from information and knowledge gate-keeping to encouraging knowledge sharing among all employees. This study will measure the extent of support through participation on ESMPs that top management offers their employees to share ideas and its impact on the stimulation of employees to produce and share novel ideas on ESMPs and thus, contribute to their exploratory and exploitative innovation.
One of the reasons behind this idea is the fact that transformational leadership from the top has been found to affect the promotion of innovation outcomes through the creation of relevant strategic orientations. More specifically, it can promote exploitative innovations through building market orientation and exploratory innovations through simulating entrepreneurial and learning orientation (Kraft & Bausch, 2016). Consequently, as ESMPs are implemented as tools for communication, collaboration and coordination among workers, top management can use these functions to promote their strategic orientations and foster the types of innovations needed.
Moreover, it has been shown that manager and coworker engagement on enterprise social media measured in recent activity affect individuals’ initiating or resuming participation on the ESMPs (Brzozowski, Sandholm & Hogg, 2009). Similarly, Chin et al.
(2014) have found that top management support and engagement signal the extent of how valuable ESMP’s use is and hence, directly affects employees’ willingness to engage on the platform. At the same time, they found that a lack of top management engagement on the ESMPs leads to employees’ low usage of the platform due to the fact that they perceive their activity as being monitored. These findings emphasize the benefits top management engagement can have on the engagement of other employees on ESMPs, which consequently can lead to enhanced knowledge sharing and idea generation.
Additionally, Farouk et al. (2016) found that top management support, as facilitator of knowledge sharing, has a positive impact on innovation capability, but only while being mediated by the employees’ willingness to collect knowledge (compared to knowledge donation). This is because an organization that encourages employees to share their knowledge within that organization is likely to produce new ideas and create new business opportunities, facilitating innovation capabilities and activities. Therefore, it is expected that top management engagement can have a positive impact on both exploratory and exploitative innovation outcomes via the use of ESMPs. This idea is reflected in the last two hypotheses:
Hypothesis 7: Top management engagement on ESMPs is a positive moderator of the relationship between ESMPs’ use and exploitative innovation.
Hypothesis 8: Top management engagement on ESMPs is a positive moderator of the relationship between ESMPs’ use and exploratory innovation.
2.5. Conceptual framework
The hypotheses presented above are depicted and visualized in the conceptual framework from below (see Figure 1). The next section describes the methodology of the study whereby the research design, sampling procedure and measurement of variables are being discussed.
3. Methodology
3.1. Research design
In order to assess the impact of ESMPs’ use on exploratory and exploitative innovation, as well as the moderating effects of top management engagement, motivation and task-learning feedback on these relationships, this study adopted a quantitative approach, collecting the relevant data through an online survey. The survey was created on
www.qualtrics.com and consisted of 20 closed-ended questions. The survey contained a vignette technique for measuring respondents’ perceptions of the quality of the two types of feedback in order to reduce the common method bias; therefore, two questionnaires were created with (1) a task-motivation feedback condition and (2) a task-learning feedback condition. The respondents were administered the questionnaires randomly. Both questionnaires were administered exclusively in English. In order to ensure that there are no missing points in the data, the option of “Force response” was applied in the survey, which implies that a respondent could not continue to a question without fully completing the previous one. The research time horizon of this study is cross-sectional.
3.2. Sample
The sample consisted of users of an Enterprise Social Media Platform implemented within the company they work for. To ensure that this criterion is respected, the survey was introduced with the following sentence: “Do you use Slack, Yammer, Chatter or any other Enterprise Social Media Platform at work? Then you can help us by completing the survey!”. Additionally, the following statement followed: “For all the questions on ESMPs within this survey, please focus on their use related to your working activities only (NOT for
personal projects)” (see Appendix A). As the study examines employees’ engagement and innovation on ESMPs, which are virtual platforms, there is no preset geographical boundary for the population and therefore also for the sample. This limitation will be reported in the generalizability of the study results.
The respondents were approached in three ways: privately by e-mail or LinkedIn, publicly via the social networks Facebook and LinkedIn, and indirectly via a contact person from the company they are working for. The sample size was preset at N=100 and as soon as this number of valid responses was collected, the survey was closed. Based on the descriptive
statistics of the sample (see Table 1) the mean age of the respondents was 31.6 (SDage = 9.9)
and 49% of the respondents were female while 51% were respectively male. From these, 93% have completed a university degree (Bachelor’s Degree = 56%, Master’s Degree = 37%, PhD = 0%), 79% had a job tenure of less than five years, and 72% had a work experience of less than 10 years. Regarding the engagement on the ESMPs, a cumulative percentage of 74% respondents used the ESMPs on an average of 5 hours per week or less, thus reporting a relatively moderate frequency of use. The ESMP experience was uniformly distributed across the predefined intervals.
3.3. Measurement of variables
3.3.1. Independent Variable
ESMP’s use
In order to measure ESMP’s use, the measurement from the study of Adams, Nelson & Todd (1992) which measures the general usage of Information Technology, was applied. As ESMPs are acknowledged to be a component of a company’s Information Technology portfolio, these measures can be translated into an ESMP context as well. The use of ESMP’s was measured by asking the respondents to indicate the number of hours they used the platform in the last week. The intervals are selected from the study of Olson, O’Brien, Rogers & Charness (2011), in which the Internet use among adults is measured in hours per week.
3.3.2. Dependent variables
Exploratory innovation
The measure for exploratory innovation was adopted from the study Jansen et al. (2006) (Cronbach’s α = 0.86). This measure is initially developed to assess exploratory innovation at the unit level. However, as the current study investigates the effect of ESMPs’ use on exploratory innovation at the individual level, the items have been rephrased as such. An example of a modified item is “Handling demands that go beyond existing products and services”, from the original “Our unit accepts demands that go beyond existing products and services.” (see Appendix A). The measure consists of five items which are measured on a 7 point Likert-scale ranging from “to a very small extent” to “to a very large extent”.
Exploitative innovation
Similar as for exploratory innovation, the measure for exploitative innovation was adopted from the study of Jansen et al. (2006) (Cronbach’s α = 0.80). The measure consists of five items (e.g. “Improving existing products and services for the local market” modified from the original “We introduce improved, but existing products and services for our local market.”) that are measured on a 7 point Likert-scale ranging from “to a very small extent” to “to a very large extent”.
3.3.3. Moderating variables
Task-motivation feedback
Both task-motivation and task-learning feedback were measured using a vignette approach
where respondents were randomly selected on www.qualtrics.com. One group of respondents
was presented a screenshot from a Slack News Feed to measure the perceptions about task-motivation feedback and the other group about task-learning feedback. This random selection was done to decrease the probability of common method bias occurrence.
The screenshot depicted an imaginary post shared by the survey respondent on his/her new ideas on a social media marketing campaign of the company they work at. The post was followed by three comments from fellow colleagues describing task-motivation feedback, in this case (e.g. “That’s a great idea”, “That’s very good input, keep up the good work”). The comments were reproduced from the examples presented in the study of Seeber et al. (2017). Consequently, to measure task-motivation feedback, the Feedback Quality subscale of the Feedback Environment Scale developed by Rosen, Hall & Levy (2006) (Cronbach’s α = 0.78) was adopted. The Feedback Quality measure has been extracted as it has been proven to have a significant impact (p < .01) on feedback utility (perception that feedback is
instrumental in achieving personal effectiveness at work), which is a predictor of increased frequency of feedback-seeking behavior and ultimately of improved performance (Whitaker & Levy, 2012). This measure consists of five items which were rephrased to fit the ESMPs’ context and were measured on a 7 point Likert-scale ranging from “strongly agree” to “strongly disagree”. Examples of the modified items are “I find this feedback helpful”, “I find this feedback generally not very meaningful.” Both the original and the modified versions of the items can be found in the Appendix.
Task-learning feedback
Similar to the previous measure, task-learning feedback was measured based on the Feedback Quality subscale developed by Rosen, Hall & Levy (2006). For this measure, the screenshot shown to the respondents depicted task-learning feedback and the comments were reproduced from the examples presented in the study of Seeber et al. (2017) as well (e.g. “Did you already think about how we can align these ideas with our overall marketing strategy?”, “Would you consider using A/B testing to compare several versions of the ideas?”)
Top Management engagement
Finally, the measure for top management engagement is similar to the one for ESMPs’ use and was adopted from the study of Adams, Nelson & Todd (1992). Respondents were asked to rate the perceived frequency of use of the ESMP by top management. The item was measured on a 7 point Likert-scale ranging from “never” to “a few times a day”.
3.3.4. Control variables
Overall work experience and organizational tenure have been shown to have a relationship with both employees’ ambidexterity and overall performance (Mom et al. 2006; Jansen et al.
2006). To control for their effects, this study included measures for respondents’ gender, age, organizational tenure (years) and overall work experience (years). Additionally, it controlled for the organizational tenure in a current position as according to Birkinshaw & Gibson (2004), this variable can increase one’s level of specialization and therefore might decrease the level of capability for exploratory innovation. Conversely, the level of education might have a positive impact on both types of innovation as it has been found that the higher the level of education, the higher one’s cognitive abilities to process information and learning (Papadakis et al., 1998). Replicating the method of Mom et al. (2006), the current paper controlled for the effects of the educational level by differentiating among respondents with a Bachelor’s degree or higher (Master’s and PhD) and those with a degree lower than Bachelor’s.
While some studies (Mom et al., 2006) investigate the levels of exploratory and exploitative behavior at the managers’ level, the current paper did not differentiate among a managerial and a non-managerial position. Therefore, a control variable for the job position was introduced. Additionally, the frequency with which an ESMP user shares posts/messages can impact the frequency of receiving feedback from his/her peers. To control for this variable, a question was introduced to measure the self-reported frequency of sharing posts/messages on the ESMP per week (same time scale as ESMP use). Also, as ESMPs are part of a company’s IT portfolio, the extent to which IT is generally used in daily work activities can impact the extent to which ESMPs are used. In other words, a job requiring less intensive use of IT might be correlated to a less intensive ESMP’s use. To control for this effect, a modified version of the measure from Tallon (2010) assessing the IT use was used. This item was measured on 7 point Likert-scale ranging from “to a very small extent” to “to a very large extent”. Last but not least, both platform experience and voluntariness might be correlated to ESMP’s use frequency. To control for the effect of platform experience, the
measure used by Kluger et al. (2015) was applied whereby respondents stated how long they have been using an ESMP by choosing a corresponding time interval (in months). To assess the voluntariness, which is the degree to which the use of the ESMP is perceived as being voluntary as opposed to required by the company, the paper adopted three out of four items from the measure developed by Moore & Benbasat (1991). An example of one item is “My boss does not require me to use the ESMP”. This variable was measured on a 7 point Likert-scale ranging from “strongly agree” to “strongly disagree”.
All in all, the control variables for which the effects were checked in this paper are: age, gender, level of education, organizational tenure, organizational position (manager vs. non-manager), platform experience, frequency of sharing posts on the ESMP and voluntariness, voluntariness, and IT use in daily work activities.
3.4. Statistical procedure
The results of the reliability analysis indicate a relatively high level of internal consistency for the five scales (exploitative innovation Cronbach’s Alpha = .83, exploratory innovation Cronbach’s Alpha = .73, task-motivation feedback Cronbach’s Alpha = .87, task-learning feedback Cronbach’s Alpha = .86 and voluntariness Cronbach’s Alpha = .81) (see Table 2). For the exploratory innovation scale, the removal of Item 5 (‘Searching for new distribution channels’) improves the reliability of the scale by .06. Even though this difference is in an acceptable range (Δ < .1), the Corrected Item-Total Correlation for Item 5 is low (.188). Moreover, from a theoretical perspective, the activity of searching for new distribution channels might be perceived as a component of a sales strategy rather than exploratory innovation with regard to a product or service. For this reason, Item 5 was removed from the exploratory innovation scale and the final Cronbach’s Alpha = .78.
Second, descriptive statistics, normality tests, kurtosis and skewness were computed. According to the results, the values for the ESMPs’ use frequency variable were skewed towards the interval “less than 1 hour” (Cumulative Percent “less than 1 hour” = 38%, Cumulative Percent “1-5 hours” = 72%). For this reason, the seven response options for ESMPs’ use were recoded into three categories in order to be able to identify more significant frequency patterns: 0 – “Infrequently used”, 1 – “Occasionally used” and 2 – “Frequently use” (Olson et al., 2010). To achieve a more approximately normal distribution, the interval “less than 1 hour” was recoded into 0 – “Infrequently used”, the interval “1-5 hours” was recoded into 1 – “Occasionally used”, the rest of the intervals (from “6-10 hours” to “> 25 hours”) were recoded into 2 – “Frequently use”. Similarly, the five response options for top management engagement were dichotomized as follows: “never” and “a few times a month” were recoded into 0 – “Infrequently used”, and “weekly”, “a few times a week” and “daily” were recoded into 1 – “Frequently used”.
The correlations and descriptive statistics are summarized in Table 1, along with the reliabilities of the variables. In order to test the direct relationships between ESMPS’ use frequency, exploratory innovation and exploitative innovation a one-way MANCOVA analysis was conducted. This particular analysis was chosen because the two hypotheses include one independent variable at the categorical level (ESMPs’ use frequency) and two dependent variables at the interval level (exploratory and exploitative innovation), which are significantly positively correlated (r = .61, p < .01). As the two outcome variables are correlated at a moderate level (.20 < r < .90), they meet the assumption of no multicollinearity and can therefore be tested in one MANCOVA (Bowerman & O’Connell, 1990; Menard, 1995). Besides the independent and dependent variables, the analysis incorporated the covariates top management engagement, voluntariness and job position to control for their effects, as they have been found to be significantly correlated with both
dependent variables (at p < .01). Next, to further investigate the differences among the groups of ESMPs’ use frequency with regard to their levels of exploratory and exploitative innovation, a MANOVA without covariates was conducted whereby Sheffe’s post hoc test was applied. This particular test was chosen because it allows for multiple comparisons of an independent variable with unequal group sizes across the dependent variables (Ruxton & Beauchamp, 2008).
Two additional MANCOVAs were performed with the covariates task-motivation feedback and task-learning feedback respectively, to test the moderating effects of these variables on the relationship between ESMPs’ use frequency and exploratory and exploitative innovation. To test the moderating effect of top management engagement a factorial MANOVA was conducted.
4. Results
4.1. Correlation analysis
According to the data from Table 1, it can been observed that ESMP’s use is positively and significantly correlated with exploratory innovation (r = .46, p < .01), exploitative innovation (r = .34, p < .01), top management engagement (r = .30, p < .01), frequency of sharing posts on ESMPs and negatively correlated with voluntariness. Top management engagement is positively correlated with exploratory innovation (r = .28, p < .01), but not with exploitative innovation. This might indicate that top management engagement might have a moderation effect on the relationship between ESMPs’ use and exploratory innovation, but not on exploitative innovation. Task-learning feedback is not significantly correlated with any of the outcome variables, whereas task-motivation feedback is significantly and positively correlated with exploitative innovation (r = .24, p < .05). Hence, task-motivation feedback
might have a moderation effect on the relationship between ESMPs’ use and exploitative innovation, whereas task-learning feedback is not expected to have a moderation effect on any of the outcome variables. Both exploratory and exploitative innovation are significantly negatively correlated with voluntariness and significantly positively correlated with the frequency of post sharing on ESMPs and job position. This means that the higher is an employee’s job function in the organizational hierarchy, the more he/she engages in both exploratory and exploitative innovation. Therefore, voluntariness, ESMPs posts sharing frequency and job position will be included as control variables in the MANCOVA analysis of the direct effects.
4.2. Pre-analyzing data
4.2.1. MANCOVA
Before performing the MANCOVA analysis several assumptions were tested first.
Firstly, the assumption regarding the normality of the outcome variables was checked. The results from the Shapiro-Wilk test of normality showed that both exploratory and exploitative innovation are normally distributed among the three groups of ESMPs’ use, with all p-values being non-significant (p > .05), hence fulfilling the assumption.
Secondly, the assumption regarding the absence of univariate or multivariate outliers was tested. To check for the univariate outliers, the z-scores of the independent variable ESMPs’ use were examined across the three groups, and none of them exceeded the cut-off value of 3.29, thus proving that no outliers were present. Next, to detect the multivariate outliers, the Mahalanobis measure was used. As no score exceeded the value of 13.82, it has been concluded that the data does not include any multivariate outliers either.
Thirdly, according to the Box’s test of equality of covariance matrices, the observed covariance matrices of the dependent variables are equal across groups (p = .04 < .01), thus meeting the homogeneity assumption as well.
4.3. Direct effects
In order to test the first two hypotheses which assumed that ESMPs’ use frequency has a positive direct effect on (1) exploratory and (2) exploitative innovation, a one-way MANCOVA was performed. After accounting for the effects of the control variables, a one-way MANOVA was conducted and the results from the Scheffe’s post hoc test were analyzed.
Firstly, when accounting for the effects of top management engagement, voluntariness and job position, the results of the Multivariate Tests showed that there is a significant difference between the effects of the different frequencies of ESMPs’ use on both exploratory and exploitative innovation, F (4, 194) = 6.64, p < .0001; V = 0.32, partial η2 = .12. With regard to the control variables, job position had a significant effect on exploratory and exploitative innovation, F (2, 93) = 12.20, p < .0001; Pillai’s V = 0.21, partial η2 = .21, whereas the other two control variables did not show a significant effect (see Table 3).
Secondly, to analyze how the impact of ESMPs’ use frequency differs for the two types of innovation, the Tests of Between-Subjects Effects were investigated. Based on the results, the differences in frequency of ESMP’s use have been found to have a significant effect on exploratory innovation, F (2, 94) = 11.38, p < .0001; partial η2 = .19 and on exploitative innovation F (2, 94) = 7.84, p < .001; partial η2 = .14 (see Table 4). Both partial η2 indicate large effect sizes. Therefore, it can be concluded that hypotheses 1 and 2 were supported which means that ESMP’s use has a significant main effect on employees’ extent of exploratory and exploitative innovation.
Next, in order to investigate the specific differences between the three groups of ESMPs’ use frequency on the two types of innovation, the Scheffe’s post hoc test was applied. According to the results, exploratory innovation was significantly higher for (1) frequent ESMP use compared to infrequent use (p < .0001) and for (2) frequent use compared
to occasional use (p = .024). No significant difference has been found for the levels of exploratory innovation between occasional compared to infrequent use. At the same time, exploitative innovation was significantly higher for (1) frequent use compared to infrequent use (p = .002) and for (2) occasional compared to infrequent use (p = .01). No significant difference has been found for exploitative innovation between frequent and occasional use. These differences are visualized on the plots from Appendix C.
4.4. Moderation effects
The next part will present the results from the analysis of the hypotheses 3 through 8, the moderation effects of task-motivation feedback, task-learning feedback and top management engagement on both exploratory and exploitative innovation.
4.3.1. Task-motivation feedback
In Section 2 it has been hypothesized that task-motivation has a positive moderating effect on the relationship between ESMPS’ use and exploratory innovation, and a positive moderating effect on the relationship between ESMPS’ use and exploitative innovation. As explained in the methodology section, a random part of the respondents received questions about task-motivation feedback and the other part about task-learning feedback. In total N = 51 valid data points were selected.
Consequently, a MANCOVA analysis was conducted, whereby the moderator task-motivation feedback was included as a covariate. Box’s test resulted in a significance level of p = .066 which is larger then the cut-off point recommended by Tabachnick & Fidell (2001) of p = .03, thus confirming the assumption regarding the equality of covariance. Based on
Table 3
the results of the Multivariate Tests, no significant interaction effect was found between ESMPs’ use frequency and perceptions of task-motivation feedback quality posted under the form of comments on the ESMPs, F (4, 90) = .52, p = .73; Pillai’s V = .05, partial η2 = .02. At the same time, the group who answered the questions about the perception of task-motivation feedback quality, no significant main effects were found for ESMPs’ use frequency F (4, 90) = 1.61, p = .18; Pillai’s V = .13, partial η2 = .07, or for perceptions of task-motivation feedback quality, F (2, 44) = 1.15, p = .33; Pillai’s V = .05, partial η2 = .05 on the levels of exploratory and exploitative innovation (see Table 5 and Table 6). Therefore, based on these results, hypotheses 3 and 4 are rejected.
4.3.2. Task-learning feedback
Hypothesis 5 proposed task-learning feedback to be having a negative moderating effect on the relationship between ESMPS’ use and exploratory innovation, and a positive moderating effect on the relationship between ESMPS’ use and exploitative innovation.
Similar to the testing of the moderating effect of task-motivation feedback, in order to measure the moderating effect of task-learning feedback, the relevant cases were selected from the data, resulting in N = 49. Then, a MANCOVA analysis was performed with task-learning motivation as covariate. Based on the non-significant p-value of Box’s test, p = .23 the assumption about the equality of covariance has been met. The results of the Multivariate Tests showed no statistical significance of the interaction effect between ESMPs’ use frequency and task-learning feedback quality on exploratory and exploitative innovation, F (4, 86) = .80, p = .53; Pillai’s V = .07, partial η2 = .04. Also, there was no statistical support found for the main effects between ESMPs’ use frequency, F (4, 86) = .92, p = .46; Pillai’s V = .08, partial η2 = .04 and perceptions of task-learning feedback quality, F (2, 42) = .26, p = .77; Pillai’s V = .01, partial η2 = .01 on exploratory and exploitative innovation (see
Table 5
Table 7 and Table 8). Based on these results no support has been found for hypotheses 5 and 6 and therefore, they are rejected.
4.3.3. Top management engagement
Last but not least, it has been hypothesized that top management engagement (TME) has a positive moderating effect on the relationship between ESMPs’ use and exploratory innovation, as well as exploitative innovation (Hypotheses 7 and 8). To test these hypotheses a Factorial MANOVA was conducted. Firstly, the non-significant p-value of the Box’s Test (p = .34) shows that the assumption about equality of covariance matrices was met. Additionally, the non-significant p-values of the Levene’s test (p = .12 for exploratory innovation and p = .05 for exploitative innovation) prove that the assumption about homoscedasticity has been met as well. Consequently, based on the results of the Multivariate Tests, neither TME, F (2, 93) = .66, p = .52; Pillai’s V = .01, partial η2 = .01, nor the interaction between TME and ESMP’s use frequency, F (4, 188) = 1.19, p = .32; Pillai’s V = 0.5, partial η2 = .03, have a significant effect on the extent of exploratory and exploitative innovation (see Table 9). For this reason, no moderation effect of top management engagement has been found and hypotheses 7 and 8 were rejected.
Table 7
5. Discussion
5.1. Theoretical implications and future research
In the following paragraphs the findings of this study will be discussed. Their theoretical implications will be pointed out along with some suggestions for future research. Next, the managerial implications of the findings will be presented and the paper will end with the limitations of the study.
The purpose of this study was to extend the findings of existing research which has focused on exploring the benefits and added value the adoption of ESMPs can bring to organizations. More specifically, this study analyzed the effects of implementing and using an ESMP on employees’ extent of exploratory innovation and exploitative innovation, and at the moderating effects of task-motivation feedback, task-learning feedback and top management engagement on these relationships.
The first hypothesis (H1) proposed that the more frequently an employee would use the ESMP, the higher her/his reported level of exploratory innovation will be. The hypothesis was supported at a significant level and with a large effect size. The results of the post hoc tests showed that employees using the ESMPs frequently, which refers to a frequency level of at least 6 hours per week, reported a significantly higher degree of exploratory innovation compared to the users using the platforms occasionally (1-5 hours per week) or infrequently (< 1 hour per week). There was no significant difference in the levels of exploratory innovation for employees using the ESMPs occasionally or infrequently. In other words, using the platforms at least six hours per week is positively related to a higher level of exploratory innovation, whereas any less frequent degree of use does not result in a significant difference. Two non-mutually exclusive explanations for these findings can be the following. Firstly, one might need a larger pool of collected knowledge from the ESMPs to
stimulate the development of radical ideas based on fundamentally new knowledge. Jansen et al. (2006) explain that the decentralization of communication in an organization leads to an increase in both quantity and quality of ideas and knowledge shared, increasing the likelihood of developing more radical ideas as well. At the same time, Fores & Camison (2016) refute this idea and claim that for an organization to boost employees’ degree of radical innovation, it is essential for it to combine existing internal knowledge with external sources of knowledge. Future research can further investigate these two views by studying the impact of ESMPs’ use on exploratory innovation while accounting for the moderating effect of the extent of knowledge absorbed from external sources. A second possible explanation can be the fact that a more frequent use of ESMPs leads to a significant increase in the connectedness among employees, which has been found to have significant positive effect on employees’ exploratory innovation (β = .20, p < .01) (Jansen et al., 2006). Future research can study whether connectedness has a mediating role in the relationship between ESMPs’ use and exploratory innovation.
Furthermore, the study found support for the second hypothesis (H2) as well. This hypothesis proposed that ESMPs’ use would have a positive effect on employees’ extent of exploitative innovation. Based on the results, employees using the ESMPs frequently and the ones using the platforms occasionally reported significantly higher levels of exploitative innovation compared to those using them infrequently. There was no significant difference in exploitative innovation between the groups of employees using the ESMPs frequently and occasionally. This means that using the ESMPs at least 1-5 hours per week is positively related to a higher degree of exploitative innovation.
To summarize, this study has shown that engaging on the ESMPs frequently is positively related to higher levels of both exploratory and exploitative innovation. The interaction between these two opposing but complementary types of innovation lay at the
core of ambidexterity theory which claims that for a firm to sustain a competitive advantage, it needs to simultaneously engage in exploring new opportunities and exploiting existing opportunities for innovation (Schulze, Heinemann & Abedin, 2008). Therefore, the implementation and use of ESMPs could enhance organizational ambidexterity at the employee level, which is a predictor of better organizational performance (Caniels & Veld, 2016). This has important implications as the use of ESMPs can change not only the way employees connect, exchange knowledge and engage with each other, but also the way they innovate due to the platforms’ inherent features of visibility, persistence, association, editability and self-organizing. An interesting finding from the MANCOVA analysis reveals that a higher job position in the organizational structure is a predictor of a higher extent of exploratory and exploitative innovation. However, even though there was a significant positive correlation between employees’ job position and age, age did not show a significant correlation for neither exploratory nor exploitative innovation. While findings from the literature claim that organizational tenure and overall experience are predictors of higher levels of exploratory and exploitative innovation, the current paper has found this effect for job position. A possible explanation can be the fact that a managerial or a higher ranked position is positively correlated with a higher power of decision making over which ideas turn or turn not into actual innovation, and therefore allows for more engagement in exploratory and exploitative innovation.
Next, this study has predicted three moderating effects on the relationship between ESMPs’ use and exploratory and exploitative innovation. Based on the results of the analysis none of these 6 hypotheses were supported.
Firstly, it was expected that peer task-motivation feedback and task-learning feedback on the platform would amplify the effect of ESMPs’ use on feedback receiver’s extent of exploitative innovation. Additionally it was proposed that while task-motivation feedback
would amplify the extent of exploratory innovation, task-learning feedback would reduce it. The correlation analysis has shown a significant positive correlation between task-motivation feedback and exploitative innovation, but based on the results of the two MANCOVAs the four hypotheses were not supported further. One possible explanation for these findings can lay in the complexity of the relationship between employees’ perceptions of feedback quality and actual innovative performance. Rosen et al. (2006) have found that one’s perceptions of quality feedback are a predictor of feedback utility (perception that feedback is instrumental in achieving personal effectiveness at work). However, in order for the feedback received through ESMPs to have an impact on the development of an idea generation process and innovation, the feedback receiver might need to firstly perceive learning (task-learning feedback) and effort (task-motivation feedback) instrumental in realizing performance. Dweck (1986) has developed the concept of mastery-oriented mindset, which reflects this idea of one’s belief that setting learning goals and demonstrate challenge-seeking behavior are needed to attain success. Luthans, Youssef & Rawski (2011) later applied this concept and found that mastery orientation is positively related to one’s degree of innovation (β = .10, p < .05). Therefore, future research can investigate whether peer motivation and task-learning feedback amplify/reduce the effect of ESMPs’ use on the extent of exploratory and exploitative innovation by accounting for the mediating effect of employees’ degree of mastery-oriented mindset.
Last but not least, this study predicted that under higher top management engagement on the platforms, the use of ESMPs would relate to a higher extent of both exploratory and exploitative innovation. Despite the fact that the correlation analysis showed a positive, moderate and significant correlation for top management engagement with ESMPs’ use and exploratory innovation, no support was found for its moderating effect. This finding could be related to issues regarding the construct validity of the measure for top management
