1 Master Thesis June 22nd, 2015
D.C.E. Suzanne van Strien S2590905
MSc Business Administration - Strategic Innovation Management University of Groningen, Faculty of Economics and Business Supervisors: Dr. J. Q. Dong (1st) and Dr. W.G. Biemans (2nd)
HOW DO ONLINE USER INNOVATION COMMUNITIES AFFECT THE CREATION OF CORPORATE SOCIAL
RESPONSIBILITY?
Abstract: Corporate Social Responsibility (CSR) is becoming more and more important in today’s economy. Nevertheless, research shows the negative association that firms have concerning the aim of it. Different sources claim the importance of involving stakeholders into CSR development and at the same time, markets are becoming more dynamic which also asks for firms being active in the external environment. Despite the many advantages of involving customers into the development of CSR, the literature on this phenomenon is currently not that elaborated. By performing a longitudinal study, the influence that online user innovation communities (OUIC) could have on the creation of CSR is researched. Results show that the development of corporate social innovations within an OUIC can be stimulated by the number of posted ideas, the rate of involvement from the focal firm and the number of supportive interaction moments by the platform users. Subsequently, results provide evidence on the statement that the outcomes of the OUIC positively influence the overall corporate social performance. This, since making use of an OUIC stimulates the CSR based media attention for the focal firm.
Word count: 10.685
2
1. INTRODUCTION
Within the last decades, the importance for an organization to involve in corporate social responsibility (CSR) activities has increased. Next to tighter environmental regulations concerning how firms should produce, society also expects from organizations that they take responsibility for the consequences of their business operations and act regarding their social duty (Waddock and Graves, 1997). Asongu (2007), describes that having a licence to operate, building sustainability, creating a good reputation and having an ethical duty, are the main reasons for firms to involve in CSR. Despite the fact that CSR is often seen as a head of expenditure, it could also be a source for an increased firm performance (Story and Neves, 2015; Porter and van der Linde, 1995; Saeidi et al., 2014; Zhu et al., 2013; Becker-Olsen et al., 2006). By involving CSR, the overall trust and loyalty of customers can increase and this can influence the eventual financial position of the firm (Story and Neves, 2015). A leading theory in the literature of firm performance is the resource based view (RBV). This theory can help with understanding the influence that the use of CSR could have on the firm (Branco and Rodrigues, 2006).
Currently the literature about CSR activities and the effect that it has on the performance of the firm is quite elaborated (Gollop and Roberts, 1983; Smith and Sims, 1985; Wu, 2006; Simpson and Kohers, 2002; Spicer, 1978). Different CSR based studies claim the importance of involving stakeholders in CSR practices to increase the corporate social performance (Turgsa et al., 2013; McWilliams and Siegel, 2001; Russo and Perrini, 2010). In addition to this, markets are becoming more and more dynamic and the focus on the external environment to respond to changes in the economy is critical in this context. Here, the dynamic capability view (DCV) is developed which refers to the urge of creating dynamic capabilities that are defined as “the ability to integrate, build and reconfigure internal and external competencies to address rapidly changing environments’’ (Teece et al., 1997, p.
516).
The claims to incorporate stakeholders into the development process together with the urge to focus on the external environment to be able to respond to fast moving changes, make researching CSR in an open innovation context interesting. Additionally, firms see developing CSR as quite a challenging task and in this context literature has not yet recognized one specific approach for improving corporate social performance (Chaudhri, 2006). For bigger firms to build a CSR strategy, they need to take care of the corporate consistency so that the firm communicates one specific CSR strategy that is the same in all the areas where the firm operates. Here, online tools like ‘online user innovation communities’
(OUIC) can be considered as a possible useful method to create this specific CSR strategy. These communities are focused on getting input from customers about new products/services/activities for the firm to focus on and can also be used to develop CSR practices.
3 A good example of such a community is the one of Starbucks named ‘MyStarbucksIdea’. Starbucks has been able to listen to the ideas of customers concerning CSR and launched different initiatives like the partnership with the charity named [RED] and a special food donation program to stop wasting pastries (Starbucks, 2015a). Subsequently, developing CSR through OUICs can be fruitful since prior research revealed that customers sometimes doubt the motivations of organizations to work on CSR (Skarmeas and Leonidou, 2013; Jahdi and Acikdilli, 2009) and therefore involving the customer in the CSR related innovation process could both increase corporate social performance and the satisfaction of the customers.
Currently, no empirical research has focused on the influence that the development of CSR through OUICs could have. Above reasons make researching this, at the current dynamic times where everybody is online, highly relevant for the evolution of the CSR literature. Therefore, the purpose of this study is to discover the influence that an OUIC could have on the creation of CSR.
This study consists of two main contributions to the CSR literature. First the platform components are researched. This to find out what conditions create the optimal OUIC for specifically developing CSR based innovations. Here different determinants concerning user interactions are researched. Secondly, the study contributes by researching the influence of OUIC based CSR innovations on the focal firms’
corporate social performance. This is done by researching the relationship between the number of ideas that are launched through the OUIC and the effect of these innovations on the number of traditional media expressions related to the firm and its performance on CSR. Above mentioned contributions are highly relevant for the CSR literature since it can provide firms insights into the relevance of OUICs within the context of developing corporate social performance. Subsequently, it also provides strategic understanding of the influence that customers could have within the development of the CSR construct. In sum, this research will focus on the following research question:
To what extent can a firm make use of online user innovation communities (OUICs) to create corporate social performance?
To research this question, data is gathered from the ‘MyStarbucksIdea’ community within a timeframe of 43 months. Subsequently, news articles from LexisNexis are used to indicate the corporate social performance of the focal firm. This master thesis is divided into six chapters. First the existing literature will be discussed in the literature review followed by several hypotheses in the theory development section. After this the methodology explains how the research is conducted, followed by the results. Finally the discussion and conclusion explain the main contributions together with the limitations and further research directions.
4
2. LITERATURE REVIEW
Corporate Social Responsibility
Over the last decades, corporate social responsibility (CSR) is viewed from different perspectives.
Friedman (1970) aims that in this world, no firm is willing to act in favour of society “[…] there is one and only one social responsibility of business- to use its resources and engage in activities designed to increase its profits so long as it stays within the rules of the game, which is to say, engages in open and free competition without deception or fraud” (Friedman, 1970, p. 6). In essence, Friedman states that being social responsible involves only acting in line within the legal system. Other, more contrary, views aim that firms are responsible for social problems and need to make sure that they fulfil their job of being a corporate citizen by engaging in voluntary activities (Hay et al., 1976;
Ackerman and Bauer, 1976). These contrary views on the construct make it hard to create one definition of CSR (Scherer and Palazzo, 2011) and this is also the cause for the fact that the evolution of the academic literature on CSR is not that elaborated (De Bakker et al., 2005). More recent views have been developed where the involvement in not-for profit-activities are more important. According to McWilliams and Siegel (2001) CSR can be described as “situations where the firm goes beyond compliance and engages in actions that appear to further some social good, beyond the interests of the firm and that which is required by law’’ (p. 22). They describe CSR as different social characteristics that can be integrated in activities like product and manufacturing processes, human resource, improving environmental attention and working together with other external parties on related problems (McWilliams and Siegel, 2001).
Often firms do not see the urge of involving CSR practices into their organization and doing so is expected to be very costly without any return on investment. Nevertheless, current research is proving these statements wrong. Lii and Lee (2012) for example, empirically test the influence of CSR and find that CSR reputation has a moderating effect on the relationship between CSR initiative and consumer attitudinal evaluations. Subsequently, Saeidi et al. (2014) found empirical evidence for CSR as influencer of the firms’ reputation and competitive advantage by sampling manufacturing and consumer product firms. A very extensive part of the existing literature on CSR focuses on the outcome of CSR activities, involving the construct CSR into the firms’ overall strategy. Where some focus on the financial pay-offs of having a CSR focussed strategy (McGuire et al., 1988; Saeidi et al., 2014), others take non-financial achievements (Zhu et al., 2013; Becker-Olsen et al., 2006), like brand loyalty, as focus area. These different movements have created a separate field that emphasises on the outcome of CSR and developed a new construct named: corporate social performance (CSP).
5 Resource Based View
Different sources strike the importance of involving CSR into the corporate strategy in order to increase the performance of the firm (Saeidi et al., 2014; Zhu et al., 2013; Becker-Olsen et al., 2006).
McWilliams and Siegel (2001) evaluated the demand for CSR and the eventual costs of satisfying these demands and found that eventually these costs will be overruled and the private returns will be increased. Porter and Kramer (2006) argue that when organizations are able to manage CSR in the right way it can be a source of innovation and a source for creating a competitive advantage.
McWilliams and Siegel (2001) agree with this and state that CSR can be used as an organizations’
differentiation strategy to create and possibly maintain a competitive advantage. A good CSR strategy could influence corporate reputation which is important since it helps to reduce transaction costs and it influences trust and loyalty of the end-user which eventually could lead to a higher overall firm performance (Zhu et al., 2013; Walsh and Beatty, 2007). McWilliams and Siegel (2001) argue that CSR can be used for profit maximizing. They compared products of which one product involved a social attribute and the other not. Eventually they found that the product that involved the social construct brought more revenues.
When researching the components that can help creating and improving the firm performance, the resource based view (RBV) is a well-known and an often used theory (Barney, 1991; Costa et al., 2013; Sun and Tse, 2009; McWilliams and Siegel, 2010). The RBV argues that organizations need to focus on internal resources which include “all assets, capabilities, organizational processes, firm attributes, information, knowledge etc. controlled by a firm to conceive of and implement strategies that improve its efficiency and effectiveness’’ (Barney, 1991, p. 101). Here, an organization can obtain a competitive advantage when it succeeds in managing a value creating strategy which is not used by any other firm. Part of building this competitive advantage could be the CSR performance and therefore the current research uses the resource based view as a theoretical baseline to measure the outcome effects of involving CSR into the firm.
Remarkable is the high number of articles that claim the importance of involving stakeholders when searching for successful CSR strategies. Torugsa et al. (2013) state that CSR can be used to create a sustained competitive advantage for the firm by focusing on shared vision, stakeholder management and strategic proactivity. Further, McWilliams and Siegel (2001) urge the importance of meeting the demands of the relevant stakeholders for increasing firm performance. Lastly, Russo and Perrini (2010) focus on the importance of interrelated relationships between the focal firm and the stakeholders which is explained in having a moral obligation towards the stakeholders as well as selecting stakeholders who are in strategic line with the focal firm. The importance of involving the stakeholders can be illustrated by taking the example of product stewardship (Hart, 1995).
6 Every production process, where raw materials are transformed into end products, have some influence on the environment. When focusing on this process to improve the firms’ performance, via CSR performance, it will mainly involve changing the process where stakeholders are involved (Hart, 1995).
Dynamic capabilities
Within today’s economy, more and more markets are becoming dynamic and the competitive landscape is shifting. This change makes it more urgent for organizations to create dynamic capabilities which are defined as “the ability to integrate, build, and reconfigure internal and external competencies to address rapidly changing environments’’ (Teece et al., 1997, p. 516). Teece et al.
emphasise that the creation of dynamic capabilities needs to be supported by internal as well as external sources. Dynamic capabilities are widely recognized as critical for firm performance since it can be of influence in the process of converting resources which eventually lead to a higher performance of the overall organization (Lin and Wu, 2014; Elsenhardt and Martin, 2000; Teece et al., 1997).
According to Robert et al. (2008), the ability to integrate knowledge of individuals is one of the most important factors for creating dynamic capabilities. The importance of involving the customer in the innovation processes on the final innovation outcome has been tested positively multiple times (Gupta et al., 2000; Fritsch and Lukas, 2001). Additionally, Porter and van der Linde (1995) argue that CSR related innovations can improve the product consistency plus the overall quality and these innovations could stimulate resource productivity by reducing the use of unnecessary materials. Next to the fact that individuals are important for an organization, the learning theory developed by Wenger (1998) states that when organizations want to learn they should participate in a social environment. Finally research also suggests that making use of external sources to increase the innovation rate may be more valuable than internal ones (Laursen and Salter, 2006). As for which involving customers into the development process is beneficial for dynamic capabilities, it could also be beneficial for the development of CSR performance. As stated before, involving the main stakeholders is defined as a valuable source for developing CSR that could eventually help increasing the firm performance.
Open Innovation
One type of stakeholder is the customer and this group can be involved into the CSR development process via open innovation (OI). Jespersen (2011) argues that making use of OI may shorten the distance between the organization and their customers and influences the involvement and engagement of the customers in the organization. OI can be defined as “the use of purposive inflows and outflows of knowledge to accelerate internal innovation, and expand the markets for external use of innovation, respectively” (Chesbrough et al., 2006, p. 1).
7 Within these outflows and inflows, different types of open innovation can be explained namely the outside-in process, the inside-out process and the coupled process (Enkel et al., 2009). First, the outside-in process can be explained by purely using external sources like suppliers and customers when expanding the current knowledge base. On the opposite, there is the inside-out process which involves selling ideas to the market by for example licencing agreements. Lastly, the coupled process of OI refers to the collaboration in a fixed partnership which could be an alliance or joint venture. In this type of OI all involved partners should give and take in order for the partnership to be fruitful (Enkel et al., 2009).
Online User Innovation Communities
When referring to open innovation as a mechanism for developing CSR we can use the knowledge that is created through the use of online user innovation communities (OUICs). These online communities are an outside-in form of OI that uses the ideation of the users on the platform. They can be defined as
“distributed groups of individuals focused on solving a general problem and/or developing a new solution supported by computer mediated communication” (Di Gangi and Wasko, 2009, p. 303).
Despite the fact that the OUIC is sometimes associated with crowdsourcing, there is a clear difference between these two constructs. Where in the OUIC the collaboration is based on gaining ideas from users within an variety of subjects (as long as it is connected to the focal firm) through giving users the opportunity to post ideas and respond on others ideas, the focus of crowdsourcing lies on obtaining the crowds’ opinion about one main subject (Martinez-Torres, 2014). Organizations that operate in OUICs build brand loyalty, increase market penetration and positive word of mouth advertising (Kim et al., 2008).
Laursen and Salter (2006) argue that open innovation is recognized as a remarkable way of searching for new ideas. In this context product search is defined as “an organizations’ problem-solving activities that involve the creation and recombination of technological ideas’’ (Laursen and Salter, 2006, p. 134). Hereby they distinguish two sorts of product search namely the ‘depth’ and ‘breadth’.
Whereas search breadth is related to the number of sources that are used to search for new ideas, search depth relates to the amount of knowledge that is drawn from one source. Laursen and Salter refer to these search breadth and depth as main influencers of the innovative performance of an organization and therefore organizations need to develop a well thought search strategy that fits the type of environment where the firm is active in. When specifically evaluating above information on OUICs, Dong and Wu (2015) state that the number of people on these platforms can help developing an posted idea of one individual. This construct is called the ideation capability which is defined as “a firm’s ability to collect user-generated ideas such as posts and comments about potential innovation from its OUIC” (Dong and Wu, 2015, p. 115).
8 Making use of OUICs provide the opportunity to jointly develop innovation activities which are normally having an economic and rational loading. This means that firms are mostly focusing here on improving their sales revenues (Charalabidis et al., 2014). In the context of the current research, the innovations created are focused on improving social related issues of which other studies refer to
‘social innovation’ and ‘corporate social innovation’ (Sanzo-Perez et al., 2015; Herrera, 2015). Sanzo- Perez et al. (2015) define social innovation as “the development and implementation of new ideas (products, services and models) to meet social needs and simultaneously create new social relationships or collaborations” (p. 431). The current research refers to corporate social innovations as innovations who are developed for social goods and are focused on improving the corporate social performance of the focal firm.
3. THEORY DEVELOPMENT
Former research draws the importance of OI as a source for creating dynamic capabilities that can help improving the competitive advantage of the firm. Nevertheless, the combination with CSR activities as main source for the innovation is not that elaborated. Making use of OI channels like an OUIC can be seen as a useful tool for several reasons. First, the involvement of customers in the social innovations can make customers be more loyal and more willing to pay for the product (Jahdi and Acikdilli, 2009).
Also, Peloza and Papania (2008), explain that when organizations are more involved within the CSR needs of their close stakeholders, it can eventually have a good influence on the firms’ performance.
The coming sections will first explain the stated hypothesis concerning the creation of corporate social innovation through the use of an OUIC and secondly the influence that this could have on the eventual corporate social performance of the focal firm.
User ideation
When referring to OUICs, the influence of network effects can be recognized. There needs to be a critical mass of users that are using the platform, for the platform to be of interest for the user (Moon and Sproull, 2008). Mc William (2012) refers to the importance of active participation of the users in an online community to be recognized as a community and mentions that organizations should also stimulate this participation. As mentioned in the literature review, Laursen and Salter (2006) refer to the importance of search breadth when there is need for innovation and explain the positive effect of the number of different sources on the innovation rate of an organization. Within this research the term
‘user ideation’ is used which refers, as stated in the literature review, to the idea generation of product users (the number of ideas that every user posts on the platform) (Dong and Wu, 2015).
9 The current research expects that the number of sources from which an organization can get information (which is in this case the number of ideas) will have a positive influence on the eventual creation of corporate social innovations. The following hypothesis for user ideation to develop corporate social innovation is developed:
H1 User ideation within an OUIC has a positive effect on the creation of corporate social innovation that is derived from the OUIC.
User interaction
Another main aspect of an OUIC contains the user interaction of which firms can easily measure the needs and preferences of their customers. Füller et al. (2009) describe that the internet created more possibilities to increase the customers’ empowerment. Here they refer to empowerment as “[…] how the new technologies enable people to interact with the world on different levels (personal, dyad, group or community) and to do or achieve things that they found difficult to do or achieve before’’
(Füller et al., 2009, p. 75). Hagel and Armstrong (1997) state that feedback can be seen as a main instrument for innovation but also refer to the fact that the platform is open in which other parties can possibly profit; “the secret of success will lie not in ownership of information but in its application”
(Hagel and Armstrong, 1997, p. 146). West and Lakhani (2008) and Wegner (1998) refer to the importance of feedback mechanisms as in which interactions among the platform users are the main ingredient of a firm to create a learning environment. Hypothesis two concerns the important role of the feedback mechanism:
H2 User interaction within an OUIC has a positive effect on the creation of corporate social innovation that is derived from the OUIC.
User Co-development
The study of Lu et al. (2011) provides evidence for the fact that the existence of a feedback mechanism is essential for innovations. Results of this study shows that the number of adopted ideas within a platform are 77% lower when the feedback mechanism is missing. Also Franke and Shah (2003) stress the importance of interaction within community and refer to the fact that the most innovative ideas came from users that were the most active in interacting with others on the platform (by responding on other ideas). Next to this, Franke and Shah state that the initial development of the idea can be done by one person, but to make the initial idea into something usable for the organization it needs the help of other users: “[…] user-innovation is not an individual task but a joint effort”
(Franke and Shah, 2003, p. 164). Di Gangi and Wasko (2009) state that the focal firm will feel more urgency towards adopting the ideas that get the most votes from the crowd and therefore it is likely that negative comments on ideas will have a negative influence on the possibility that an idea will be implemented.
10 Next to this, the involvement of the focal firm in the innovation process is also of importance to optimize the environment for innovation (Von Hippel, 1976; Foss et al., 2011). In conclusion, the following hypotheses concerning co-development are established:
H3a Supportive interactions on posted ideas within an OUIC will positively influence the corporate social innovation that is derived from the OUIC.
H3b Negative interactions on posted ideas within an OUIC will negatively influence the corporate social innovation that is derived from the OUIC.
H3c Focal firm interactions on posted ideas within an OUIC will positively influence the corporate social innovation that is derived from the OUIC.
H3d Interactions on the OUIC between users and the idea generator will positively influence the corporate social innovation that is derived from the OUIC.
Corporate social performance
As explained in the literature review, the relationship between corporate social innovation and the influence of this on the corporate social performance is expected to have a positive influence. First, different sources describe the importance of involving stakeholders when developing CSR (Torugsa et al., 2013; McWilliams and Siegel, 2001; Russo and Perrini, 2010). Additionally, externally developing CSR is beneficial for developing dynamic capabilities which gives the firm the opportunity to respond easier on changes in the environment (Teece et al., 1997).
H4 Corporate social innovations that are developed through an OUIC will positively influence the corporate social performance of the focal firm.
Figure 1: Conceptual framework
11
4. METHODOLOGY
Research design
During the research two different sources are used to collect data from which are the OUIC of Starbucks called ‘MyStarbucksIdea’ and LexisNexis to get insight into the media expressions concerning the corporate social performance of the firm. The current research can be recognized as a longitudinal study since it analyses data within a time period of 1263 days (43 months) starting from March 18th, 2008 (the launch date of the platform) up till September 2nd, 2011.
Research site
In order to test the hypotheses, the multinational Starbucks is selected for analysis. Starbucks is an international coffee company that is active since 1971 and operates in 65 countries with more than 21.000 stores (Starbucks, 2015b). Since 2008 Starbucks is having its own online innovation platform named ‘MyStarbucksIdea’ (Starbucks, 2015c). Appendix two (p.29) shows an infographic with the achievements of the platform within the first five years. The platform is divided into three parts; an area for posting new ideas, an area for viewing ideas in which the user can also comment and vote on other ideas and lastly the ‘ideas in action’ area where users can see which ideas are truly executed by Starbucks. On the platform, customers can submit ideas that they have for Starbucks with regard to all sorts of categories, for example ideas about new coffee flavours or ideas about creating a new experience. One specific category is about social responsibility and here people can submit ideas that are in line with this concept. Currently 11,106 ideas are submitted in this section and 97 ideas from all the submitted ideas are launched by Starbucks. In total, 5157 ideas have been analysed within the assigned timeframe which are all part of the ‘social responsibility’ section on the ‘MyStarbucksIdea’
platform.
Starbucks is seen as a firm with a leading OUIC and is used in multiple studies that are focussed on innovation, dynamic capabilities and firm related performances (Dong and Wu, 2015; Lee et al., 2014). Starbucks as focal firm will fit this research well since the platform is sufficiently developed with multiple interactions each day. Next to this, Starbucks is focussing on acting as a social responsible organization whereas they are one of the biggest Fairtrade certified purchasers and have also developed their own policy called ‘coffee and farmer equity’ (C.A.F.E.) to make sure that the coffee is ethically sourced (Starbucks, 2013). The fact that Starbucks is very active in CSR related activities creates a useful environment to test the influence of an OUIC on the overall CSR performance over time.
12 Measures1
Even though the conceptual model indicates a mediating effect for corporate social innovation, the research does not treat it as one. As the theoretical sections describe, two different relationships are expected. First the relationship between the OUIC components and the influence of this on the creation of corporate social innovation and secondly the relationship between the corporate social innovations that derive from the OUIC and the focal firms’ corporate social performance. In short, the variable social innovation will be used as the dependent variable for one part of the research, and as an independent variable for the other part.
Dependent variable: Corporate Social Performance
In this research the corporate social performance of the focal firm will serve as the dependent variable whereas only the non-financial performance is taken into account. More specifically, the corporate social performance of the focal firm is evaluated in which the focus will lie on news publications.
Different sources refer to measuring media attention when researching performance. Zyglidopoulos et al. (2012) state that more media attention will have a positive influence on corporate social performance. Further, Winter and Eyal (1981) found that the concerns of the public about social issues were significantly correlated with the number of news articles. The corporate social performance is measured by making use of LexisNexis. This database shows all published news articles of traditional media and gives the possibility to measure the corporate social performance on a daily basis. First, relevant articles were selected by filtering on the right time period (18/03/2008 – 02/09/2011) and publications related to CSR and Starbucks. The keywords that were used for this search were
“Starbucks” and “social responsibility”. For this specific search 786 articles were shown and used for further research. During the second step, articles were reviewed on behalf of their relevance, which meant that only the articles that had as main subject CSR in combination with Starbucks were selected which resulted in 297 articles. Next to this selection, some articles turned out to be duplicates of each other and these were also removed which resulted in the final number of 234 articles. All these articles were counted on the day that it was published and for the analysis it was measured if the number of published news articles per day was influenced by the number of social innovations that were developed via the OUIC. When performing the analyses, a time lag of 1 day was used, because it is impossible for newspapers to publish the news on the same that that it was posted. Measuring corporate social performance via traditional media is proven to be very reliable since the news articles are posted by different newspapers. Therefore the outcomes do not depend on what only one source claims.
1 Other studies that researched OUICs used characteristics of users like age and gender as control variables for their research. Unfortunately this information was not openly available on the platform of MyStarbucksIdea (user names were available but often it was hard to identify the gender of this user) and therefore no control variables are involved.
13 Dependent/Independent variable: Corporate Social Innovation
Corporate social innovation is measured by analysing the number of ideas opted by the platform users on the platform within the pre-defined timeframe. Next to that, these launched ideas are coming from ideas that were posted in the social responsibility section. To identify which ideas were launched two different methods were used. First, the ideas that were marked with a checkmark on the platform, meant that they were identified by Starbucks as launched and so these were counted on the day of launch. Secondly, as described earlier, one part of the ‘MyStarbucksIdea’ platform contains the ideas that are launched. Within this page, the launched ideas in the social responsibility category in the targeted period were included in the research. In order to relate the launched ideas to the time when the idea was opted, a ‘time lead’ was used in SPSS. To make sure that the right time lead was used, a random sample of ten launched ideas was used to measure the number of months between the post of the idea and the launch moment by Starbucks. Evaluating these ten launched ideas gave an average period of 3.85 months and therefore a time lead of 4 months was used.
Independent variables: user ideation, user feedback and user co-development
As has been stated before, the current literature on feedback mechanisms and OUIC describes different factors that can influence the creation of innovations. For this research, three different independent variables are used which are counted on a daily basis.
1. User Ideation
For all of the independent variables, data is gathered within the given time period. User ideation consists of the total number of ideas that were posted in the social responsibility section, on a daily level. In total 5157 ideas were analysed with a mean of 4.08 ideas per day.
2. User Feedback
Users of the ‘MyStarbucksIdea’ platform are able to respond to every idea that is posted (when they are logged in). Concerning the rules of conduct there are some cases when user feedback gets deleted:
“[…] we will not allow profanity or other inappropriate conduct. Posts that are inappropriate will be removed from the site and users that repeatedly post inappropriate conduct will be asked not to post at all’’ (Starbucks, 2015d). The user feedback construct is measured by counting the total number of comments with all sorts of emotional loadings (negative, neutral and positive) on the involved ideas each day in the given period of time.
3. User Co-development
The variables of the construct co-development are part of the variable ‘user feedback’ and these variables contain the emotion of every comment given by idea developers, platform users and employees of Starbucks.
14 As has been measured, feedback moments can involve different emotions like positive, negative, neutral and supportive. In this context supportive comments involve feedback that is aimed to further develop the original idea by giving suggestions to further improve the idea. These supportive comments can give a positive, negative but also neutral comment about the idea, as long as it gives a further suggestion for improvement. Further, another variable used in the research is the total number of negative comments per day since previous research states that this is important when measuring the success of an OUIC.
The literature on platform interaction and feedback constructs also show that the person who writes the comment can be of influence on the eventual innovation rate (Franke and Shah, 2003; Von Hippel, 1976; Foss et al., 2011) and therefore the number of times per day that Starbucks and the idea generator itself responded on the idea was measured. To avoid possible biases of multicollinearity (since this data comes forward from the variable ‘user feedback’), the variables were counted on a daily level first and after this the percentage of each variable on user feedback was calculated.
5. RESULTS
5.1 Descriptives
Table 1 gives an overview of the descriptives of the different used variables. As can be seen, the standard deviations of the variables ‘number of ideas’ and ‘number of comments’ are rather high. This means that the data contains a lot of variation which is certainly the case since the variance in comments and ideas per day could differ a lot. When analysing the variables further, some extreme values and outliers were detected but since the dataset is elaborated and removing outliers would only detect new ones, the decision was made to not delete them. Subsequently, deleting these variables meant that a lot of valuable data was removed and therefore the decision is made to include them.
Obs. Mean Sum SD Min Max
Number of ideas 1263 4.08 5157 7.540 0 131
Number of comments 1263 7.62 9624 15.683 0 435
Number of launched ideas 1263 .02 26 .142 0 1
Number of negative interactions 1263 .99 1248 2.100 0 46
Number of supportive interactions
1263 .88 1117 2.966 0 89
Number of idea generator interactions
1263 .08 99 .318 0 4
Number of Starbucks interactions
1263 .39 498 1.058 0 18
Number of news articles 1263 .19 234 .502 0 5
Table 1: Descriptives
15 5.2 Reliability analysis
To examine multicollinearity, the variance inflation factors (VIF) were tested and results showed that it was below the limit of 10. The outcomes ranged from 1.052 till 4.619 and therefore no multicollinearity problems were found. Also the tolerance variable is for every variable above 0.1 which also proves that no multicollinearity exists (Menard, 2002). Next to this, the scatterplot indicated that the variables in the regression were normally distributed.
5.3 Correlations
First, the correlations between the included variables were tested via a bivariate correlation. The results reported in table 2 show different significant correlations. The results in the correlation matrix are only providing insights into which variables are correlated with each other but give no insight into which direction they are correlated. First, the results show a significant positive correlation between the total number of ideas and the total number of comments r(1263) = .49, p= 0.01. Next to this correlation also some other correlations are statistically significant but these effects are not that strong.
The most important result of this analysis is that except for the positive relationship between launched ideas and number of news articles r(1263) = .069, p=0.05, none of the variables that are determined in the hypotheses to influence each other are statistically significant correlated.
5.4 Regression analyses
To test the effect of the OUIC on the eventual corporate social performance, single and multiple regression is used to test the relationships. The conceptual model will be divided into two parts that are analysed independently from each other. First, regression analyses were performed to measure the effect that the predictors related to the OUIC had on the total ideas that are launched. Equation [1]
shows this model with the dependent variable ‘LAUNCHED’ (total number of launched ideas with a time lead of 4 months) and independent variables ‘IDEA’ (total number of ideas), ‘COMMENT’ (total number of comments) and ‘INTERACT’ (co-development variables). The second equation [2]
measures the effect of the launched ideas (LAUNCHED) on the dependent variable ‘TME’ (traditional media expressions) with an added time lag of 1 day.
LAUNCHEDt-1 = 𝛽0 + 𝛽1IDEA + 𝛽2 COMMENT + 𝛽3 INTERACT + ɛ [1]
TMEt+1 = 𝛽0 + 𝛽1LAUNCHED + ɛ [2]
Model one, two and three in table 3 are conducted to research which constructs influence the amount of CSR ideas that are launched within an OUIC. Secondly, table 4 shows the performed test to see whether the corporate social innovations derived from the OUIC influence the corporate social performance of the focal firm.
16 5.4.1 The effect of OUIC variables on the corporate social innovations created
User Ideation
Model 1 in table 3 shows that the amount of ideas that are posted on the platform have a positive influence on the ideas that are launched. This relationship is statistically significant (B = ,001) at a p<0.05 significance (sig = ,013). Also the significance level of the model, which is defined by the F significance value is accepted (P < 0.1). Nevertheless, the positive influence of the amount of ideas stays rather low since the model shows that the R2 is only 0.005, which means that only 0.5% of the total variance on launched idea is explained by an increase in ideas posted on the platform. Despite the low influence of this variable on the launched ideas, we can accept hypothesis 1 which stated that an influence in the amount of ideas posted on an OUIC positively influences the ideas that will be launched on the platform.
User feedback
The second possible influencer of the CSR related ideas that are launched on the OUIC is the total amount of feedback that is given on the ideas. As can be seen in table 3, the results of the second model show a negative relationship between the feedback mechanism and the ideas launched (B = -,005). This negative relationship is not statistically supported (sig= ,967) and therefore the second hypothesis, which stated that an influence in the total number of feedback interactions within an OUIC positively influence the ideas that will be launched on the platform, is rejected.
User co-development
As the theory argues, the type of feedback and the type of interactor on the platform, can also influence the ideas launched on the platform. These relationships are tested in the third model of table three. When first looking at the type of feedback, the regression results support the stated hypothesis 3a, and shows that negative interaction on an idea has a statistically significant negative influence on the number of ideas that are launched (β = -,062). The relationship is proved to be significant at a p<0.1 significance level. Secondly, hypothesis 3b argued that supportive feedback will have a positive effect on the number of ideas that are launched within the OUIC and also this hypothesis is statistically significant on a p<0.1 significance level (β = ,057). When evaluating the influence that the type of respondent on the posted ideas could have, hypothesis 3c argued that the involvement of the focal firm within the idea interaction, will positively influence the number of ideas that are launched within the OUIC. As can be seen, the model shows that this relationship is confirmed (β= ,061) on a statistical significance level of p<0.1. Next to the influence of the focal firm and the crowd, also the idea generator itself can interact on its own idea and the relationship of this interaction on the number of ideas launched is expected to be positive. The result of the regression nevertheless shows a negative relationship (β= ,029) but this is not statistically supported (sig= ,381), therefore h3d is not supported.
Striking is the overall model variance of 0.8% (R2 adjusted= ,008) which is rather low.
17
Table 2 Bivariate Correlation Matrix
1 2 3 4 5 6 7 8
1. Total ideas 1
2. Total feedback ,494** 1
3. Negative interaction -,072* -,007 1
4. Supportive interaction ,040 ,008 -,105** 1
5. Idea generator interaction ,014 -,020 -,045 ,010 1
6. Starbucks interaction ,069* -,083** -,185** -,103** ,017 1
7. Launched ideas ,014 ,022 -,027 -,033 ,018 ,007 1
8. News articles ,15 -,006 -,026 ,075* -,005 ,032 ,069* 1
** Correlation is significant at the 0.01 level (2-tailed).
* Correlation is significant at the 0.05 level (2-tailed).
Table 3 Regression coefficients with ideas launched2 as dependent variable
Variables Model 1 Model 2 Model 3
B Stand. β B Stand. β B Stand. β
Total ideas ,001**
Total feedback -,005
Negative interaction -,062*
Supportive interaction ,057*
Starbucks interaction ,061*
Idea generator interaction -,029
R2 ,005
,005 6,206*
,000 ,013
Adjusted R2 -,001 ,008
F (p) ,218 2,925*
*Correlation is significant at the 0.1 level
** Correlation is significant at the 0.05 level
2 The variable ‘launched ideas’ is performed with a time lead of four months
18 5.4.2 The effect of OUIC based corporate social innovations on the corporate social performance of the focal firm
Hypothesis 4 stated that the number of CSR ideas launched within an OUIC would positively influence on the focal firms’ corporate social performance. As stated in the method section, within this research the corporate social performance of the firm is measured by evaluating news articles that are in relation with CSR and the focal firm. The regression results as reported in table 4 found a significant (p <0.05) positive relationship between the number of ideas launched on the OUIC and the number of articles that were published with as main subject Starbucks in relation with CSR (B= ,203).
This result indicates that the number of news articles that are posted (with as main subject Starbucks and CSR) are positively influenced by the number of launched ideas from the OUIC and therefore we can accept hypothesis 4.
Table 4 Regression coefficients with number of articles3 as dependent variable B
Launched ideas ,203**
R2 ,003
Adjusted R2 ,003
F (p) 4,179**
*Correlation is significant at the 0.1 level
** Correlation is significant at the 0.05 level
Table 5 summary of proposed hypotheses
Relationship Hypothesis Supported/not supported
Relationship Significance level
User Ideation – Corporate Social Innovation H1 Supported + **
User Feedback – Corporate Social Innovation H2 Not supported n.s.
Supportive Interaction – Corporate Social Innovation
H3a Supported + *
Negative Interaction – Corporate Social Innovation
H3b Supported - *
Focal Firm Interaction – Corporate Social Innovation
H3c Supported + *
Idea Generator Interaction – Corporate Social Innovation
H3d Not supported n.s.
Corporate Social Innovation – Corporate Social Performance
H4 Supported + **
*p < .1 **p <.05
3 The variable ‘number of articles’ is performed with a time lag of one day
19
6. DISCUSSION & CONCLUSION
6.1 Main findings and contributions
This research aims to discover the extent to which online user innovation communities (OUICs) can help strengthen the overall corporate social performance. The findings present two main contributions to the CSR literature. First, results from the regression analysis provide insights in the community components that are critical for stimulating corporate social innovation within an OUIC. Secondly, the results provide statistical evidence for the influence that corporate social innovations that are developed via an OUIC can have on the firms’ overall corporate social performance.
Different community components turned out to be important when seeking to make optimal use of an OUIC when developing CSR. Results showed that the number of ideas that are posted on an OUIC have a positive effect on the eventual innovation rate. This result supports the existing innovation literature that state that a firm should have a certain search breadth of different sources when searching for innovations (Laursen and Salter, 2006) and provides hereby evidence for the CSR context.
Subsequently, this outcome can be linked to the theory of network effects where first a minimum number of users should be active on the platform for the platform to start creating value (Moun and Sproull, 2008).
A very extensive part of the platform consists the different types of feedback interactions. Findings show that the influence of the total number of feedback moments (negative, positive, supportive or neutral) per day on the development of corporate social innovation within the OUIC was not significant. A possible reason for this could be the fact that the values of feedback per day were too much distributed. Results on the influence that supportive feedback could have showed a positive significant effect. This means that when people are co-developing together to improve a posted idea by giving suggestions for improvement, the possibility that the idea is going to be launched, will increase.
This result is consistent with the OUIC literature (Franke and Shah, 2003; Lu et al., 2011) and provides support for the CSR literature when developing corporate social innovation. From this result it could also be likely to conclude that making use of OUIC seems to be more useful than making use of a joint venture when developing CSR. This because the result argues that the development of corporate social innovation needs to be done via the interaction of multiple users and is not just a result of a dyadic interaction. Important is the result that negative feedback proved to be of negative influence on the number of corporate social innovations. This is consistent with the literature of OUIC which states that only the ideas that get the most votes are having the attention of the focal firm (Di Gangi and Wasko, 2009). Concerning the influence of the focal firm on the platform, results showed a positive relationship between the involvement of the focal firm and the number of ideas that were launched.
20 This result is in line with OUIC research (Von Hippel, 1976, Foss et al., 2011) which states that the involvement of the firm in the platform is of sufficient importance and provides evidence that firm involvement is also crucial in the development of CSR.
When measuring the possible link between OUIC based corporate social innovations and the effect of these on the overall corporate social performance, results showed that the number of corporate social innovations had a positive influence on the number of news articles that were posted concerning the focal firm and social responsibility. From this we can conclude that stimulating corporate social innovation via an OUIC has a positive effect on the attention that the firm gets concerning CSR related news and has therefore a positive effect on the corporate social performance. This is in line with research that claims that involving stakeholders can positively influence the corporate social performance of the firm (Turgsa et al., 2013; McWilliams and Siegel, 2001; Russo and Perrini, 2010).
Now the main results have been evaluated, the research question can be answered: To what extent can a firm make use of online user innovation communities (OUICs) to create corporate social performance? The results of this research show that making use of an OUIC to develop CSR, can be evaluated as highly relevant. The corporate social innovations coming forward from these OUICs have a positive effect on the CSR based media attention and therefore influence the corporate social performance in a positive way. In order to make use of OUICs when developing CSR, the interactions within the platform should be managed in the right way to create a high corporate social innovation rate and to influence the corporate social performance in a positive way.
6.2 Theoretical implications
The results of this research contain some important theoretical implications. First, the research provides evidence for the development of CSR via a specific type of open innovation. Specifically, the results reveal the importance of developing corporate social innovation through OUICs where the customer becomes the main source for obtaining new knowledge. Referring to the resource based view, improvements in the corporate social performance are thought to be of sufficient influence on the eventual firm performance. Nevertheless the increase in corporate social performance is obtained via external sources which is not in line with the resource based view but with the dynamic capability view. The active position in the external environment can therefore, next to a better corporate social performance, also influence the development of dynamic capabilities. In short, these outcomes give insights into the up till now unexplored involvement of OUICs when building CSR performance and show that open innovation should be taken serious as a construct within the CSR literature.
