The Electronic Journal for Virtual Organizations and Networks Volume 8, March 2007

CURRENT PRACTICE AND STRUCTURAL PATTERNS IN VIRTUAL ORGANIZATIONS – A QUALITATIVE ANALYSIS OF 30 CASES Marita Haas*, Sabine Koeszegi* & Matthias Noester** *UNIVERSITY OF VIENNA School of Business, Economics, and Statistics Bruenner Strasse 72; A-1210 Vienna, Austria **SFC Sylvie Feindt Consulting 4, rue de la Presse; B-1000 Brussels, Belgium

{marita.haas, sabine.koeszegi}@univie.ac.at, [email protected]

ABSTRACT Virtual Organizations (VO) are seen as possible solutions for Small and Medium Enterprises (SME) to overcome disadvantages attributed to their size. Although several examples are well described in the literature, the VO domain still needs clarification in terms of VO types and characteristics. This paper aims at exploring different designs of VOs by investigating structural patterns of 30 VOs. The authors develop an analysis framework, the Virtual Profile, which allows profiling each case along structural dimensions. The analysis shows that, in practice, most of the VOs are based on prior relationships of collaborators with a strong focus on core competencies that intend to cooperate on a long-term basis. Apart from these common characteristics, structural differences between two types of VOs are identified. The resulting typology of two main types – “emerged” (internally driven) and “initiated” (externally driven) VOs - contributes to an explanation of VO types in literature as well as in practice.

Keywords: Virtual Organizations, Virtual Profile, Exploratory Research

Copyright: © 2007 eJOV. This is an open-access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by-nd/2.5/), which permits unrestricted use and distribution, in any medium, provided the original author and source are credited, and prohibits derivatives.

Current practice and structural patterns in VOs

1

INTRODUCTION AND THEORETICAL BACKGROUND

The ongoing globalisation, fast changing technologies and a turbulent business environment characterize today’s economy. These changes affect today’s enterprises’ production processes as well as their internal structures. Especially for small and medium-sized enterprises (SMEs), this is not without problems. The main disadvantages of SMEs, i.e. the lack of sophisticated know-how and the financial background to respond to new market requirements, are attributed to their size. While large enterprises can easily customize their products and services according to their clients’ needs, SMEs are likely to be more vulnerable in a changing business environment. A possible solution to this dilemma is to enhance cooperation between SMEs. Cooperative structures and joint production, selling or marketing processes are able to strengthen the competitive position of SMEs by giving them virtual size. This means that networks of smaller companies are able to act like larger ones. Virtual Organizations (VOs) have been proposed as an adequate cooperation business model by offering a maximum of flexibility for the individual companies. Byrne et al. (1993, p. 36) define a VO as “(…) a temporary network of independent companies – suppliers, customers, and even rivals – linked by information technology to share skills, costs, and access to one another’s markets.“ Sieber and Griese (1998, p. 213) similarly describe VOs as “(…) an appropriate kind of cooperative organization to explore business opportunities that one enterprise itself would not be able to work out.” Bultje and van Wijk (1998, p.9) also point out that in VOs „(…) companies quickly unite to exploit a specific opportunity and will disperse afterwards.” The organizational concept of a VO does not follow a generally accepted definition but is represented by its main characteristics: (a) Collaborating companies focus on their individual core competencies, but (b) they act with one face to the customer. Ideally, partners contribute their core competencies to the chain of value creation and thereby form a best-of-everything network (Balint & Kourouklis, 1998). (c) The partners remain legally and economically independent (Bultje & van Wijk, 1998; Mertens et al. 1998). (d) To be able to react quickly, the companies refrain from hierarchical integration. (e) Instead, the cooperation between partners is based on self-organization and mutual trust (Koeszegi, 2001). (f) Sophisticated Information and Communication Technology (ICT) support facilitates coordination among partners and reduces communication costs (Sieber, 1998; Wuethrich et al., 1997). (g) Finally, VOs are originally seen as temporary networks. Only those companies cooperate that are able to fulfil a certain business opportunity. The next business opportunity may be different and therefore, there might be a need for different competencies and companies (see also the first ideas of VOs of Davidow & Malone, 1992). Although the benefits of VOs and networks - like cost sharing, access to new markets and increased flexibility - are widely known and discussed (Davidow & Malone, 1993; Sieber & Griese, 1998; Mertens et al., 1998), the realization of the concept might be challenging in

eJOV – The Electronic Journal for Virtual Organizations and Networks, Volume 8, March 2007.

84

Marita Haas, Sabine Koeszegi, Matthias Noester

connection with SMEs. One main problematic issue is the risk associated with sharing knowhow with potential competitors. The competitive advantage is often based on unique and tacit production procedures (Sherer, 2003) and entrepreneurs are not willing to share them. Thus, trusted relationships among cooperating companies are not easily established. This in turn leads to the situation that the coordination of the partners’ activities within the VO is in reality organized more hierarchically than suggested by theory. Second, the process of forming and coordinating an enterprise that is able to react quickly to market opportunities not only demands flexibility but also predefined rules and roles for daily business and conflict resolution (Freitag & Winkler, 2000). Establishing these rules and cooperation agreements is difficult, especially for SMEs that usually dispose of little formalization and depend strongly on the owner of the business (Mintzberg, 1999). Scholars have presented Airbus1 or Smart Car2 as examples for VOs in the manufacturing sector (Wuethrich et al., 1997). Other often-cited examples are Rosenbluth International3, a VO in the transport and travel sector with subsidiaries all over the world (Mertens & Faisst, 1997) and Amazon4, an on-line retailer. Sieber and Griese (1998) and Bremer et al. (2001) highlight the example of VIRTEC, a cooperation platform of Brazilian SME – manufacturers as best-practice example. The Virtual Factory Nordwestschweiz Mittelland5 is a popular example for a vertical cooperation network (Wuethrich et al., 1997; Mueller-Stewens, 1997; Goeransson & Schuh, 1997). Recent research shows that, in practice, the VO-model varies according to the specific business situation of cooperation partners (Huber, 2005). Furthermore, VOs consisting of SME-partners like the Virtual Factory Nordwestschweiz - Mittelland show other structural patterns than cooperation between larger enterprises. Nevertheless, examples do not resemble each other: while Dell6 is a huge supplier network with partner companies distributed all over the world, the Virtual Factory Nordwestschweiz-Mittelland is a regional network of SMEs founded to explore joint innovation and production. Different from both seems VW Resende (Wuethrich et al., 1997), where a huge plant unites the competencies of eight suppliers from different countries in order to establish trucks and busses. Nevertheless, all three of them are mentioned as successful VOs. This discussion shows that a) there is no one definition for what a VO is and that b) various differently structured VOs are highlighted as being exemplary. What is more, previous research on VO examples has been restricted to single case studies, providing information about one specific application, but at the current state of knowledge there exists no comparative study of VOs within or between industries. The authors therefore take an

1

http://www.airbus.com, accessed in October 2006. http://www.smart.com, accessed in October 2006. 3 http://www.rosenbluth.com/home.html accessed in August 2004. 4 http://www.amazon.com, accessed in October 2006. 5 http://www.virtuelle-fabrik.ch, accessed in October 2006. 6 http://www.dell.com, accessed in October 2006. 2

eJOV – The Electronic Journal for Virtual Organizations and Networks, Volume 8, March 2007.

85

Current practice and structural patterns in VOs

exploratory approach and carry out an in-depth analysis of 30 VO cases reported in the literature by characterising and comparing the examples. By analyzing qualitative case study data, the structure of VOs is investigated. Dominant structural patterns are identified and the result reveals how VOs present themselves in practice. The remainder of this paper is organized as follows: in the next section, the authors present the cases and discuss the method and construct measurement. Section three presents the results of the analysis and finally, section four comprises discussion and conclusions.

2

METHOD AND MEASUREMENT

Based on the research approach to detect structural patterns of VOs, an exploratory metaanalysis was used to carry out the investigation. A combination of qualitative and quantitative methods assures both, the inductive interpretation process, but also the procession of data for quantitative analysis and hypotheses testing7. In the analysis, we followed the following procedure: 1. Based on an extensive literature review, the authors developed a “Virtual Profile” including the ten most often cited characteristics of a VO. 2. In a second step, two independent expert coders profiled 30 selected cases according to the established scheme.8 The sample was drawn from a list of more than 65 VOs, discussed in recent academic or industrial literature (cf. Wuethrich et al., 1997; Mertens & Faisst, 1997; Sieber & Griese 1998). Selection of cases was based on the purpose of the study to focus on SME VOs in the manufacturing industry. The selected sample represents a strong emphasis on the manufacturing sector (77 percent). Furthermore, 57 percent of the included cases contain mainly SMEs. The VOs are located in 10 different countries. Two thirds of the VOs consist of a maximum of 50 companies. 3. The information used for profiling was based on existing reports on the selected cases, but was completed with available information on the Internet, reports in literature and magazines and personal interviews with member companies. If necessary, the coders conducted additional telephone interviews with managers of the selected VOs in order to get missing data for profiling. Every single case was therefore drawn from multiple sources, consisting of currently available but also former documented sources. Each case was rated on each item of the profile on a five-point scale (see table 1). The coding results were compared and discussed to establish consistency between coders.

7 8

The model we used was a Generalization Model (Srnka & Koeszegi, 2004). A detailed list is applied in the Appendix of this paper.

eJOV – The Electronic Journal for Virtual Organizations and Networks, Volume 8, March 2007.

86

Marita Haas, Sabine Koeszegi, Matthias Noester

4. Qualitative comparative analysis method proposed by Ragin (1987) was used to identify structural patterns from the case study data. Following this approach the authors clustered cases in an iterative process according to different external (for example country of origin, industry) and internal (for example coordination processes, distribution of power) dimensions to compare Virtual Profiles and identify similarities or differences. 5. After the qualitative part, descriptive statistics, Exploratory Factor Analysis and Cluster Analysis were used to quantitatively test statistical significance of findings. The Virtual Profile (table 1) consisted of the ten items belonging to four different dimensions: (a) criteria concerning the Set-up environment of the VO, (b) its Purpose, (c) the Organizational framework and (d) the Time horizon. The selection of the criteria is based on the general characteristics and the definition of a VO, as well as criteria of its set-up process. VIRTUAL ORGANIZATION 5 a) Set-up Evolution (emerged - initiated environment General industry ties (strong - weak Ties between Partners (strong - weak b) Purpose Aim / Vision (well defined - non defined c) Organizational Focus on Core Competencies (high - low framework Scope (international - loca Standardization of Governance Structure (low - high Dispersion of Power (equal - foca Use of ICT (extensive - limited d) Time horizon Duration (temporary - long-term

4

3

2

Table 1. Virtual Profile

(a) This dimension describes the cooperative environment where the VO evolved: the authors assume that, for example, patterns of cooperation in an industry may affect the formation as well as the success of VOs (Gulati et al., 2000). The first item is Evolution (emerged vs. initiated). It differentiates between VOs that are purely based on a business opportunity recognized by one of the partners from those that are established on the basis of external funding or support. Second, General Industry Ties (strong vs. weak) describe the general cohesion in the industry, reflecting a certain established attitude towards cooperation, whereas the third item, Ties between Partners (strong vs. weak) measures whether the VO partners had prior relationships to each other before the VO was formed. (b) The Purpose of a VO is identified with the item Aim / Vision (well defined vs. nondefined) and it quantifies whether the network follows an explicitly and jointly defined vision. Like in any other organization, the strategic direction of a VO serves as a guideline for joint business (Huber et al., 2005) (c) The criterion Organizational framework comprises five items classifying the core structure of a VO. The authors differentiate between the following factors: Focus on Core Competencies (high vs. low) shows the concentration of every partner on his key

eJOV – The Electronic Journal for Virtual Organizations and Networks, Volume 8, March 2007.

87

1

Current practice and structural patterns in VOs

products/services that are combined with each other’s key products/services in the cooperation network. Several authors argue that the focus on and the dynamic combination of core competencies enable VOs to meet the customers’ requirements efficiently by reacting in a fast and flexible way (Sieber, 1998; Bauer & Koeszegi, 2003). Scope (international vs. local) aims at classifying the geographical extension of a VO. Interdependencies between collaborating partners require coordination (Laux & Liermann, 1997). VOs, however, are supposed to act with a minimum of rules or contracts and are coordinated on the basis of mutual trust, which is seen as an instrument to reduce opportunistic behaviour (Granovetter, 1985; Koeszegi, 2001). Therefore, the authors have added the item Standardization of Governance Structure (low vs. high), which measures the degree of formalization of coordination. Additionally, the authors differentiate the degree of governance structure from the Dispersion of Power (equal vs. focal) to analyse the power distribution and dependence between partners. Theoretically, VOs should be characterized by value-orientated relationships where all partners have equal rights and no partner dominates (Wuethrich et al., 1997). The fifth item is Use of ICT (extensive vs. limited) since information and communication technology is seen as a driving force for VOs facilitating communication and resource sharing (cf. Davidow & Malone, 1992). (d) The time aspect is represented by the fourth dimension Duration (temporary vs. long-term), reflecting the life span of VOs, which are often seen as temporary phenomena dissolving at the end of a fulfilled project. For further analysis, the Virtual Profiles are also graphically displayed: a case’s profile more on the left side of shows a higher degree of virtualisation than a profile more on the right side (see table 1).

3

FINDINGS

The findings section contains a descriptive analysis of data, a section on Virtual Profiles and structural patterns, as well as a section on the identification of critical success factors for the establishment of VOs.

3.1 Descriptive Analysis After individual coding of the cases by two independent coders, the case-profiles have been compared and aligned if necessary. The final profiles of 30 VOs deliver, a diverse picture (see table 2).

eJOV – The Electronic Journal for Virtual Organizations and Networks, Volume 8, March 2007.

88

Marita Haas, Sabine Koeszegi, Matthias Noester

Virtual Profiles Evolution General Industry Ties Ties between Partners Aim / Vision Focus on Core Competencies Standardization of GS Dispersion of Power Scope Use of ICT Duration

(emerged - initiated) (strong – weak) (strong - weak) (well defined - not defined) (high - low) (low - high) (equal - focal) (international - local) (extensive - limited) (temporary - long-term)

15 19 11 15 24 10 18 14 17 3

6 4 8 6 3 5 3 4 7 1

9 7 11 9 3 15 9 12 6 26

Table 2. Frequencies

Table 2 exhibits the final profiling of the 30 cases and describes how many of the examples disposed of certain structural dimensions. Originally, the authors used a 5-point scale for each dimension but for further analyses the extreme values of each side of the continuum were merged so that only 3 groups remained. Set-up environment There are two possible paths how VOs develop in practice: (1) companies identify a business opportunity and arrange a network based on partners they know from previous business (=emerged VOs), and (2) the network is established through the active support of a third party (=initiated VOs) (compare also to Hannus, 2005). From the analysed cases, half of the VOs emerged, while the others were supported or funded by an external institution (industry association, university, consultant, regional authority). 63 percent of the analysed VOs have been established in an industry where companies tend to form clusters regularly or where at least the concept of cooperation is known in a wider context (could be partnerships, joint projects or outsourcing). This indicates that some industries facilitate cooperation more than others. In eleven cases (see table 2), the organizations have participated in a cluster or network before and had therefore established personal or business-related ties (see for example the Virtual Team of Rauser Entertainment or the former suppliers in the VW Resende network (both in Wuethrich et al., 1997). In the interviews, several network managers asserted that they knew their collaborators personally from working groups or even university before the respective VO was established. This helped them to create good business relationships and mutual trust. Furthermore, all interviewed network managers confirmed that prior relationships facilitated successful cooperation. Purpose The criterion vision/aim of the VO seems to have increased in importance in the last decade. While the VOs founded during the 1990s were basically driven by the exploitation of a concrete business opportunity (like the establishment of the milking system Prolion (Bultje &

eJOV – The Electronic Journal for Virtual Organizations and Networks, Volume 8, March 2007.

89

Current practice and structural patterns in VOs

van Wijk, 1998) or the helicopter design team Sikorski (Goranson, 1999)), today’s collaboration concepts are almost always operating on basis of a clearly defined strategy for the VO as an organizational entity itself. This holds especially for the group of initiated VOs, where external experts support the network formation process, like the Virtual Factory Nordwestschweiz-Mittelland or Swiss Microtech. The explicit formulation of a joint strategy is a fundamental step in the set-up process of externally funded projects (see for example EU framework program no. 6)9. The interviews revealed that VOs’ strategies are, especially in externally initiated VOs, based on a combination of its individual member’s strategies. The vision serves as the basis for joint marketing activities and includes the establishment of a brand name. Only after successful network formation, the members enter into the intelligence and production phase in which joint business opportunities are developed and realized. Organizational framework A strong focus on the core competences was identified in four fifth of the cases (see table 2). Twelve VOs of the sample recruit partners locally, four regionally, and in fourteen VOs the partners are from different countries. However, from the interviews we know that all VOs want to operate on international markets. The Standardization of the Governance Structure is one of the most diverse factors in the sample. While half of the VOs explicitly established rules and defined roles, one third disposed only of very few formalized governance mechanisms. VOs can chose from a number of formalized and informal coordination mechanisms. Formalized mechanisms define the frame of the cooperation and include contracts, cooperation agreements, and joint ICT infrastructures. Informal mechanisms, in contrast, are based on self-organization and require personal and direct contact of partners’ activities. In interviews, the network managers claimed that it was essential for network success to find the right balance of formal and informal coordination. If general rules are in place and companies trust that they are adhered to, the set-up of concrete business projects can be organized informally and coordination becomes easier. All of the seven interviewed VOs installed a network manager, who is responsible for the coordination of the partners. The manager organizes, for instance, meetings on a regular basis, supports partners, and/or distributes information. Additionally, six of the seven VOs have installed the role of a network broker, who looks actively for business cases, markets the VO, and acts as a boundary person. The scope of activities and invested time, however, is very diverse. In some VOs of the sample, the members themselves or the network manager act as brokers and only one case additionally employed a full-time network broker.

9

Details about the 6th framework program: http://europa.eu.int/comm/research/fp6/index_en.html.

eJOV – The Electronic Journal for Virtual Organizations and Networks, Volume 8, March 2007.

90

Marita Haas, Sabine Koeszegi, Matthias Noester

As far as power relationships are concerned, 60 percent of the sample shows an equal power distribution among the partners, 30 percent of the VOs are controlled by one dominating company. In the remaining 10 percent, there is more than one company having a strong influence in the network (see table 2). About one half of the VOs of the sample use basic ICT and only seven cases use Intranet or more sophisticated collaboration tools. 20 percent of the VOs coordinate their work solely using traditional communication media like telephone, fax, and sometimes email. In the interviews, several network managers confirmed the lack of interest in ICT tools: for example only one interview partner uses a sophisticated Internet platform and members are required to check it everyday. In another case, the manager contacts the members by phone or email when information on the Intranet is updated. In yet another case, the use of ICT was literally abandoned after a “trial period”. Time horizon While in earlier research VOs were purely regarded as a temporary phenomenon, recently several authors acknowledge the co-existence of both temporary and more permanent forms (cf. Aken et al., 1998 or Wuethrich et al. 1997). In the sample, almost all (26 out of 30) VOs were established with a long-term perspective (see table 2). In summary, the descriptive analysis shows that some of the characteristics in the profile seem to be very similar for all VOs: almos. all examples have a high focus on corecompetences, the use of ICT is not excessively high, almost all networks are based on a long-term perspective, and the personal ties between the partners in the networks seem to be important. On the other hand, there are some factors, in which the VOs seem to be very diverse. Especially in the items Evolution (emerged vs. initiated), Scope (international vs. local), and Dispersion of Power (equal vs. focal) we find different values in the sample. Furthermore, the formalization of government structures, i.e. the use of formalized and informal management tools and their role within networks, is apparently not the same in all cases. The authors therefore assume that there are at least two or even more types of VOs in the sample.

3.2 Virtual Profiles and structural patterns In order to identify structural patterns, an exploratory factor analysis and a cluster analysis with the 30 profiles was carried out. The results of the factor analysis are displayed in table 3:

eJOV – The Electronic Journal for Virtual Organizations and Networks, Volume 8, March 2007.

91

Current practice and structural patterns in VOs

Exploratory factor analysis Main components analysis with Varimax-rotation and Kaiser-normalization. KMO=546; Bartlett-test: p