IMM Manuscript 10-562RR
Understanding Network Picture Complexity:
An Empirical Analysis of Contextual Factors
Carla Ramos*
manchester IMP Research Group,
Manchester Business School - The University of Manchester
Booth Street West, Manchester M15 6PB, UK
Tel: +44 161 306 3528
and
Faculty of Economics, Universidade do Porto
Rua Dr. Roberto Frias, 4200-464 Porto, Portugal
[email protected]
Stephan C. Henneberg
manchester IMP Research Group,
Manchester Business School - The University of Manchester
Booth Street West, Manchester M15 6PB, UK
Tel: +44 161 306 3463
[email protected]
Peter Naudé
manchester IMP Research Group,
Manchester Business School
The University of Manchester
Booth Street West, Manchester M15 6PB, UK
Tel: + +44 161 275 7782
[email protected]
Industrial Marketing Management
Initial Submission November 2010
Revised Submission June 2011
Second Revised Submission November 2011
* Corresponding Author, Tel.: +44 7796124630, E-mail address: [email protected]
IMM Manuscript 10-562RR
Bio sketches
Carla Ramos is a Lecturer at Manchester Business School (MBS), University of Manchester, UK.
Previously, she was a post-doctoral research fellow in marketing jointly at MBS, UK, and at the Faculty
of Economics, Universidade do Porto, Portugal. She joined the mIMP Research Group after she finished
her PhD at the University of Bath. Her research interests are in the area of business-to-business
marketing, network pictures, and services, business and innovation networks.
Stephan C. Henneberg is Professor of Marketing and Strategy at Manchester Business School. He
obtained his Ph.D. in Marketing from the University of Cambridge, Judge Business School. His current
research interests are in the areas of strategic marketing, relational marketing, consumer behavior,
strategic competences, and social and political marketing.
Peter Naudé is Professor of Marketing at Manchester Business School, Manchester University, UK. He
gained his Ph.D. in Marketing from the University of Manchester. His research interests are in
quantitative modelling and B2B Marketing.
IMM Manuscript 10-562RR
Research Highlights
 There is a growing body of research on the relationship between managers’ perceptions




and how they interact with each other.
Past evidence shows how complexity of perceptions conditions firms’ actions. We
examine network picture complexity.
A model of network picture complexity is derived and used to analyze forty-seven
network pictures from two networks.
We show the relationship between complexity and specific individual managers and
organizational characteristics.
This paper allows better understanding of how contextual factors condition sense-making
in business networks.
IMM Manuscript 10-562RR
Understanding Network Picture Complexity:
An Empirical Analysis of Contextual Factors
Abstract
There has recently been increasing interest in the relationship between managers’ perceptions of
their surroundings and their interactions with other actors. This sense-making issue is linked to
the development of the concept of network pictures. Our paper investigates a hitherto neglected
aspect of network pictures: their complexity. In several bodies of literature complexity has been
found to affect firms’ action and performance. We theoretically derive a model of complexity,
which is then used to analyze forty-seven network pictures collected in seventeen companies
from two distinct networks. Complexity is assessed on a number of dimensions at the individual,
and organizational levels, and we show the relationship between complexity and an individual
manager’s characteristics (number of years in a company, as well as experience in internally or
externally oriented functions). We also provide evidence for a relationship between cognitive
complexity and the number of years a company has been established in business, and the
complexity of companies’ internal structures. In doing so, this article contributes to a better
understanding of the contextual factors that drive sense-making in business networks.
Keywords
Business Networks, Cognitive Complexity, Contextual Factors, Network Picture, Sensemaking
IMM Manuscript 10-562RR
Understanding Network Picture Complexity:
An Empirical Analysis of Contextual Factors
1 Introduction
Companies are embedded in complex networks of interconnected relationships, with most
research on industrial systems based on this dictum (Axelsson and Easton, 1992; Håkansson and
Snehota, 1989; Håkansson and Johanson, 1992; Knight, 2002). Within these relationships, it is
through networking that a company can access the resources that are required to develop and
sustain business activity (Pfeffer and Salancik, 1978). Networking is defined as all interactions
of an actor in a business network, comprising all activities concerning the management of the
existing relationships, the management of the position occupied in the surrounding network and
the strategies of how to network (Ford et al., 2002; Håkansson and Ford, 2002).
Such activities within the network are grounded on the underlying beliefs about the network by
relevant actors. Based on principles of organizational cognitive science, an organizational actor’s
actions can be understood as being conditioned by their perceptions of the world around them
(Weick, 1979). In business networks, this can be translated into how actors’ cognitive schemata,
i.e. views of their surroundings, condition their networking activities, either by restricting or
expanding their available options (Anderson et al., 1994; Ford et al., 2002; Gadde et al., 2003;
Ritter, 1999). Recently, there has been growing interest in this relationship between actors’
perceptions of their surroundings and their networking activities (Corsaro et al., 2011;
Henneberg et al., 2010; Ramos, 2008; Welch and Wilkinson, 2002). Specifically, the concept of
IMM Manuscript 10-562RR
network pictures was developed to capture an organizational actor’s beliefs about his/her
network (Colville and Pye, 2010; Mouzas et al., 2008; Öberg et al., 2007; Rohrmus and
Henneberg, 2006).
Our article contributes to understanding aspects of sense-making in business networks, especially
network pictures. Specifically, we focus on understanding one characteristic of network pictures
(or an actor’s cognitive schemata) that has been shown to influence networking activities (or
decision making) and consequently firms’ performance (see e.g., McNamara et al., 2002; Walsh,
1995): their underlying complexity. The aim is to develop a multi-dimensional construct of
complexity that can be used to assess the cognitive conditions of actors’ views regarding their
surrounding network. Additionally, the article seeks to explore the relationship between certain
contextual factors and network picture complexity. This is done by applying the developed
multi-dimensional complexity construct to the network pictures of forty-seven managers, in
seventeen companies from two distinct networks. Degrees of network picture complexity are
subsequently associated with different independent factors, for example managers’ personal
characteristics, and their company’s characteristics. This analysis allows us to contribute to the
literature on sense-making in business networks, which can then be used to further understand
networking activities in business relationships (Brown et al., 2008; Henneberg et al., 2010).
Therefore, although the core purpose of this research does not rely on investigating the
relationship that is believed to exist between network pictures and networking, future
investigation on this relationship can be anchored in this piece of research. The findings show
that complexity is higher for individuals with more experience in externally-oriented functions
than in internally-oriented ones, and that the more complex companies’ internal structures are,
IMM Manuscript 10-562RR
the more complex are the individuals’ network pictures. The data analysis also indicates that
there is an inverted-U shaped relationship between network picture complexity and the number
of years a company has been established in a business. Finally, and unexpectedly, we find a Ushaped relationship between network picture complexity and the number of years of an
individual’s working experience.
The paper is structured as follows: stressing the key role that actors’ perceptions of the network
play in bringing about networking activities, we begin by providing an overview of research on
strategizing in business networks. We then explore the meaning of complexity by uncovering its
twofold perspective: the objective and subjective definition of complexity. This is complemented
by a review of research specifically on complexity of business networks. From this we derive a
definition of network picture complexity as an analytical concept, and we develop a multidimensional construct to capture the degree of complexity in an actor’s network pictures. This
leads to an empirical analysis based on several propositions regarding the relationship between
network picture complexity and contextual factors. We outline our analyses and empirical
findings, and conclude by discussing contributions, implications, and suggest further avenues of
research.
2 Relationship between Network Pictures and Strategizing Activities in Business Networks
The network metaphor has been commonly used by researchers over the past two decades to
describe and analyze the characteristics and organization of complex industrial systems, such as
stability and change, and the creation of value (Håkansson and Johanson, 1992; Möller and
Svahn, 2006). Actors with heterogeneous resources, challenges, and goals, interact with other
IMM Manuscript 10-562RR
actors in the development of their business activity (Ford and Håkansson, 2006b; Pfeffer and
Salancik, 1978). Business actors are not able to use all the information available to them through
their scanning activities. Actors’ cognitive frameworks play a key role in helping them to select
what information to process and how to interpret it (Gioia and Chittipeddi, 1991; Weick, 1979).
Each actor therefore holds a subjective perception or network picture of his/her surroundings
based upon previous experiences and memories, as well as beliefs about the future (Ford and
Håkansson, 2006a; Mattsson, 1985, 1987). Distinct network pictures can be expected to co-exist
in the same company and throughout the network (Ford and Ramos, 2006; Henneberg et al.,
2006).
The processes of interaction between actors in business networks are guided by network pictures
(Ford et al., 2002; Ford and Håkansson, 2006a; Mattsson, 1987). This relationship between
network pictures and networking within business networks is associated with basic principles
from information processing and strategic management theories: actors’ actions and reactions are
guided by the output of their interpretations of the information they choose to select (Daft and
Weick, 1984; Gioia and Chittipeddi, 1991). As such, we follow Rohrmus and Henneberg (2006)
and understand network pictures in line with the concept of cognitive cycles defined by Neisser
(1967). There is therefore a recognized need to better understand individuals’ network pictures,
so that in the future it can be possible to understand the relationship between actors’ sensemaking, in this case their network pictures and their networking activities (Ford et al., 2002).
Before this interdependence between network pictures and networking is discussed in further
detail (more specifically how the complexity of actors’ cognitive structures affects actors’
decisions and performance), the multi-faceted nature of network pictures is discussed, as several
IMM Manuscript 10-562RR
complementary concepts can be found in the literature (Ford et al., 2002; Ford and Ramos, 2006;
Henneberg et al., 2006).
2.1 Network Pictures
The concept of network pictures was first introduced in the business network literature by Ford
et al. (2002) and since then, it has been defined or interpreted in different ways (Geiger and
Finch, 2010; Ramos and Ford, 2011). Rohrmus and Henneberg (2006) link it to Neisser’s
theories (1967, 1976) and anchor it in cognitive psychology. Network pictures can be understood
as an organizational actor’s subjectively perceived network (i.e. actors’ picturing of the network;
Henneberg et al., 2010): managers make choices about what they believe is important in
characterizing the network and its actors, activities, and resources. On a different level, network
pictures can be understood as a research tool: on one hand they reflect researchers’ efforts to
make sense of managers’ sense-making of the network (i.e. researchers’ picturing of actors’
network pictures or what Geiger and Finch (2010) define as ‘mentalist network pictures’) and on
the other hand, they are associated with researchers’ needs to have an independent and
objectified perspective on business network characteristics (i.e. a researchers’ own network
picturing which corresponds to Geiger and Finch’s notion of ‘representationalist network
pictures’). These two forms of network picturing are contrasting in the sense that, whilst the
former refers to researchers’ attempt to understand the cognitive characteristics of network
pictures of individual managers, the latter relates to the capturing of the objectified
characteristics of the network (Henneberg et al., 2010). The different levels of network
pictures/picturing described provide a kaleidoscopic view of networks, capturing different
IMM Manuscript 10-562RR
aspects of the overall constructs, and relating to each other in what we call an ‘epistemological
dialogue’. This dialogue is represented in figure 1.
_________________
Insert Figure 1 about here
_________________
Research always involves ‘picturing’ of some nature; there cannot be empirical research on
actors’ network pictures without the researcher’s exercise of picturing in order to capture those
pictures. Most studies conducted on network pictures are related to a particular level of network
picturing. Henneberg et al. (2006), Ford and Håkansson (2006b), and Ramos (2008) have all
developed research tools that can be used by researchers to grasp business actors’ understanding
of their surroundings, and other researchers have used these tools to look into specific business
network-related phenomena (Abrahamsen et al., 2009; Henneberg et al., 2009; Holmen et al.,
2008; Kragh and Andersen, 2009; Leek and Mason, 2008). On the other hand, Ford and
Redwood (2005) explicitly employed the third level of network pictures/picturing: they use their
own picturing of a single business and surrounding network over a period of one century to grasp
the dynamics of that business as well as the development of the overall network. However,
research that develops some kind of representation of a business network also relates to this third
level of picturing, although in an implicit way (see for example Salmi, 1999; Baraldi and
Strömsten, 2006).
For the purpose of this article we employ the concept of network pictures in a twofold way: on
one hand, we focus on ‘mentalist network pictures’, i.e. we use it as a research tool to conduct
IMM Manuscript 10-562RR
the empirical study and to capture the idiosyncratic network pictures of individual managers; on
the other hand, throughout most of the paper we talk about network pictures as actor’s
subjectively perceived network, i.e. actors’ picturing of the network. The latter provides a
shortcut for the reader (i.e. instead of talking about the ‘complexity of actors’ subjective views of
the network’, we simplify and talk about ‘network picture’s complexity’).
2.2 Network Pictures and Networking Activities
Researchers are interested in how different managers perceive their surrounding network, and
how the perceptual differences affect their actions. While these issues have recently been
emphasized by several researchers from the business network area (Henneberg et al., 2010;
Welch and Wilkinson, 2002), to date, there exists only limited empirical research (Corsaro et al.,
2011). Other areas of management research have also identified similar issues as important,
namely the cognitive strategic group literature in the area of strategic management (Walsh et al.,
1988; Jackson and Dutton, 1988). Moreover, there is clear evidence of the importance of
cognitive models for strategic decision-making and actions in this other area of management
(Brown et al., 2008), evidence that can be transferred to the relationship between network
pictures and strategic networking. For example, the way managers perceive and label issues (i.e.
as threats or opportunities, and as controllable or not controllable) condition their choices for
strategic change (Thomas et al., 1993).
Research in the cognitive strategic group literature in the area of strategic management shows the
effect that complexity of actors’ knowledge structures has on action and performance
(McNamara et al., 2002; Miller, 1993; Tetlock, 1984; Walsh, 1995). In summary, this research
IMM Manuscript 10-562RR
theme argues that complexity of actors’ knowledge structures affects the way they interpret the
surrounding environment, and thereby conditions their strategic responses (Porac et al., 1989;
Porac and Thomas, 1990).
There is also evidence of the impact of cognitive complexity on managers’ decision-making and
analytical skills (Streufert and Swezey, 1986; Streufert and Nogami, 1989). Decision-makers’
actions and firms’ performance are thus affected by the degree of cognitive complexity. Several
authors argue that the cognitive complexity of top-managers’ mental models needs to increase in
order to ensure the survival of organizations operating in complex networks (Bartunek et al.,
1983; Hooijberg, 1996; Miller, 1993; Weick, 1995; Yiu and Saner, 2000). This idea draws on the
notion of variety and on the principle that a more differentiated knowledge structure (i.e. with
more categories and thus more complex sense-making schemata) leads to optimize information
processing in ill-structured situations and thus to better performance (Ashby, 1956). However,
Walsh (1988) found that groups that based their decisions on a small number of belief categories
performed better than other groups that draw on more differentiated belief structures. This hints
at the existence of a critical point where cognitive complexity begins to have a negative impact
on performance: overly complex frameworks may act as an obstacle to managerial performance
in certain circumstances (Lurigio and Carroll, 1985; McNamara et al., 2002).
Understanding cognitive complexity is therefore pivotal in order to grasp the relationship
between actors’ sense-making, networking, and performance. Given that this paper focuses on
understanding the complexity of network pictures, in the following section we provide a
IMM Manuscript 10-562RR
discussion of the meaning of complexity, specifically focusing on cognitive complexity and
complexity in business networks.
3 Complexity of Cognitive Models
3.1 Perspectives and Definitions of Complexity
Reviewing the extant literature, it becomes obvious that there is no common definition of
complexity. It is possible to find multiple definitions within the same body of research, including
management-related studies. We identify two levels of analysis for complexity, namely objective
and subjective, each of which is explored below.
Objective complexity: complexity (and open systems) theory
Developed primarily in the field of natural sciences in the late eighties, complexity theory was
adopted by social sciences and specifically by economics, and then management (Byrne, 1998;
Holland, 1987). According to this theory, organizations interact within a complex system,
influenced by the contingencies of their surroundings. This results in emergent patterns at the
system or network level (Maguire and McKelvey, 1999). Consequently, understanding industries
as complex systems became an important means to enhance understanding of managerial science
(Levy, 1994; Maguire and McKelvey, 1999). The focus of complexity theory is on the whole
system, and on the dynamic relationships between the agents that constitute the system; there is
also a focus on the uncertainty of a system’s initial conditions, which can result in unexpected
emergent behaviors (Lissack, 1999; Mason, 2008; Thompson, 1967).
IMM Manuscript 10-562RR
Complex systems are characterized by on-going re-organization of multiple constituent parts into
new emergent structure, reflecting their adaptive orientation (Mason, 2008; Morrison, 2002). The
complexity of a system and the number of possible outcomes or emergent properties and
behaviors increase with the addition of new constituents (Kauffman, 1992; Mason, 2008).
However, a system’s emergent behavior and complexity do not derive solely from the
relationships between a system’s entities or from their ‘essence’. For instance, Stacey (1996) has
identified three parameters that condition complex adaptive systems: the rate of information flow
(and feedback) in the system, the richness and connectivity between entities in the system, and
the extent of diversity within and between the entities’ schemata.
In terms of structural features, complexity is dependent on the organization’s size and number of
sub-systems such as R&D, manufacturing and sales (Lawrence and Lorsch, 1967). It also
depends on the nature of the relationships that exist between organizational units (Thompson,
1967). Some researchers identify the complexity of an organization with the number of
organizational parts or structural components (e.g. occupational specialties or specialized
subunits) into which organizations and their employees can be categorized (Jablin, 1987). On the
other hand, in terms of organizational behavior, complexity can be seen as “intricate aggregate
patterns emerging from the interaction of the constituent parts of an organization that
themselves followed relatively simple behavioral rules” (Fioretti and Visser, 2004, p. 12). These
two objective approaches to complexity have some common grounds: both make reference to the
number of parts (i.e. numerosity), and the linkages between them (i.e. connectedness) (Eden and
Ackermann, 1992; Fioretti and Visser, 2004).
IMM Manuscript 10-562RR
Subjective complexity: a cognitive interpretation of complexity
A different notion of complexity is that of ‘second-order’ complexity (Tsoukas and Hatch,
2001). This refers to complexity that is not an intrinsic or objective property of a system, instead
being dependent on the modes of thinking that individuals use to make conjectures about it
(Fioretti and Visser, 2004). Complexity in this case is therefore observer-dependent (Casti,
1994). Drawing on the recognized lack of consensus on when a system should be considered as
complex (Casti, 1994; Waddington, 1977), Tsoukas and Hatch (2001) claim that “it is better to
explore complex thought processes (second-order complexity) in relation to an assumed
objective world (first-order complexity)” (p. 981). Thus, authors on ‘second-order’ complexity
undertake a narrative perspective of reality, according to which the vocabulary used by the
observer reflects the extent of complexity that he/she believes exists in a system, therefore not
reflecting an objective property of that system. A system’s complexity is thus dependent on the
complexity of the language or representation that is used by the observer to describe it (Brown et
al., 2008; Casti, 1994; Rescher, 1998; Simon, 1999), being dependent on the observer’s
perception. The narrative approach to complexity is congruent with the social constructionist
approach undertaken by Weick (1979) to define organizations. Subjective complexity itself is
mediated by the complexity of the cognitive structure of individuals’ cognitive systems; this
notion evolves from Kelly’s (1955) theory of personality, and results in differences in the way
individuals perceive their social environment (Carraher and Buckley, 1996).
From the different perspectives on complexity described, the subjective or ‘second-order’
complexity is the one that is closer to our aim of developing an understanding of network picture
complexity. Nevertheless, objective complexity definitions can also inform the development of a
IMM Manuscript 10-562RR
model in that they provide observers with a vocabulary that they may choose to use in their
description and interpretation of complexity. With this in mind, the next section looks into
definitions of objective complexity specifically in the business-to-business network area of
literature.
3.2 Complexity in Business Networks – the Industrial Network Approach
There is an extensive and wide-ranging body of research on business networks. Literature can be
found on strategic nets, i.e. networks which are intentionally created and which can be managed
(Jarillo, 1988; Gulati et al., 2000), organizational networks (Nohria and Eccles, 1992), or interfirm social networks (Granovetter, 1973; Scott, 2000; Uzzi, 1996). Given that the aim of this
paper is to understand a specific aspect of network pictures (i.e. their complexity), and because
this construct has been theoretically (and empirically) outlined within concepts linked to the IMP
Group (Industrial Marketing and Purchasing Group) who draw on the industrial network
approach (Ford et al., 2002; Håkansson et al., 2009; Ramos and Ford, 2011), we ground our
research in this specific approach to business networks.
The notion of complexity is closely associated with the industrial network approach (Axelsson
and Easton, 1992; Håkansson and Snehota, 1989). References to a business network’s
complexity are common in this body of literature, e.g. “we can regard the global industrial
system as one giant and extremely complex network since there exist always some path of
relationships that connect any two firms” (Mattsson 1988, p. 313). Networks of companies are
perceived by Ritter et al. (2004) as complex adaptive systems, where companies take part in a
self-organizing emergent process that results from multiple micro-interactions between them
IMM Manuscript 10-562RR
(Easton et al., 1997; Wilkinson and Young, 2002). References such as this are associated with
the logico-scientific or objective definition of complexity (i.e. first-order complexity).
However, the notion of subjective complexity (i.e. ‘second-order’ complexity) can also be found
in the industrial network approach body of literature. Lowe et al. (2008) draw on a narrative
approach to highlight how the adoption of a new language amongst business researchers, one
that draws on a discourse analysis of practitioners’ speech, would result in a new and probably
better understanding of the complexities of business networks. This is in line with the idea that
‘second-order’ complexity and a narrative approach are essential for the definition of system
complexity (Brown et al., 2008; Tsoukas and Hatch, 2001). There have been several calls for
further research on subjective complexity and sense-making in business systems, so far largely
unheeded (Ford and Håkansson, 2006b; Mattsson, 2005). The notion of subjective complexity of
business networks is closely associated with the idea of complexity of network pictures as
developed in this paper.
As in other fields of research, there is no agreement on how to define business network
complexity. For example, Halinen and Tornroos (2005) posit that complexity is a multidimensional construct in terms of structure and embeddedness. Networks are structurally
complex because they involve multiple actors and multiple links of diverse nature between them,
such as weak and strong, indirect and direct, and loosely or tightly coupled ones. In terms of
embeddedness, complexity derives from the difficulty in defining a business network’s
boundaries, as well as the informal nature of some agreements that link actors, and it also results
from networks being embedded in different spatial, social, political, and technological structures.
IMM Manuscript 10-562RR
Therefore, business networks are unique and context specific. Complexity is also seen as a result
of networks dynamics, being strongly dependent on actors’ positioning (Håkansson, 1982). This
is in line with Easton (1995) who argues that it is connectedness and dynamism that confer
complexity on networks. Baraldi and Nadin (2006) posit that complexity derives from the several
entwined and concurrent types of exchanges (i.e. social and economic) between actors. Each
actor’s sets of bonds, resource ties and activity links are embedded in a web of actors, resource
constellations and activity patterns, the result of their involvement in multiple business
relationships with other actors. Complexity is also conditioned by the diversity of roles played by
each actor. A relationship’s content, substance and strength, and underlying (intensive) flows of
information, also account for business network complexity. Baraldi and Nadin (2006) posit that
increasing the number of involved relationships in an already complex network leads to greater
complexity. On the other hand, Johnston et al.’s (2006) definition of complexity focuses mainly
on the communication flows between companies, the ‘glue’ that connects relationships (Mohr
and Nevin, 1990; Samli and Bahn, 1992). Complexity is perceived by these authors as being
dependent upon the number of network interactions and the number and diversity of levels of
communication through which actors connect with each other. Another approach to complexity
is that of Johnston et al. (1999) who relate complexity to the different nature of the actors that
interact in the network (e.g. firms, governmental entities and multinationals/multi-firms).
The different conceptualizations of complexity in business networks discussed in this section
have common grounds, and to a certain extent can be said to be complementary to each other.
The differences between them are mostly a result of the emphasis that the authors place on
IMM Manuscript 10-562RR
different aspects of complexity. In the following section, we synthesize the existing literature and
identify and operationalize a set of dimensions that define network picture complexity.
4
Defining and Capturing Network Picture Complexity
4.1 Definition of network picture complexity
A business network is not objectively given; it is subjective and varying, dependent on what
actors believe it to be, or dependent on their network pictures (Ford et al., 2002; Håkansson and
Snehota, 1995). A ‘second-order’ approach to complexity is therefore appropriate for analysing
network picture complexity (Ford and Håkansson, 2006b; Lowe et al., 2008; Mattsson, 2005).
By analyzing the representations (the pictorial/spatial understanding of the network) and the
language (the textual description of the network) that individuals use to describe their
surroundings, researchers can evaluate the degree of complexity of an actor’s network picture.
However, objective complexity also plays a role in the definition of a model of network picture
complexity: it provides actors (i.e. the observers) with an idea of what elements can be used for
defining a system’s complexity. We therefore use a tool for picturing of actors’ network pictures,
with a ‘double-hermeneutical’ logic being involved in this process, which is based on capturing
and interpreting a respondent’s interpretation of a social phenomenon.
Drawing on the definition of subjective complexity and on the theoretically identified
dimensions of network picture complexity introduced above, we put forward the following
definition of network picture complexity:
IMM Manuscript 10-562RR
Network picture complexity is defined by the language and representational dimensions that
actors use to describe their surrounding business network, specifically the number and nature of
actors, number and nature of relationships, and dynamism and flexibility. These dimensions are
dependent on actors’ sense-making and interpretation of that network.
In line with the literature on complexity, we operationalize network picture complexity as a
multi-dimensional construct. Three dimensions can be delineated: the number and nature of
actors, the number and nature of relationships, and finally dynamism and flexibility. Each of
these dimensions can be expanded into several sub-dimensions (see table 1). These dimensions
(and corresponding sub-dimensions) of network picture complexity are further operationalized in
section 6; furthermore, a more detailed overview is provided in Appendix 1.
----------------------------------Insert Table 1 about here
-----------------------------------
4.2 Number and nature of actors
According to complexity theory, adding new constituents to a system increases its complexity
(Eden and Ackermann, 1992; Kauffman, 1992; Mason, 2008). Complexity is said to be
dependent on the system’s size and number of sub-systems (Lawrence and Lorsch, 1967) or
structural components (Jablin, 1987). This finding is mirrored in the literature on business
networks (Halinen and Tornroos, 2005; Johnston et al., 2006). We suggest that the greater the
number of actors identified by an individual when representing and describing the surrounding
business network, the more complex is his/her network picture.
IMM Manuscript 10-562RR
Complexity is also associated with the nature of the involved parties: the extent of diversity
within and between the entities’ schemata is a parameter that conditions complex adaptive
systems (Stacey, 1996). This idea is reinforced in the business network literature, with business
network complexity being linked with different qualities of actors that interact in the network
(Johnston et al., 1999). We therefore expect that complexity also increases when an individual
identifies more diverse types of actors.
4.3 Number and nature of relationships
One basic principle underlying complexity theory is that if the number of potential connections
and interactions increases, so too does the complexity of the system (Eden and Ackermann,
1992; Kauffman, 1992; Mason, 2008). The number of relationships therefore conditions the
complexity of a system. This idea can also be traced in the business network literature (Halinen
and Tornroos, 2005; Johnston et al., 2006). Increasing the number of relationships involved in a
business network leads to greater complexity (Baraldi and Nadin, 2006; Samli and Bahn, 1992).
We therefore suggest that the complexity of an individual’s network picture increases with the
number of relationships included in his/her representation and description of the surrounding
business network.
Moreover, the nature of the relationships included also affects complexity. Complexity depends
on the nature of relationships that exist between organizational units (Thompson, 1967), being
associated with the adaptive behavior that results from multiple parts of a complex system
(Fioretti and Visser, 2004; Stacey, 1996). One of the reasons why business networks are said to
IMM Manuscript 10-562RR
be complex is that they are embedded in multidisciplinary structures, which makes them unique
and context specific (Easton, 1995; Halinen and Tornroos, 2005). Relationships’ content,
substance and strength, and underlying flows of information, all influence complexity (Baraldi
and Nadin, 2006). Business networks complexity is also said to be dependent upon the diversity
of levels of communication through which they can connect with each other (Johnston et al.,
2006). We therefore conceive that a network picture is more or less complex depending on the
variety of the nature of relationships that are included in that picture, i.e. greater variety leads to
greater complexity (e.g. perceiving not only weak ties or strong ties, but both, confers greater
complexity).
4.4 Dynamism and flexibility
A system’s complexity is associated with its underlying dynamism and flexibility. In complex
systems, the focus is on the dynamic relationships between the agents that constitute the whole.
An on-going organization and re-organization of a system’s multiple constituents into new
emergent structures takes place in these complex systems, reflecting these systems’ adaptive
orientation (Mason, 2008). Unexpected emergent behaviors created by networking activity are
expected to affect a system’s initial conditions (Lissack, 1999; Mason, 2008; Thompson, 1967),
with the system exhibiting emergent new structural patterns (Maguire and McKelvey, 1999).
Business networks complexity is also associated with networks being flexible, meaning that
change is one of their intrinsic features; there is therefore a temporal dimension of complexity
(Halinen and Tornroos, 2005). Easton (1995) adds to this idea by claiming that complexity is a
result of the dynamism that characterizes networks. We suggest that the complexity of an
IMM Manuscript 10-562RR
individual’s network picture will vary according to the extent of changes he/she perceives and
identifies.
5 Propositions Regarding Contextual Factors of Network Picture Complexity
Previous research shows that the complexity of managers’ mental models is conditioned by
specific contextual factors such as individuals’ domain of expertise (Carraher and Buckley,
1996; Hershey et al., 1990; Lurigio and Carroll, 1985; Vandenberg and Scarpello, 1990).
Considering that individuals’ “interpretations and response choices are not made independent of
decision makers’ contextual perspectives” (Ford and Bacus, 1987, p. 373), we posit several
propositions on the relationship between network picture complexity (i.e. our construct capturing
mental models) and different factors relating to context, namely individual manager
characteristics, and company characteristics.
5.1 Network picture complexity and characteristics of managers
Variations in the experiences of individuals result in differences in the content of individual
knowledge structures (Fiske et al., 1983; Larkin et al., 1980; Wagner, 1987). Lurigio and Carroll
(1985) show that more experienced individuals use fewer but more detailed cognitive schemata
than others with less experience. When entering a new business or company, a manager can be
expected initially to be more committed to learning in unfamiliar surroundings, being more
sensitive to manifold phenomena. As the individual becomes more experienced and familiar with
the situation, he/she may be expected to become less sensitive to certain issues but at the same
time to be able to identify new phenomena, and to fit them into their understanding of the
surroundings (Starbuck and Milliken, 1988). Thus, whilst at the beginning of a learning
IMM Manuscript 10-562RR
experience, managers’ cognitive structures have more schema categories, they have fewer
information units within each of those categories, compared to later stages of that learning
experience (Lurigio and Carroll, 1985; Rentch et al., 1994).
According to Starbuck and Milliken (1988), with experience and success, managers generalize
more by capitalizing on their familiarity with several domains of expertise and on their access to
rich and filtered sources of information. There is a learning process associated with managers’
experience, and the reinforcement of perceptions results in managers developing strong views
with regard to what works, what does not work, and how things ought to work in business
networks (Levitt and March, 1988; Reger and Palmer, 1996).
We therefore posit that the relationship between network picture complexity and individual
experience has the shape of an inverted-U: early in their working life, managers hold very
simplistic views of their organization’s network. As they deal with ever increasing numbers of
actors, relationships, and changes, perceptions become increasingly granulated and complex.
Thus, initially network picture complexity increases with experience. After a certain point, i.e.
with more experience, mental pictures then become more abstract and therefore less granulated
(i.e. less differentiated), and thus less complex (Elster, 2000).
Proposition 1: There exists an inverted-U shape relationship between network picture
complexity and the number of years of an individual’s working experience
IMM Manuscript 10-562RR
Different job roles require different levels of knowledge or insight regarding what is taking place
in the firm’s environment. Hodgkinson and Johnson (1994) posit a relationship between a
managers’ job role, and managers’ cognitive taxonomies of the environment. Their research
found that the complexity of cognitive structures depends on the extent of knowledge that
individuals require to fulfill their organizational role. Managers’ perceptions as well as sensemaking structures can thus be expected to be conditioned by their functional areas (Dougherty,
1992; Hambrick and Mason, 1984; Lawrence and Lorsch, 1967).
However, Walsh’s (1988) study on selectivity and selective perception provided empirical
evidence that managers with a strong functional-area orientation could also work successfully in
a different area from that reflected in their perceptions. Managers were able to identify issues and
use information from domains that were dissimilar to the functions they were currently involved
in. One of the explanations provided by Walsh (1988) for this phenomenon was that managers’
belief structures could be conditioned by their previous experience in other functions. Poole et al.
(1990) argue that this could be explained by individuals’ direct (repetitive) and indirect (stories,
role models) experiences, given that both affect individuals’ knowledge structure development.
We therefore propose that managers’ network pictures are conditioned by their overall
experience in diverse functions, namely in mainly internally versus externally-oriented functions.
To develop this proposition, we draw strongly on Walsh’s (1988) study that shows evidence of
these expected differences in individuals’ belief structures: “managers with a marketing belief
structure identified more external management problems and requested more external
management information than managers in the human relations and generalists groups” (p.
IMM Manuscript 10-562RR
888). It can therefore be expected that individuals who are more experienced in externallyoriented functions (e.g. marketing, sales, business development, supply-chain and procurement
management) hold belief structures that are well informed regarding the company’s
surroundings. The network pictures of these managers’ are therefore assumed to be more
complex than those of individuals with internally-oriented functions (e.g. production, finance,
HRM) whose main interest and remit lie in operational issues within their own company.
Proposition 2: Network pictures are more complex for individuals with more experience in
externally-oriented functions (e.g. marketing) than in internally-oriented ones
(e.g. production)
5.2 Network picture complexity and company characteristics
The organizational context of decision makers (i.e. strategy and strategic capabilities of a firm,
organizational structure, and culture) informs their interpretation of what is taking place in their
surroundings (Ford and Bacus, 1987). This idea is reinforced by Isabella (1990) and by Öberg et
al. (2007) who provided empirical evidence that change in a manager’s information environment
(e.g. an acquisition, or company reorganization) results in changes in his/her frame of reference.
Through organizational memory, an organization is able to preserve its past through its culture,
structures, and changes, as well as in the minds and experiences of individuals (Levitt and
March, 1988; Walsh and Ungson, 1991). Organizational success results in individuals relying on
established institutionalized practices and engaging in a socializing process with the
organizational culture (Ford and Bacus, 1987): beliefs become ossified through repeated
IMM Manuscript 10-562RR
successful actions in a specific social context (Argyris and Schon, 1978; Daft and Weick, 1984;
Hedberg et al., 1976; Sandelands and Stablein, 1987).
Thus, the principle underlying proposition 1 on the individual level also applies at the
organizational level: as a result of organizational experience and success or failure, shared
understandings about what does or does not work in a network are developed and manifested at
the organizational level (Stewart and Latham, 1980). We therefore posit that there exists an
inverted-U shape relationship between an individual’s network picture complexity and the
number of years his/her company has been operating in a specific network or sector. Individuals
working in an organization which has been in a particular business sector only for a short time
are expected to exhibit low network picture complexity; with increasing company tenure in a
network this is expected to become gradually more granulated. Again, after time, individuals in
the company will exhibit less granulated and thus more abstract and less complex network
pictures due to ossifications of belief structures.
Proposition 3: There exists an inverted-U shape relationship between an individual’s network
picture complexity and the number of years that his/her company existed or
operated in a specific network/sector
It was argued above that internal organizational structure informs individuals’ sense-making
frameworks (Ford and Bacus, 1987). Additionally, previous research on organizational change
points to a reciprocal relationship between organizational structural and individual interpretative
schemes (Bartunek, 1984; Bartunek and Franzak, 1988; Greenwood and Hinings, 1988). Thus,
IMM Manuscript 10-562RR
organizational structure affects and is also affected by individuals’ frameworks. According to
Ranson et al. (1980), a structural change is more closely associated with changes in action than
with changes in interpretative schemes. This principle reflects Giddens’ (1979) notion of a
reciprocal relationship between organizational structure and individuals’ actions (and
understanding). This is echoed by Bartunek (1984) who argues that “organizations’ structural
properties both legitimize and constrain action. When interpretative schemes and their
expression in action change, then, structure will also undergo modification, which in turn will
legitimize and constrain later action and interpretative schemes” (p. 366). If the organizational
context conditions how individuals think about their surroundings (Ford and Bacus, 1987;
Isabella, 1990), then we posit that the more complex the organizational structure of a company,
the more complex the mental frameworks of individuals that work in that company.
Proposition 4: Network pictures are more complex for individuals in companies that have a
more complex internal structure
6 Research Method and Design
6.1 Empirical Grounding
In line with the second-order approach to complexity, we used a social constructionist,
interpretative frame (Guba and Lincoln, 2000; Isabella, 1990; Schwandt, 2000); thus capturing
individuals’ network pictures is not to “gather facts and measure how often certain patterns
occur, but to appreciate the different constructions and meanings that people place upon their
experience” (Easterby-Smith et al., 2003, p. 30). Eden et al.’s (1979) approach to exploring
IMM Manuscript 10-562RR
individuals’ cognitive structures, one which draws on personal construct theory (Kelly, 1955),
was identified as the most appropriate way to elicit and code individuals’ network pictures.
Individual network pictures are defined as the unit of analysis, and specific contextual factors are
introduced at two levels of analysis: the individual and company level. Two networks that were
expected to exhibit distinct relational exchange features were selected (Lambe et al., 2000). One
of the networks corresponded to a traditional view of relational exchange (i.e. as opposed to a
transactional exchange; Williamson, 1975), where there exists a long-term collaborative
relationship between business actors that can be explained with the evolutionary model of
business relationships (Dwyer et al., 1987; Ford, 1980; Håkansson, 1982; Morgan and Hunt,
1994): a non-project-based network. The second network was associated with a short-term
relational exchange situation characterized by low expectations about the existence of possible
future transactions: a project-based network, i.e. an interimistic relational exchange situation
(Purchase and Olaru, 2004).
The two selected networks were also expected to present approximate (and high) levels of
complexity. The high complexity of the non-project based network resulted, amongst other
factors, from the ‘hole-through-the-wall’ production and delivery system that it was based on;
this system usually implies high and tailored investments in machinery and equipment for singleclients, as well as joint planning and a tight collaboration at diverse levels between the client and
the supplier (Perks, 2004). On the other hand, the high complexity of the project based network
is associated, with its public nature, intensively involving actors of very diverse nature such as
City Councils, the Government, and private companies. A second reason is the nature of the
construction industry that it was grounded on: it is highly complex, due to its underlying
IMM Manuscript 10-562RR
uncertainty (as there are numerous individual tasks that have to be performed), and strong
interdependence between tasks (as diverse components have to be brought together to develop a
work-flow) (Gidado, 1996). More detailed information is provided below.
These two specific networks were chosen not only to cover diversified business activity settings
(i.e. networks with different relational exchange features), but also to have the level of
complexity as a control variable: both networks were considered to have similar levels of high
complexity. The data collection process was facilitated by the fact that the researchers were
given privileged access to some of the companies in both networks. The selected networks
therefore fulfilled what Brito (1999) classified as the two basic requirements for a selection of
this nature: appropriateness and accessibility. These two networks corresponded to two rich case
studies (Yin, 1994). A total of forty-seven respondents from seventeen companies were included
in the data collection: twenty-six respondents were interviewed in ten companies for the nonproject network, and the remaining twenty-one respondents were selected from a total of seven
companies for the project-based network. The companies included in the non-project based
network were selected by a snowballing or chain process (Gray, 2004; Ritchie and Lewis, 2003);
in the project based network, all companies directly included in the conception and execution of
the project were included in the network. The companies included in the non-project based
network were somehow associated with the practice of integrated outsourcing in the production
and delivery of plastics containers (i.e. suppliers, producers, clients), whilst the companies
selected for the project-based network were all involved in a project related to the transportation
sector (i.e. the sole client, consortiums of companies and some of the companies that were
involved in the conception, construction and operation of a civic Metro system). All respondents
IMM Manuscript 10-562RR
played a key role in the decision making process of their companies (e.g. CEOs, company
directors), and companies were chosen by adopting a snowball approach. The data collection
process took place over four months in 2005 and 2006. All interviews took place in Portugal and
were conducted in the native language. All interviews were transcribed verbatim into English, a
consequence of there being researchers involved in the data analysis process who were not
Portuguese native speakers.
Following Ramos et al.’s (2005) and Henneberg et al.’s (2006) procedure to collect data on
actors’ perceptions of their surrounding network, a combination of visual and verbal data
eliciting techniques was used to gather the data (Meyer, 1978; Tufte, 1983). Respondents were
asked to represent on a blank A4 sheet what they ‘saw’ in their business surroundings. This was
followed by a semi-structured interview during which respondents were asked to explain what
they had represented, as well as to answer some questions that would help to understand how
they perceived the network (i.e. questions regarding the scale and structure of the perceived
network, as well as the processes between the identified actors and their positioning). This is in
line with the approach of Eden et al. (1979) in eliciting cognitive structures. Some questions
regarding individual objective features, such as work experience, were also included in the
interview. Secondary data was collected to identify each company’s features (e.g. number of
years it had been in business), as well as the networks’ features (e.g. expected continuity of
relationships).
6.2 Analysis
IMM Manuscript 10-562RR
The data analysis encompassed two main stages. In the first, the complexity underlying each
respondent’s network picture (i.e. visual and textual data) was assessed. For this, we used the
three-dimensional construct of network picture complexity. Different complexity scales were
developed for each of the sub-dimensions (for a summary, see Appendix 1). Classifying each
respondent’s network picture complexity implied first of all scoring each sub-dimension on a
corresponding complexity scale. After scoring all complexity sub-dimension scales by
respondent, the overall complexity score for each main dimension (i.e. number and nature of
actors, number and nature of relationships, and dynamism and flexibility) was assessed as the
average of their corresponding sub-dimensions. Based on this, the complexity of a given network
picture was assessed, firstly, as a profile based on the scores for the three main complexity
dimensions, and secondly as an averaged overall complexity score (i.e. the average of the three
main complexity dimensions). Complexity could therefore vary between respondents overall, or
depending on their profile relating to the three complexity dimensions. Illustration of network
pictures with a high overall complexity score and with a low complexity score are provided in
Appendix 2 and in Appendix 3, respectively.
The scores for each complexity sub-dimension resulted from an initial rater interpretation. To
assure inter-rater reliability (Brito, 1999; Kaplan and Goldsen, 1965), two further researchers
were asked to classify independently the complexity for a subset of randomly selected network
pictures. No major differences between these three researchers’ interpretation was found, and
this replicability provided a sufficient indicator for the reliability of the analysis (Krippendorff,
2004; Ritchie and Lewis, 2003). We specifically checked that our ‘concepts-in-use’ overlapped
between raters (Gephart, 2004). Furthermore, as our research, although it uses ‘scores’, is based
IMM Manuscript 10-562RR
on interpretations of cases (which are represented through pictorial and textual data), the quality
and trustworthiness of such a case approach is important. By using several elements associated
with ‘innovative’ case research (Piekkari et al., 2010; Woodside, 2010), we ensure that the case
is relevant and rigorously analyzed. These elements are: population studies (within the selected
case networks), multiple method data collection (i.e. textual and pictorial), within and cross-case
analysis, and a combination of qualitative and quantitative elements (Gephart, 2004).
Furthermore, through a theory-driven development of dimensions of complexity, we ensure
construct validity (Dubois and Gibbert, 2010). In presenting our data and analysis, we follow the
seminal process suggested by Pratt (2009) for qualitative research.
A second stage of the data analysis specifically used the developed propositions and looked for
support for the associations between contextual factors (i.e. individual – years of working
experience and level of experience in internally versus externally oriented functions -, and
company specific characteristics – number of years existing or operating in a specific network or
sector and level of complexity of the internal structure), and network picture complexity for the
47 respondents. This implied capturing the contextual factors used in the four propositions that
were to be tested.
At the individual level, to assess respondents’ working experience, we considered the number of
years respondents had been working in the company in the non-project based network. For
respondents in the project based network, we considered solely the number of years they had
been involved in the project (due to the assumption of project particularity; Cova and Hoskins,
1997). Regarding respondents’ experience in different functions, several situations needed to be
IMM Manuscript 10-562RR
considered. For example, some respondents were currently working in internally-oriented
functions and had in the past (i.e. in the same company or in previous companies) also held
internally-oriented functions. However, some internally focused respondents had held externally
oriented functions in the past, and vice versa. Furthermore, some functions can be classified as
being simultaneously internally and externally oriented (e.g. CEOs). Thus, five groups of
respondents were considered: consistently internally oriented; consistently externally oriented;
simultaneously internally and externally oriented; currently internally/previously externally; and
currently externally/previously internal. As for the company level, to analyze the relationship
between network picture complexity and the number of years a respondents’ company had been
in working in the industry or sector, we only included those respondents from companies in the
non-project based network, i.e. companies which varied regarding their time span in the industry
or sector. The decision not to include in this analysis companies from the project based network,
resulted from the fact that all companies had been involved in the project since its inception.
Moreover, because most companies had been through processes of acquisitions or mergers, we
considered only the period after its last major restructuring. Finally, to assess the complexity of
companies’ internal structure, we draw mostly on respondents’ descriptions. Thus, this
contextual factor was based on content analysis whilst all other contextual factors that were
considered were objectively obtained through secondary research. We categorized all firms into
three levels of internal structure complexity: high, intermediate, and low.
This research design implies ‘quantifying’ the qualitative data. When discussing some of the
aspects of writing (‘good’) qualitative research, Pratt (2009) identifies this practice as a
potentially dangerous path to follow. However, the development and usage of complexity scales
IMM Manuscript 10-562RR
to assess the sense-making of respondents has enabled us not only to integrate into the analysis
multiple constructs but also to achieve a more precise idea of the differences and similarities that
existed between respondents. Furthermore, it was never our intention to generalize the findings
from the proposition-support but instead, to provide rich insights about network pictures
complexity specific to two specific networks. The quantification of the qualitative data does not
change the qualitative nature of the research; it represents instead a mere structuring device that
we have implemented to interpret qualitative data. The research remains anchored in a caseresearch approach.
Moreover, the research method and design specifically developed for this project can be
classified as an ‘innovative practice’ in terms of the application of the case study method to the
study of business networks (Piekkari et al., 2010). In line with requirements for ‘innovative
practice’, we use ‘proposition-analysis’, ‘non-traditional’ data gathering techniques (a
combination of in-depth interviews combined with visual eliciting techniques), as well as a
‘population study’ (i.e. all significant players for each network were included) (Piekkari et al.,
2010). Furthermore, both qualitative and quantitative approaches were applied by combining
qualitative content analysis and quantitative pattern matching.
6.3 Dimensions of network picture complexity: operationalization
In this section, each of the sub-dimensions identified in Table 1 is operationalized and the
relationships between overall complexity and each sub-dimension are explored. We also outline
the rationale underlying the complexity scales adopted for assessing network picture complexity
for each complexity sub-dimension. Appendix 1 provides an outline of each construct or
IMM Manuscript 10-562RR
complexity sub-dimension, as well as a summary of the complexity scales that were used for
each construct. Whilst the network pictures dimensions and corresponding sub-dimensions
described in section 4 were mostly theoretically derived (i.e. deductively identified; EasterbySmith et al., 2003), the final set of elements resulted from a refinement process where our
previous empirical work on network pictures played a central role. This previous experience
resulted in the development of certain sensitivity to the matter, allowing us to also introduce an
inductive approach to the development process.
Number and nature of actors
In our model we posit that network picture complexity increases with the number of actors,
either specific actors (sub-dimension 1.1.1) or groups of actors (sub-dimension 1.1.2). To score
the sub-dimensions, the minimum value of the scale (i.e. 1) was given to the lowest number of
actors identified in the forty-seven different network pictures. Consequently, the maximum value
of the scale (i.e. 5), corresponds to the highest number of actors identified in the sample. In
between these extreme points, the scale was used to score the remaining respondents, in effect
providing a distribution of complexity on a scale of 1 to 5.
We also suggest that network picture complexity increases with the diversity in the nature of the
actors that are identified. For this sub-dimension, we include business counterparts (subdimension 1.2.1) and non-business counterparts (sub-dimension 1.2.2). The scale, anchored in 1
(least complex) and 5 (most complex), was derived as above. Thus, a rating of 5 was attributed to
those respondents who identified the most diverse set of actors, whilst the minimum value of 1
was given to those actors who identified the less diverse set of actors. The identification of a set
IMM Manuscript 10-562RR
of actors is informed by stakeholder theory (Donaldson and Preston, 1995; Freeman and Reed,
1983; Smithee and Lee, 2004), more precisely by the ‘six markets’ model suggested by
Christopher et al. (1991) and later developed by Payne et al. (2005). The model includes the
following market domains: customer markets, referral markets, influencer markets, supplier
markets and internal markets. However, our conceptualization of potential stakeholders goes
beyond this body of literature, given that we undertake an industrial network approach to
business relationships (Håkansson et al., 2009). For example, the categories included in the ‘six
markets’ model do not make a clear distinction between business (e.g. supplier and clients) and
non-business counterparts (e.g. government, local or global community, competitors, and
industry associations). Such different domains are nevertheless a tested and recognizably reliable
way of identifying whom the potential stakeholders may be, and are therefore included in the set
of actors used.
Number and nature of relationships
Network picture complexity is proposed to increase with the number of relationships an actor
includes in his/her network. This number represents specific relationships (sub-dimension 2.1.1),
as well as groups of relationships, i.e. relationships between identified groups of actors (subdimension 2.1.2). Again, the scales for these two sub-dimensions are defined as above. Seven
levels of analysis are used to assess the extent of complexity regarding the perceived variety in
the nature of relationships outlined in table 1 (sub-dimension 2.2.), namely: variety regarding the
strength, immediateness, formality, directionality, multidisciplinarity, and substance of
relationships, as well as nature of the system of relationships. A summary of the literature used in
the development of these seven complexity sub-dimensions and the rationale underlying each
IMM Manuscript 10-562RR
adopted complexity scale is presented Appendix 1. All complexity scales are anchored in 1 (least
complex) and 5 (most complex). Three different logics are considered in the development of the
complexity scales for the seven levels of analysis included in sub-dimension 2.2, depending on
the number of aspects underlying each level of analysis. Each of the first four variables (i.e.
variety in the strength, immediateness, formality, and directionality of relationships) have two
different aspects. For example, for the variety in the strength of relationships, we consider both
weak and strong ties. When assessing both variety in the multidisciplinarity, and in the substance
of relationships, three possible aspects were considered: social, economic, and technological ties
for the former, and actor bonds, activity links and resource ties for the latter. Finally, for the
nature of the system of relationships, there are three alternatives: loose systems, tight systems, or
a hybrid of the two.
Dynamism and flexibility
We propose that an individual’s network picture is complex in terms of dynamism and flexibility
if it includes network changes that have taken place, are taking place, or are going to take place
in the future. Thus, we subsume a comparative static characteristic under dynamics. In our
operationalization, sub-dimension 3.1.1 identifies all changes of the number or nature of actors,
sub-dimension 3.1.2 measures all changes of the number or nature of relationships, and finally
sub-dimension 3.1.3 identifies the periods of time over which changes have occurred. The 5point complexity scales developed for these three sub-dimensions followed the same rule: the
richer (i.e. more elements included) the representation and description of the surrounding
business network is, the more complex is the respondent’s network picture (i.e. higher point in
the complexity scale). For example, an individual that talks solely about changes that took place
IMM Manuscript 10-562RR
in the past is considered having a less complex network picture than one that also mentions
changes taking place in the present.
7 Understanding Network Picture Complexity
To understand network picture complexity (i.e. to assess how and if complexity varies with
specific individual, and organizational characteristics), we analyze the complexity scores that
resulted from the first stage of the data analysis and look for support for the propositions that we
have posited. We look for associations between the overall network picture complexity, and the
individual, and organizational objective features used in the proposition development. The
suggested associations are also analyzed for the three main dimensions of network picture
complexity. The analysis is carried out considering the average complexity values found for all
respondents from both types of networks combined (i.e. for the 47 respondents), as well as for
each network separately, for each possible situation within each considered contextual factor.
---------------------------------------------Insert Tables 2, 3, and 4 about here
----------------------------------------------
7.1 Analyses at the individual level
Proposition 1 posits that there exists an inverted-U shaped relationship between network picture
complexity and the number of years of work experience. The analysis of the individual network
picture complexity scores indicates that, contrary to what was predicted, the data revealed a Ushaped relationship between work experience and complexity. This was valid when analyzing
both networks together and also separately (see tables 2, 3, and 4). For example, when analyzing
IMM Manuscript 10-562RR
both networks together (table 2), the average overall network picture complexity of individuals
with limited experience scored 2.22, decreasing to a score of 2.15 for individuals with
intermediate experience, and increasing to a score of 2.54 for individuals with more experience.
This situation can be illustrated with the following example from the non-project based network:
one of the involved company’s CEO who had joined that company (and industry) very recently
scored 2.76 for network complexity; the international director for a different company who had
been in employed for what we classified as an intermediate period of time scored 1.71; finally,
the financial director for another company in that network who has worked for that company for
a long period of time scored 2.57. This was a somewhat counter-intuitive finding, reflecting an
initially complex network picture that is reduced in time in terms of its complexity before it
increases again with more managerial experience. This finding was supported for the overall
network picture complexity for all respondents from both networks, as well as for the averages
for each separate network. A similar pattern was observed for two of the individual dimensions
of network picture complexity: number and nature of actors, and number and nature or
relationships. Thus, proposition 1 was not supported.
Considering respondents’ experience in different functions, the second characteristic considered
for the individual level, we considered how respondents could be, for example, consistently
internally oriented or simultaneously internally and externally oriented (as discussed above).
Considering both networks together, the average score of network picture complexity shows that
consistently or currently internally oriented respondents held less complex network pictures that
those respondents that were consistently or currently externally oriented (means scores of 2.09
and 2.12 versus 2.37 and 2.53; see table 2). The complexity for simultaneously internally and
IMM Manuscript 10-562RR
externally respondents’ network pictures was high and similar to those of respondents within
externally oriented functions (i.e. score of 2.37). These patterns were also found for most
individual dimensions of network picture complexity. However, for the dimension of flexibility
and dynamism, there was an exception: the average scores of complexity for this dimension
indicate that consistently or currently internally oriented respondents held more complex beliefs
than respondents that were consistently externally oriented (means scores of 2.47 and 2.50 versus
2.33; see table 2). Thus, the data generally supports proposition 2: network pictures are more
complex for individuals with more experience in externally-oriented functions (e.g. marketing)
than in internally-oriented ones (e.g. production).
However, analyzing both networks separately shows some differences: while the relationship
between complexity and functional orientation within the non-project based network was as
expected, the project-based networks shows a more specific pattern: the overall network picture
complexity for individuals that currently or consistently held an internally oriented functions
were higher than the complexity averages found for respondents that had always held externally
oriented functions. This finding occurs because our research characterized respondents working
for the organizational entity coordinating the overall project consortium as having externallyoriented functions. However, some of these respondents perceived themselves as internallyoriented. If one adopts this perspective, the proposed relationships are again supported.
7.2 Analysis on the company level
Our analyses show that the average network picture complexity for respondents that worked in
companies that had been in business for the shortest period of time (between 1 and 11 years) was
IMM Manuscript 10-562RR
lower than the average found for respondents in companies that had been in business for what we
defined as an intermediate period of time (between 12 and 20 years) (see table 3 or 4, section 2.1.
number of years in business). The average complexity score again decreased for respondents in
companies that had been in business for the longest period of time (i.e. between 21 and 30 years).
The same consistent pattern was also observed for each individual dimension of network picture
complexity. This can be exemplified for the non-project based network: the CEO and every other
manager in one company that had been in the market for what we classified as the shortest period
of time, scored on average 1.76 for network picture complexity, while the financial director of a
company which had been in business for an intermediate period of time, scored 2.41; and finally,
the production director at a company that had been present in the network for the longest period
of time scored 1.54. This finding supports proposition 3, that there exists an inverted-U shape
relationship between network picture complexity and the number of years a manager’s company
has existed or operated in a specific sector, or network.
Regarding the complexity of companies’ internal structure, Proposition 4 suggested that network
pictures are more complex for individuals in companies which have a complex internal structure
or organization. The data from both networks, when taken together as well as separately,
supports this proposition (see tables 3, 4 and 5). For example, we observed that in the projectbased network the CEO for a company that was categorized as having a less complex structure
scored 1.83; moreover, while at a company classified as having an intermediate complexity, the
CEO scored 2.05; and finally, the construction manager working at a company with a highly
complex structure scored 2.36. The same consistent relationships were also observed for each
IMM Manuscript 10-562RR
individual dimension of network picture complexity. Again, we replicated this analysis within
each network, and the results were robust in both cases.
Table 5 provides a summary of the outcome of the proposition-support considering the level of
complexity of the captured network pictures and the underlying contextual factors: Out of the
four propositions developed in section 5, only one was not supported.
-----------------------------------Insert Table 5 about here
-----------------------------------
8 Discussion
The findings from this investigation shed light on some of the factors that condition a specific
characteristic of network pictures, i.e. their complexity. According to the industrial network
theory and the concepts of management in network by Ford et al. (2002), to understand the
complexity of actors’ sense-making in networks is important: these subjective beliefs influence
actors’ decisions regarding their networking activities (Ford and Håkansson, 2006a; Mattsson,
1987; Ramos and Ford, 2011), i.e. belief conditions action (Weick, 1995). Whilst the link
between the complexity of individuals’ cognitive frameworks on the one hand, and decisionmaking on the other, has been extensively analyzed in the cognitive strategic group literature in
the area of strategic management (Lurigio and Carroll, 1985; McNamara et al., 2002; Streufert
and Nogami, 1989), this is not the case in business marketing, particularly vis-à-vis the interorganizational system of business networks. Therefore, this study provides a specific
understanding of network-related phenomena of sense-making in business marketing which,
IMM Manuscript 10-562RR
qualify some of the previous findings from strategic management regarding how contextual
factors influence the complexity of managers’ cognitive frameworks (Ford and Bacus, 1987;
Lurigio and Carroll, 1985).
The empirical analysis showed that network picture complexity is higher for individuals with
more experience in externally-oriented functions than in internally-oriented ones, and that the
more complex companies’ internal structures are, the more complex are the individuals’ network
pictures. The data analysis also indicated that there is an inverted-U shaped relationship between
network picture complexity and the number of years a company has been established in a
business. Finally, and unexpectedly, we found a U-shaped relationship between network picture
complexity and the number of years of an individual’s working experience. The latter finding
contradicts expectations by literature in the area of strategic management (Levitt and March,
1988; Reger and Palmer, 1996). We found that when joining a company, managers quickly
acquired complex views of their organization’s network. This may be associated with the need
they experience to become familiar with their working environment, as fast and comprehensibly
as possible. With managers’ growing experience, their views become more abstract and thus less
complex (Lurigio and Carroll, 1985; Starbuck and Milliken, 1988); however, with even more
experience in the company, the complexity of managers’ perception of their surrounding
business network increased again. This reinforces the idea that is put forward by several authors
such as Weick (1995) or Hooijberg (1996), who claim that for achieving greater effectiveness,
there is a need for greater diversity or higher complexity. Further research is required to
understand this unexpected finding.
IMM Manuscript 10-562RR
We focused on contextual factors at two levels: individual and organizational. However, in other
bodies of literature, the features of the business network where companies are embedded have
also been found to condition individual interpretations of the surroundings (Barr et al., 1992).
Fahey and Narayanan (1989) found no relationship between changes in the objective features of
companies’ macro environments and their employees’ frames of reference, while Barr et al.
(1992) provide empirical evidence for the impact of environmental discontinuities on changes in
managers’ belief systems. Moreover, it is commonly accepted that there is a “shared knowledgebase that those socialized in an industry take as familiar professional common sense” (Spender,
1989, p. 69), such as a strategic frame or industry recipe. Porac et al. (1989), and Stubbart and
Ramaprasad (1988) are some of the authors whose work provides empirical evidence for the
existence of shared mental representations within an industry or network. For these reasons, the
systemic characteristics of the network or the context in which business activity takes place may
also be expected to influence actors’ cognitive complexity.
Business networks can be classified in different ways. For example, considering the aim that
networks set to achieve, de Man (2004) distinguishes between quasi-integration networks, supply
or customer oriented networks, and technology oriented networks. On the other hand, Möller and
Svahn (2006) consider specific value creation logics underlying network creation and
development, and suggest the existence of current business nets, business renewal nets, and
emerging business nets. The classification that we have employed in our empirical analysis, with
the aim of covering diverse network contexts in the analysis, was that suggested by Lambe et al.
(2000), which draws on the dominant relational exchange logic of networks: the authors
differentiate between enduring relational exchanges (i.e. our non-project based network), and
IMM Manuscript 10-562RR
interimistic relational exchanges that are specific to a defined interaction period or project (i.e.
our project-based network). For the reasons discussed below, which draw on the extant project
literature as well as on our empirical investigation, one may expect network pictures to be more
complex for actors in project-based networks than in non-project based ones. This is the case if
all other contextual factors remain ceteris paribus (e.g. individual tenure).
There are essentially three project based networks features that appear relevant for understanding
network picture complexity. First, actors involved in projects rely strongly on formalized and
contractually agreed explicit norms (Cova et al., 1994). This happens because there is not much
time available to develop relational rules or norms (i.e. implicit or emergent rules) between the
actors involved. It is also a consequence of projects’ particularity (Cova and Hoskins, 1997):
managers cannot capitalize extensively on their experiences from other projects, and thus
investing in the development of non-transferable knowledge or explicit rules could lead to a
waste of resources. Secondly, as a safe-guard to those involved, project contracts can be
expected to be as comprehensive as possible, with high levels of complexity being explicitly
stipulated in written documents and contracts. As a result of these two factors, whatever is
defined in the contract will typically need to be extensively incorporated by individuals in their
sense-making processes. This is bound to affect the level of complexity underpinning actors’
cognition: individuals may be expected to identify more actors, more relationships, a greater
variety of relationships and more changes (e.g. changes that take place in the consortium
agreement throughout time and that are all registered through amendments), than in a non-project
based network; in the latter, individuals rely more on what they “hear” and “experience”. This
IMM Manuscript 10-562RR
association between network structures and the complexity of actors’ cognition is therefore an
area that deserves further investigation.
9 Conclusions, Limitations, and Future Research
In this study, we have investigated issues around sense-making in business systems, specifically
one aspect of business actors’ network pictures that had not been addressed before in the
literature: their complexity. Previous research on network pictures has mainly concentrated on
identifying their dimensions, on analyzing the contextual factors that condition their content, and
on demonstrating how the concept could be used better to understand business network related
phenomena such as network change. Introducing the construct of complexity to the study of
network pictures provides additional understanding of network picture characteristics.
We have developed a multi-dimensional construct of network picture complexity, based on
existing complexity theory, thereby drawing on the extant literature on both sense-making
cognitive models and business networks. Given that previous research shows that the complexity
of managers’ cognitive models is conditioned by contextual factors, we focused on
understanding how such factors related to complexity underlying business actors’ network
pictures. We developed multiple propositions about the associations between cognitive
complexity on the one hand, and individual, and organizational factors on the other. Given the
nature of the research (i.e. qualitative and exploratory), we did not try or aim at testing the
posited propositions; instead, the suggested propositions were tentatively supported (or not) and
illustrated (or not) with the collected and analyzed data. Whilst three of the propositions were
empirically illustrated, thus corroborating what had been previously found in the strategic
IMM Manuscript 10-562RR
management literature, this was not the case for the propositions relating to the relationship
between network picture complexity and the number of years of an individuals’ working
experience. While some tentative explanations for this have been put forward, this issue merits
further investigation, e.g. regarding the contextual and cultural dimensions which influence this
relationship.
9.1 Contributions and Managerial Implications
This research borrows from other bodies of literature the notion that the complexity of actors’
knowledge structures conditions how they perceive their surroundings, thereby conditioning
actors’ strategic responses, decision making and analytical skills, and ultimately firms’ actions
and performance. We build on this notion to develop a similar argument in the business network
area of research, and we therefore argue that the complexity of actors’ network pictures
represents an important and pivotal aspect of understanding business marketing performance.
This study therefore contributes to a growing multidisciplinary approach to business marketing,
specifically resulting from a cross-fertilization with the cognitive strategic group literature in the
area of strategic management (Yiu and Saner, 2000; Streufert and Nogami, 1989; Walsh, 1988).
The development of a multi-dimensional construct of network picture complexity is also an
important contribution to the extant literature on cognition in business networks. The analytical
tool that is theoretically derived can be used to assess the complexity of network pictures.
Furthermore, we have provided empirical evidence on the relationship between network picture
complexity and several contextual factors. The extent to which managers understand their
surrounding business network can thus be said to be dependent on diverse individual and
organizational related factors.
IMM Manuscript 10-562RR
Although there is no final agreement on how complex an actor’s cognitive structure should be to
achieve better decision-making and firm performance (McNamara et al., 2002; Walsh, 1988;
Weick, 1995), we believe that it is important for managers to be aware of the current levels of
network picture complexity (their own, their top management team, as well as across the network
in different companies). This may be used by managers to develop strategies for achieving a
portfolio of different levels of network picture complexity with regard to their top decisionmakers (at the company level) and/or counterparts (at the network level). Such a strategy would
allow managers to avoid situations previously described in the literature as ‘collective ignorance’
(Weick, 1979), ‘cognitive oligopolies’ (Porac et al., 1989) or ‘common stereotyping’ (Ford et al.,
2002; Halinen and Salmi, 1999; Henneberg et al., 2006), i.e. situations in which
comprehensively shared collective mental maps result in decisions being made based on a
restricted view of the surroundings. Kragh and Andersen (2009) investigated the relation
between the commonality of actors’ network pictures (i.e. actors from different companies but in
the same network) and the likelihood of success resulting from the implementation of a changerelated activity. These authors identified an inverted-U shape relationship between these two
variables (i.e. network picture commonality, and business relationship success), concluding that
too much overlap between actors’ network pictures can be problematic. It may be that a very
similar relationship also characterizes the level of complexity overlap between the network
pictures held by the different actors within a company, as well as between companies. Moreover,
it is also relevant for managers to understand how different levels of complexity are associated
with specific contextual factors, as well as the relationship between network picture complexity
IMM Manuscript 10-562RR
and networking. This information may also be used by managers when developing the portfolio
strategy described above.
9.2 Limitations and Implications for Future Research
Further research should be conducted using the research methodology and design applied in this
paper (including the developed complexity scales) with respondents from other project and nonproject based networks, as well as integrating companies and individuals with diverse features.
The aim would be to verify if our findings on the association between network picture
complexity and contextual factors were corroborated. Additionally, other independent variables
that could potentially explain differences in network picture complexity, such as hierarchical
levels, should be added (Day and Lord, 1992; Ireland et al., 1987). Moreover, drawing on
findings from other bodies of literature that show the association between an actor’s environment
and his/her frame of reference (Barr et al., 1992), we suggest extending the analysis of the
relationship between network picture complexity and contextual factors to include network
structures.
Furthermore, given that a way to assess the complexity of actors’ network pictures has been
developed and work has been done to grasp how complexity can be dependent on several factors,
further research needs to link network picture complexity on the one hand, and strategic
networking on the other. This can be done, for example, by conducting experiments that simulate
different cognitive maps (e.g. by providing stimuli in the shape of network pictures which can be
manipulated regarding the complexity they exhibit) (Corsaro et al., 2011; Nadkarni and Barr,
2008). This needs to be linked to decision situations and different strategic decision outcomes
IMM Manuscript 10-562RR
regarding network activities (Turnbull et al., 1996), e.g. the distinction between consolidating,
conforming, and conceding relational activities, and creating, confronting, and coercing
approaches (Ford et al., 2003). Alternative dependent constructs for strategic networking
activities could also be developed from the resource-acquisition strategies postulated in Ivens et
al. (2009) or Hoffmann’s (2007) discussion of shaping, adopting, and stabilizing strategies in
alliance networks. Ideally, instead of conducting experiments to investigate the relationship
between different levels of network picture complexity and networking choices, research could
be conducted in real settings, capturing the level of complexity of managers’ network pictures
and relating those levels to the observed interaction decisions and practices on an organizational
level. Moreover, drawing on the ‘cinematographic’ metaphor proposed by Purchase et al. (2010)
for the concept of network pictures, we suggest that this investigation should be process-based:
to introduce a stronger element of dynamism into the analysis, network pictures and networking
decisions would be collected over a period of time. In order to carry this out, the network picture
complexity analysis suggested here would possibly have to be simplified. If this were the case,
the most appropriate approach would be to leave the third dimension of complexity (i.e.
dynamism and flexibility) out of the equation, simply because this dimension would already be
captured with this processual approach.
Additionally, following Kragh and Andersen’s (2009) research, it would also be interesting to
investigate the impact that high or low levels of network picture complexity overlap between
actors in the same network (within and between companies) may have on the success of
networking activities.
IMM Manuscript 10-562RR
IMM Manuscript 10-562RR
REFERENCES
Abrahamsen, M., Naudé, P., & Henneberg, S. (2009). Explaining Change in Networks Using Network
Pictures: The Norwegian/Japanese Seafood Distribution System. 4th International Conference on
Business Market Management, Copenhagen, Denmark.
Anderson, J., Håkansson, H., & Johanson, J. (1994). Dyadic Business Relationships within a Business
Network Context. Journal of Marketing, 58, 1-16.
Argyris, C. & Schon, D. (1978). Organizational Learning: A Theory of Action Perspective. Reading,
Massachusetts: Addison-Wesley.
Ashby, W. (1956). An Introduction to Cybernetics. London: Chapman and Hall.
Axelsson, B. & Easton, G. (1992). Industrial Networks: A New View of Reality. New York and London:
Routledge.
Baraldi, E. & Nadin, G. (2006). The Challenges in Digitalising Business Relationships. The Construction
of an IT Structure for a Textile-Related Business Network. Technovation, 26(10), 1111-1126.
Baraldi E., & Strömsten, T. (2006). Embedding, Producing and Using Low Weight: Value Creation and
the Role of the Configuration of Resources Interfaces in the Networks around Ikea's Lack Table and
Holmen's Newsprint. IMP Journal, 1(1), 39-69.
Barr, P., Stimpert, J., & Huff, A. (1992). Cognitive Change, Strategic Action, and Organizational
Renewal. Strategic Management Journal, 5, 129-37.
Bartunek, J. (1984). Changing Interpretative Schemes and Organizational Restructuring: The Example of
a Religious Order. Administrative Science Quarterly, 29(3), 355-372.
Bartunek, J. & Franzak, F. (1988). The Effects of Organizational Restructuring on Frames of Reference
and Cooperation. Journal of Management, 4(4), 579-592.
Bartunek, J., Gordon, J., & Weathersby, R (1983). Developing Complicated Understanding in
Administrators. Academy of Management Review, 8(2), 273-284.
Brito, C. (1999). Issue-Based Nets: a Methodological Approach to the Sampling Issue in Industrial
Networks Research . Qualitative Marketing Research: An International Journal, 2(2), 92-102.
Brown, A., Stacey, P., & Nandhakumar, J. (2008). Making Sense of Sensemaking Narratives. Human
Relations, 61(8), 1035-1062.
Byrne, D. (1998). Complexity Theory and the Social Sciences. London and New York: Routledge.
Carraher, S. & Buckley, M. (1996). Cognitive Complexity and the perceived Dimensionality of Pay
Satisfaction . Journal of Applied Psychology, 81(1), 102-109.
Casti, J. (1994). Complexification: Explaining a Paradoxal World through the Science of Surprise. New
York: Harper Perennial.
Christopher, M., Payne, A. & Ballantyne, D. (1991). Relationship Marketing: Bringing Quality, Customer
Service and Marketing Together. Oxford: Butterworth-Heinemann.
Colville I. & Pye, A. (2010). A sensemaking perspective on network pictures. Industrial Marketing
Management, 39(3), 372-380.
Corsaro, D., Ramos, C., Henneberg, S. & Naudé, P. (2011). Actor Network Pictures and Networking
Activities in Business Networks: An Experimental Study, Industrial Marketing Management,
Special Issue, 40(6), 919-932.
Cova, B. & Hoskins, S. (1997). A Twin-Track Networking Approach to Project Marketing. European
Management Journal, 15(5), 546-556.
Cova, B., Mazet, F., & Salle, R. (1994). From Competitive Tendering to Strategic Marketing: An
Inductive Approach for Theory-Building. Journal of Strategic Marketing, 2(1), 29-48.
Cova, B. & Salle, R. (2008). Creating Superior Value through Network Offerings. In A. Woodside, F.
Golfetton, & M. Gibbert (Eds.), Creating and Managing Superior Customer Value, Advances in
Business Marketing and Purchasing, Vol. 14 (pp. 317–342). Bingley: JAI Press.
Daft, R. & Weick, K. (1984). Toward a Model of Organizations as Interpretation Systems. Academy of
Management Review, 9(2), 284-295.
IMM Manuscript 10-562RR
Day, D. & Lord, R. (1992). Expertise and Problem Categorization: The Role of Expert Processing in
Organizational Sense-Making. Strategic Management Studies, 29(1), 35-47.
de Man, A. (2004). The Network Economy. Strategy, Structure and Management. Cheltenham, UK,
Edward Elgar.
Donaldson, T. & Preston, L. (1995). The Stakeholder Theory of the Corporation: Concepts, Evidence and
Implications, Academy of Management Review, 20(1), 65-91.
Dougherty, D. (1992). Interpretative Barriers to Successful Product Innovation in Large Firms.
Organization Science, 3(2), 179-202.
Dubois, A. & Gibbert, M. (2010). From Complexity to Transparency: Managing the Interplay between
Theory, Method, and Empirical Phenomena in IMM Case Studies, Industrial Marketing
Management, 39(1), 129-136.
Dwyer, R., Schurr, P., & Oh, S. (1987). Developing Buyer-Seller Relationships. Journal of Marketing,
51(2), 11-27.
Easterby-Smith, M., Thorpe, R., & Lowe, A. (2003). Management Research (2nd ed.). Sage Publications.
Easton, G. (1995). Methodology and Industrial Networks. in K. Moller & D. Wilson (eds.), Business
Marketing: An Interaction and Network Perspective (pp. 411-492). Massachusetts: Kluwer
Academic Publishers.
Easton, G. & Araujo, L. (1994). Market Exchange, Social Structures and Time. European Journal of
Marketing, 28(3), 72-84.
Easton, G., Wilkinson, I., & Georgieva, K. (1997). On the Edge of Chaos: Towards Evolutionary Models
of Industrial Networks. in H. Gemunden, T. Ritter, & A. Walter (eds.), Relationship and Networks
in International Markets (pp. 273-293).Oxford: Elsevier/Pergamon.
Eden, C. & Ackermann, F. (1992). The Analysis of Cause Maps. Journal of Management Studies, 29(3),
309-324.
Eden, C., Jones, S., & Sims, D. (1979). Thinking in Organizations. London: MacMillan.
Elster, J. (2000). Ulysses Unbound. Cambridge: Cambridge University Press.
Fahey, L. & Narayanan, V. (1989). Linking Changes in Revealed Causal Maps and Environmental
Change: an Empirical Study. Journal of Management Studies, 26(4), 361–378.
Fioretti, G. & Visser, B. (2004). A Cognitive Interpretation of Organisational Complexity. Emergence,
Special Double Issue, 6 (1/2), 11-23.
Fiske, S., Kinder, D., & Larter, W. (1983). The Novice and the Expert: Knowledge-Based Strategies in
Political Cognition. Journal of Experimental Social Psychology, 19(4), 381-400.
Ford, D. (1980). The Development of Buyer-Seller Relationships in Industrial Markets. European
Journal of Marketing, 14(5/6), 339-353.
Ford, D., Gadde L.-E., Håkansson H., Snehota I. (2002). Managing Networks. 18th IMP Conference.
Perth, Australia.
Ford, D., Gadde L.-E., Håkansson H., & Snehota I. (2003). Managing Business Relationships, (2nd ed.).
Chichester: John Wiley & Sons Ltd.
Ford, D. & Håkansson, H. (2006a). IMP - Some Things Achieved: much more to do. European Journal of
Marketing, 40(3/4), 248-258.
Ford, D. & Håkansson, H. (2006b). The Idea of Interaction. The IMP Journal, 1(1), 4-27.
Ford, D. & Ramos, C. (2006). Network Pictures: Conceptual Development and Preliminary Findings.
22nd IMP Annual Conference, Milan, Italy.
Ford, D. & Redwood, M. (2005). Making Sense of Network Dynamics through Network Pictures: A
Longitudinal Case Study. Industrial Marketing Management, 34(7), 648-657.
Ford, J. & Bacus, D. (1987). Organizational Adaptation to Performance Downturns. Academy of
Management Review, 12(2), 366-380.
Freeman, R. & Reed, D. (1983). Stockholders and Stakeholders: a New Perspective on Corporate
Governance. California Management Review, 25(3), 88-106.
Gadde, L.-E., Huemer, L., & Håkansson, H. (2003). Strategizing in Industrial Networks. Industrial
Marketing Management, 32(5), 337-354.
IMM Manuscript 10-562RR
Geiger, S. & Finch, J. (2010). Networks of Mind and Networks of Organizations: The Map Metaphor in
Business Network Research. Industrial Marketing Management, 39(3), 381-389.
Gephart, R. P. (2004). Qualitative Research and the Academy of Management Journal, Academy of
Management Journal, 47(4), 454-462.
Gidado, K. (1996), Project Complexity: The Focal Point of Construction Production Planning.
Construction Management and Economics, 14(3), 213-225.
Giddens, A. (1979). Central Problems in Social Theory. Berkeley, CA: University of California Press.
Gioia, D. & Chittipeddi, K. (1991). Sensemaking and Sensegiving in Strategic Change Initiation.
Strategic Management Journal, 12(6), 433-448.
Granovetter, M. (1973). The Strength of Weak Ties. The American Journal of Sociology, 78(6), 13601380.
Granovetter, M. (1985). Economic Action and Social Structure: The Problem of Embeddedness.
American Journal of Sociology, 91(3), 481-510.
Granovetter, M. (1992). Problems of Explanation in Economic Sociology. In N. Nhoria & R. Eccles
(eds.), Networks and Organizations: Structure, Form and Action (p. 25-56). Boston: Harvard
Business School Press.
Gray, D. (2004). Doing Research in the Real World. London: Sage Publications.
Greenwood, R. & Hinings, C. (1988). Organizational Design Types, Tracks and the Dynamics of
Strategic Change. Organization Studies, 9(3), 293-316.
Guba, E. & Lincoln, Y. (2000). Competing Paradigms in Qualitative Research. in N. Denzin & Y.
Lincoln (eds.), Handbook of Qualitative Research (pp. 105-117). Thousand Oaks: Sage.
Gulati, R. (1995). Social Structure and Alliance Formation Patterns: A Longitudinal Analysis.
Administrative Science Quarterly, 40(4), 619-652.
Gulati, R., Nohria, N., & Zaheer, A. (2000). Strategic Networks. Strategic Management Journal, 21(3):
203-215.
Håkansson, H. (1982). International Marketing and Purchasing of Industrial Goods: An Interaction
Approach. IMP Group: John Wiley and Sons, Ltd.
Håkansson, H. & Ford, I. D. (2002). How Should Companies Interact in Business Networks? Journal of
Business Research, 55(2), 133-39.
Håkansson, H. & Johanson, J. (1992). A Model for Industrial Networks. In B. Axelsson & G. Easton
(eds.), Industrial Networks - A New View of Reality (pp. 28-34). London and New York: Routledge.
Håkansson, H. & Snehota, I. (1989). No Business is an Island: The Network Concept of Business
Strategy. Scandinavian Journal of Management, 5(3), 187-200.
Håkansson, H. & Snehota, I. (1995). Developing Relationships in Business Networks. London and New
York: Routledge.
Håkansson, H., Ford, D., Gadde, L.-G., Snehota, I. & Waluszewski, A. (2009). Business in Networks.
Chichester: John Wiley.
Halinen, A. & Salmi, A. (1999). From Dyadic Change to Changing Business Networks: An Analytical
Framework. Journal of Management Studies, 36(6), 779-795.
Halinen, A. & Salmi, A. (2001). Managing the Informal Side of Business Interaction: Personal Contacts
in the Critical Phases of Business Relationships. IMP Annual Conference. Oslo, Norway.
Halinen, A. & Tornroos, J.-A. (1995). The Meaning of Time in the Study of Industrial Buyer-Seller
Relationships. In K. Moller & D. Wilson (eds.), Business Marketing: An Interaction and Network
Perspective (p. 493– 529). Boston, Dordrechst and London: Kluwer Academic Publishers.
Halinen, A. & Tornroos, J.-A. (2005). Using Case Methods in the Study of Contemporary Business
Networks. Journal of Business Research, 58(9), 1285-1297.
Hambrick, D. & Mason, P. (1984). Upper Echelons: The Organisation as a Reflection of its Top
Managers. Academy of Management Review, 9(2), 193-206.
Hedberg, B., Nystrom, P., & Starbuck, W. (1976). Camping on Seesaws: Prescriptions for a SelfDesigning Organization. Administrative Science Quarterly, 21(1), 41-65.
IMM Manuscript 10-562RR
Hedaa, L. & Tornroos, J.-A. (2002). Towards a Theory of Timing: Kairology in Business Networks. In R.
Whipp, B. Adam, & I. Sabelis (eds.), Making Time: Time and Management in Modern
Organisations (p. 31-45). Oxford: Oxford University Press.
Henneberg, S., Mouzas, S., & Naudé, P. (2006). Network Pictures: Concepts and Representations.
European Journal of Marketing, 40(3/4), 408-429
Henneberg, S., Mouzas, S., & Naudé, P. (2009). Going Beyond Customers: A Business Segmentation
Approach Using Network Pictures to Identify Network Segments. Journal of Business Market
Management, 3(2), 91-113.
Henneberg, S., Mouzas, S., & Naudé, P. (2010). Sense-Making and Management in Business Networks Some Observations, Considerations, and a Research Agenda. Industrial Marketing Management,
39(3), 355-360.
Hershey, D., Walsh, D., Read, S., & Chulef, A. (1990). The Effects of Expertise on Financial Problem
Solving: Evidence for Goal-Directed, Problem-Solving Scripts. Organizational Behavior and
Human Decision Processes, 46(1), 77-101.
Hodgkinson, G. & Johnson, G. (1994). Exploring the Mental Models of Competitive Strategists: The
Case for a Processual Approach. Journal of Management Studies, 31(4), 525-51.
Hoffmann, W. (2007). Strategies for Managing A Portfolio of Alliances. Strategic Management Journal,
28(8), 827-856.
Holland, J. (1987). The Global Economy as an Adaptive Process. In P. Anderson & D. Pines (eds.), The
Economy as an Evolving Complex System (p. 117-124). Redwood City: Santa Fe Institute,
Addison-Wesley.
Holmen, E., Aune, T., & Pederson, A.-C. (2008). Network Pictures and Network Potentials in Supply
Management. 24th IMP Conference. Uppsala, Sweden.
Hooijberg, R. (1996). A Multidirectional Approach Towards Leadership: An Extension of the Concept of
Behavioral Complexity. Human Relations, 49(7), 917-946.
Ireland, R., Hitt, M., Bettis, R. & de Porras, D. (1987). Strategy Formulation Processes: Differences in
Perceptions of Strength and Weaknesses Indicators and Environmental Uncertainty by Managerial
Level, Strategic Management Journal, 8(5), 469-458.
Isabella, L. (1990). Evolving Interpretations as Change Unfolds: How Managers Construe Key
Organizational Events. Academy of Management Journal, 33(1), 7-41.
Ivens, B., Pardo, C., Salle, R., & Cova, B. (2009). Relationship Keyness: The Underlying Concept for
Different Forms of Key Relationship Management. Industrial Marketing Management, 38(5), 513519.
Jablin, F. (1987). Formal organization structure. In F. Jablin, Porter, L., Roberts, K. (eds.), Handbook of
Organizational Communication. An interdisciplinary perspective (pp. 389-419). Newbury Park:
Sage Publications.
Jack, S. (2005). The Role, Use and Activation of Strong and Weak Network Ties: A Qualitative Analysis.
Journal of Management Studies, 42(6), 1233-1259.
Jackson, S. & J. Dutton (1988). Discerning Threats and Opportunities. Administrative Science Quarterly,
33(3), 370-387.
Jarillo, C. (1988). On Strategic Networks. Strategic Management Journal, 9(1), 31-41.
Johnston, W., Lewin, J., & Spekman, R. (1999). International Industrial Marketing Interactions: Dyadic
and Network Perspectives. Journal of Business Research, 46(3), 259-271.
Johnston, W., Peters, L., & Gassenheimer, J. (2006). Questions about Network Dynamics:
Characteristics, Structures and Interactions. Journal of Business Research, 59(8), 945-954.
Kaplan, A. & Goldsen, J. (1965). The Reliability of Content Analysis Categories. In Lasswell, H., Leittes,
N. and Associated (Eds.) The Language of Politics: Studies in Quantitative Semantics (p. 83-112).
Cambridge, MIT Press.
Kauffman, S (1992). Origins of Order: Self-Organisation and Selection in Evolution. Oxford: Oxford
University Press.
Kelly, G. (1955). The Psychology of Personal Constructs. New York: Norton.
IMM Manuscript 10-562RR
Knight, L. (2002). Network Learning: Exploring Learning by Interorganizational Networks. Human
Relations, 55(4), 427-454.
Kragh, H. & Andersen, P. (2009). Picture this: Managed Change and Resistance in Business Network
Settings. Industrial Marketing Management, 38(6), 641-653.
Krippendorff, K (2004). Content Analysis: An Introduction to its Methodology (2nd edn.).California: Sage
Publications.
Lambe, C., Speckman, R., & Hunt, S. (2000). Interimistic Relational Exchange: Conceptualization and
Propositional Development. Journal of the Academy of Marketing Science, 28(2), 212-225.
Larkin, J., McDermott J., Simon D., Simon H. (1980). Expert and Novice Performance in Solving Physics
Problems. Science, 208 (4450), 1335-1342.
Larson, A. (1992). Network Dyads in Entrepreneurial Settings: A Study of the Governance of Exchange
Relationships. Administrative Science Quarterly, 37(1), 76-104.
Lawrence, P. & Lorsch, J. (1967). Differentiation and Integration in Complex Organisations.
Administrative Science Quarterly, 12(1), 1-47.
Leek, S. & Mason, K. (2008). Network Pictures: Building an Holistic Representation of a Dyadic
Business-to-Business Relationship. 24th IMP Conference. Uppsala, Sweden.
Levitt, B. & March, J. (1988). Organizational Learning. Annual Review of Sociology, 14, 319-40.
Levy (1994). Chaos Theory and Strategy: Theory, Application and Managerial Implications. Strategic
Management Journal, 15, Issue Supplement S2, 167-78.
Lissack, M. (1999). Complexity: the Science, its Vocabulary and its Relation to Organizations.
Emergence, 1(1), 110-1126.
Lowe, S., Ellis, N., & Purchase, S. (2008). Rethinking Language in IMP Research: Networking Processes
in Other Words. Scandinavian Journal of Management, 24(2), 295-307.
Lurigio, A. & Carroll, J. (1985). Probation Officers Schemata of Offenders: Content, Development, and
Impact on Treatment Decisions. Journal of Personality and Social Psychology, 48(5), 1112-1126.
Maguire, S. & McKelvey, B. (1999). Complexity and Management: Moving from Fad to Firm
Foundations. Emergence, 1(2), 19-61.
Marschan, R., Welch, D., & Welch, L. (1996). Control in Less-Hierarchical Multinationals: the Role of
Personal Networks and Informal Communication. International Business Review, 5(2), 137-150.
Mason, M. (2008). What is Complexity Theory and What are its Implications for Educational Change?
Educational Philosophy and Theory, 40(1), 35-49.
Mattsson, L-G. (1985). An Application of a Network Approach to Marketing: Defending and Changing
Market Positions. In N. Dholakia & J. Arndt (eds.), Changing the Course of Marketing: Alternative
Paradigms for Widening Marketing Theory, Research in Marketing: a Research Annual,
Supplement 2 (pp. 263-88). London: Jay Press.
Mattsson, L-G. (1987). Management of Strategic Change in a "Market-as-Networks" Perspective. In A.
Pettigrew (ed.), The Management of Strategic Choice (pp. 234-260). Oxford: Basil Blackwell.
Mattsson, L.G. (1988). Interaction Strategies: A Network Approach. AMA Summer Marketing Educators
Conference. San Francisco, California: AMA, 310-315.
Mattsson, L-G. (2005). Managing in Markets as Interacting with Market Theories and Market Practice.
1st IMP Journal Seminar Series. Oslo, Norway, 1-18.
McNamara, G., Luce, R., & Tompson, G. (2002). Examining the Effect of Complexity in Strategic
Groups on Firm Performance. Strategic Management Journal 23, 153-170.
Medlin, C. & Tornroos, J.-A. (2007). Inter-Firm Interaction from a Human Perspective. Finanza
Marketing and Produzione, 41-56.
Meyer, A. (1978). Management and Strategy. In R. Miles & C. Snow (eds.), Organizational Strategy.
New York: McGraw-Hill.
Miller, D. (1993). The Architecture of Simplicity. Academy of Management Review, 18(1), 116-138.
Mohr, J. & Nevin, J. (1990). Communication Strategies in Marketing Channels: a Theoretical
Perspective. Journal of Marketing, 54(4), 36-52.
IMM Manuscript 10-562RR
Möller, K. & Svahn, S. (2006). Role of Knowledge in Value Creation in Business Nets. Journal of
Management Studies, 43(5), 985-1007.
Morgan, M. & Hunt, S. (1994). The Commitment-Trust Theory of Relationship Marketing. Journal of
Marketing, 58(3), 20-38.
Morrison, K. (2002). School Leadership and Complexity Theory. London and New York:
Routledge/Falmer.
Mouzas, S., Henneberg, S., & Naudé, P. (2008). Developing Network Insight. Industrial Marketing
Management, 37(2), 167-180.
Nadkarni, S., & Barr, P. (2008). Environmental Context, Managerial Cognition, and Strategic Action: an
Integrated View. Strategic Management Journal, 29(13), 1395-1427.
Neisser, U. (1967) Cognitive Psychology. New York: Appleton-Century-Crafts.
Neisser, U. (1976) Cognition and Reality: Principles and Implication of Cognitive Psychology. San
Francisco: W.H. Freeman and Company.
Nohria, N. & Eccles, R. (1992). Networks and Organizations: Structure, Form and Action. Boston,
Massachusetts, Boston: Harvard Business School Press.
Öberg, C., Henneberg, S., & Mouzas, S. (2007). Changing Network Pictures: Evidence from Mergers and
Acquisitions. Industrial Marketing Management, 36(7), 926-940.
Payne, A., Ballantyne, D. & Christopher, M. (2005). A Stakeholder Approach to Relationship Marketing
Strategy: The Development and Use of the “Six Markets” Model. European Journal of
Marketing, 39(7/8), 855-871.
Perks, H. (2004). Exploring Processes of Resource Exchange. International Journal of Innovation
Management, 8(1), 37-61.
Perrow, C. (1984). Normal Accidents: Living with High-Risk Technologies. New Jersey: Princeton
University Press.
Pfeffer, J. & Salancik, G. (1978). The External Control of Organizations: A Resource Dependence
Perspective. New York: Harper and Row, Publishers.
Piekkari, R., Plakoyiannaki, E., & Welch, C. (2010). Good Case Research in Industrial Marketing:
Insights from Research Practice. Industrial Marketing Management, 39(1), 109-117.
Poole, P., Gray, B., & Gioia, D. (1990). Organizational Script Development Through Interactive
Accommodation. Group and Organization Studies, 15(2), 212-232.
Porac, J. & Thomas, H. (1990). Taxonomic Mental Models in Competitor Definition. The Academy of
Management Review, 15(2), 224-240.
Porac, J., Thomas, H., & Baden-Fuller, C. (1989). Competitive Groups as Cognitive Communities: The
Case of the Scottish Knitwear Manufacturers. Journal of Management Studies, 26(4), 397-416.
Pratt, M. (2009). For the Lack of a Boilerplate: Tips on Writing Up (and Reviewing) Qualitative
Research. Academy of Management Journal, 52(5), 856-62.
Purchase, S. & Olaru, D. (2004). Substance in Business-to-Business Relationships. Journal of Businessto-Business Marketing, 11(3), 23-52.
Purchase, S., Lowe, S., & Ellis, N. (2010), From "Taking" Network Pictures to "Making" Network
Pictures: A New Metaphorical Manifesto for Industrial Marketing Research, Journal of
Organizational Change Management, 23(5), 595-615.
Ramos, C. (2008). Developing Network Pictures As A Research Tool: Capturing The Output Of
Individuals Sense-Making In Organisational Networks. Unpublished PhD Thesis, University of
Bath, School of Management.
Ramos, C. & Ford, D. (2011). Network Pictures as a Research Device: Developing a Tool to Capture
Actors Perceptions in Organizational Networks. Industrial Marketing Management, 40(3), 447464.
Ramos, C., Naudé, P., & Ford, D. (2005). Developing Network Pictures as a Conceptual Device. 1st IMP
Meeting of the IMP Journal. Oslo, Norway.
Ranson, S., Hinings, B., & Greenwood, R. (1980). The Structuring of Organizational Structures.
Administrative Science Quarterly, 2(5), 1-17.
IMM Manuscript 10-562RR
Reger, R. & Palmer, T. (1996). Managerial Categorization of Competitors: Using Old Maps to Navigate
New Environments. Organization Science, 7(1), 22-39.
Rentch, J., Heffner, T., & Duffy, L. (1994). What you Know is What You Get from Experience: Team
Experience Related to Teamwork Schemas. Group and Organization Management 19(4), 450474.
Rescher, N. (1998). Complexity: A Philosophical Overview. New Brunswick: N.J.: Transaction.
Ritchie, J. & Lewis, J. (2003). Qualitative Research Practice: A Guide for Social Sciences Students and
Researchers. London: Sage Publications.
Ritter, T. (1999). The Networking Company: Antecedents for Coping with Relationships and Networks
Effectively. Industrial Marketing Management, 28(5), 467-479.
Ritter, T., Wilkinson, I., & Johnston, W. (2004). Managing in Complex Business Networks. Industrial
Marketing Management, 33(3), 175-183.
Rohrmus, D. & Henneberg, S. (2006). Sense-Making and Cognition in Business Networks:
Conceptualisation and Propositional Development. Second IMP Journal Seminar. Chalmers
University, Gothenburg, Sweden.
Salmi, A. (1999). Russian Networks in Transition: Implications for Managers. Industrial Marketing
Management, 25(1), 37-45.
Samli, A. & Bahn, K. (1992). The Market Phenomenon: an Alternative Theory and Some Metatheoretical
Research Considerations. Journal of the Academy of Marketing Science, 20(2), 143-153.
Sandelands, L. & Stablein, R. (1987). The Concept of Organizational Mind. In C. Bacharach & N.
DiTomaso (eds.), Research in the Sociology of Organizations (pp. 135-161). Vol. 4; Greenwich,
Connecticut: JAI Press.
Saner, R. & Yiu, L. (2005). Swiss Executives as Business Diplomats in the New Europe: Evidence from
Swiss Pharmaceutical and Agro-Industrial Global Companies. Organizational Dynamics, 34(3),
298–312.
Schwandt, T. (2000). Constructivist, Interpretivist Approaches of Human Enquiry. In N. Denzin & Y.
Lincoln (eds.), Handbook of Qualitative Research (pp. 118-137). Thousand Oaks: Sage.
Scott, J. (2000). Social Network Analysis: A Handbook. London: Sage Publications.
Simon, H. (1999). The Sciences of the Artificial (3rd ed.). Cambridge: Mass.: The MIT Press.
Smithee, A. & Lee, T. (2004). Future Directions in Marketing Knowledge”, Journal of Business and
Industrial Marketing, 19(2), 149-54.
Spender, J.-C. (1989). Industry Recipes: An Enquiry into the Nature and Sources of Managerial
Judgement . Cambridge: MA: Basil-Blackwood, Inc.
Stacey, R. (1996). Complexity and Creativity in Organizations. San Francisco: Berrett-Koehler
Publishers.
Starbuck, W. & Milliken, F. (1988). Executives Perceptual Filters: What they Notice and How They
Make Sense. In D. Hambrick (ed.), The Executive Effect: Concepts and Methods for Studying Top
Managers (pp. 35-65). Greenwish, Connecticut: JAI Press.
Stewart, D. & Latham, D. (1980). A Preference Mapping of Organizational Objectives of Sports
Franchise Executives. Journal of Applied Psychology, 65(5), 610-615.
Streufert, S. & Nogami, G. (1989). Cognitive Style and Complexity: Implications for I/O Psychology. In
C. Cooper & I. Robinson (eds.), International Review of Industrial and Organizational
Psychology (pp. 93-143). London: Wiley.
Streufert, S. & Swezey, R. (1986). Complexity, Managers, and Organizations. London: Academic Press.
Stubbart, C. & Ramaprasad, A. (1988). Probing Two Chief Executives Beliefs about the Steel Industry
using Cognitive Maps. In Lamb, R. and Srivastava, P. (ed.), Advances in Strategic Management
(p. 139-164). Vol. 5; Greenwich, Connecticut: JAI.
Tetlock, P. (1984). Cognitive Style and Political Belief Systems in the British House of Commons.
Journal of Personality and Social Psychology, 48(2), 1227-1239.
IMM Manuscript 10-562RR
Thomas, J., Clark, S., & Gioia, D. (1993). Strategic Sensemaking and Organizational Performance:
Linkages among Scanning, Interpretation, Action, and Outcomes. Academy of Management
Journal, 36(2), 239-270.
Thompson, J. (1967). Organizations in Action, Social Science Bases of Administrative Theory: McGrawHill Book Company.
Tsoukas, H. & Hatch, J. (2001). Complex Thinking, Complex Practice: The Case for a Narrative
Approach to Organizational Complexity. Human Relations, 54(8), 979-1013.
Tufte, E.R. (1983). The Visual Display of Quantitative Information. Cheshire, Ct.: Graphics Press.
Turnbull, P., Ford, D., & Cunningham. M. (1996). Interaction, Relationships and Networks in Business
Markets: An Evolving Perspective. Journal of Business & Industrial Marketing, 11(3/4), 44-62.
Uzzi, B. (1996). The Sources and Consequences of Embeddedness for the Economic Performance of
Organizations: The Network Effect. American Sociological Review, 61(4), 674-689.
Vandenberg, R. & Scarpello, V. (1990). The Matching Model: An Examination of the Processes
Underlying Realistic Job Previews. Journal of Applied Psychology, 75(1), 60-67.
Waddington, C. (1977). Tools for Thought: How to Understand and Apply the Latest Scientific
Techniques of Problem Solving. New York: Basic Books.
Wagner, R. (1987). Tacit Knowledge in Everyday Intelligent Behavior. Journal of Personality and Social
Psychology, 52(6), 1236-1247.
Walsh, J. (1988). Selectivity and Selective Perception: An Investigation of Managers Belief Structures
and Information Processing. Academy of Management Journal, 31(4), 873-896.
Walsh, J. (1995). Managerial and Organizational Cognition: Notes from a Trip Down Memory Lane.
Organization Science, 6(3), 280-321.
Walsh, J., Henderson, C., & Deighton, J. (1988). Negotiated Belief Structures and Decision Performance:
an Empirical Investigation. Organizational Behavior and Human Decision Processes, 42(2), 194216.
Walsh, J. & Ungson, G. (1991). Organizational Memory. Academy of Management Review, 16(1), 57-91.
Weick, K. (1979). The Social Psychology of Organizations (2nd ed.). New York: Random House, Inc.
Weick, K. (1995). Sensemaking in Organizations, Thousand Oaks, CA: Sage Publications, Inc.
Welch, C. & Wilkinson, I. (2002). Idea Logics and Network Theory in Business Marketing. Journal of
Business-to-Business Marketing, 9(3), 27-48.
Wilkinson, I. & Young, L. (2002). On Cooperating Firms, Relations and Networks. Journal of Business
Research, 55(2), 123-32.
Williamson, O. (1975). Markets and Hierarchies. New York: The Free Press.
Woodside, A. G. (2010). Bridging the Chasm between Survey and Case Study Research: Research
Methods for Achieving Generalization, Accuracy, and Complexity, Industrial Marketing
Management, 39(1), 64-75.
Yin, R. (1994). Case Study Research: Design and Methods (2nd ed.). Applied Social Research Methods
Series, 5: Sage Publications.
Yiu, L. & Saner, R. (2000). Determining the impact of cognitive styles on the effectiveness of global
managers: Propositions for further research. Human Resource Development Quarterly, 11(2), 31924.
IMM Manuscript 10-562RR
IMM Manuscript 10-562RR
Appendix 1
Construct Description and complexity scales underlying the sub-dimensions of network picture complexity
SUB-DIMENSIONS
Sub-Dimension 1.1
Number of Actors
Sub-dimension 1.1.1
Number of actors
Sub-dimension 1.1.2
Number of groups
of actors
Sub-Dimension 1.2
Nature of Actors
Sub-dimension 1.2.1
Business Counterparts
Sub-dimension 1.2.2
Non-Business Counterparts
Sub-Dimension 2.1 Number
of Relationships
Sub-dimension 2.1.1
Number of relationships
Sub-dimension 2.1.2
Number of groups of
relationships
Construct Description
Nature: Large or small number of specific actors
Definition: Specific actors are those that are identified by a respondent by their names (e.g.
name of a person, of a company, of an association) (Cova and Salle, 2008).
Rationale: The greater the number of specific actors identified, the more complex is the
respondent’s network picture.
Nature: Large or small number of groups of actors
Definition: Groups of actors are those that are described by a respondent as a category of
actors (e.g. clients, suppliers, competitors, business sectors) (Cova and Salle, 2008).
Rationale: The greater the number of groups of actors identified, the more complex is the
respondent’s network picture.
Nature: Large or small number of business counterparts
Definition: Business counterparts are all entities with which a respondent has business
relations with (e.g. clients, suppliers) (Medlin and Tornroos, 2007).
Rationale: The greater the number of business counterparts identified, the more complex is
the respondent’s network picture.
Nature: Large or small number of non-business counterparts
Definition: Non-business counterparts are all entities that somehow affect or are affected by
a respondent’s business activity, despite the nonexistence of business relationships (e.g.
government, local or global community, competitors, industry associations) (Saner and Yiu,
2005).
Rationale: The greater the number of non-business counterparts identified, the more
complex is the respondent’s network picture.
Nature: Large or small number of relationships between specific actors
Definition: Relationships between specific actors are all relationships between actors that
are identified by a respondent by their names (Cova and Salle, 2008).
Rationale: The greater the number of relationships between specific actors identified, the
more complex is the respondent’s network picture.
Nature: Large or small number of relationships between groups of actors
Definition: Relationships between groups of actors are all relationships between actors that
are identified by a respondent as a category or group (Cova and Salle, 2008).
Rationale: The greater the number of relationships between groups of actors identified, the
Complexity Scales (5-point)

Value 1 (minimum complexity) when
the lowest number of actors (for Subdimension 1.1) / nature of actors (for
Sub-dimension 1.2) / number of
relationships (for Sub-dimension 2.1)
is identified in the sample

Value 5 (maximum complexity)
when the highest number of actors
(for Sub-dimension 1.1) / nature of
actors (for Sub-dimension 1.2) /
number of relationships (for Subdimension 2.1) is identified in the
sample

Values 2, 3 and 4 (intermediate
complexity) for situations in-between
the two described above, with the
score increasing with complexity.
IMM Manuscript 10-562RR
more complex is the respondent’s network picture.
Sub-Dimension 2.2 Nature
of Relationships
Sub-dimension 2.2.1
Variety in the strength of
relationships
Sub-dimension 2.2.2
Variety in the immediateness
of relationships
Sub-dimension 2.2.3
Variety in the formality of
relationships
Sub-dimension 2.2.4
Variety in the directionality
of relationships
Sub-dimension 2.2.5
Variety in the
multidisciplinarity of
relationships
Sub-dimension 2.2.6
Nature: Strong and/or weak ties.
Definitions: Weak ties are nodes in a wider social network that can be used to provide
resources, generate business and enhance reputations; Strong ties are ties that are developed
to serve a certain function, that are useful and that provide trustworthy information (Jack,
2005).
Rationale: The more variety of tie characteristics identified regarding strength, the more
complex is the respondent’s network picture.
Nature: Direct and/or indirect ties.
Definitions: Direct ties are the personal relationships between a decision maker and the
party about whom the decision is being made; Indirect ties are relationships between two
individuals, who are not directly connected but through whom a connection can be made
through the social network of each party’s direct ties (Granovetter, 1992; Gulati, 1995;
Larson, 1992).
Rationale: The more variety of tie characteristics identified regarding immediateness, the
more complex is the respondent’s network picture.
Nature: Formal and/or informal ties.
Definitions: Formal ties are ties that are legally established or defined by some hierarchy;
Informal ties are mostly social, person-centered and not organization-centered, not
contractually defined (Easton and Araujo, 1994; Granovetter, 1985; Halinen and Salmi,
2001; Marschan et al., 1996).
Rationale: The more variety of tie characteristics identified regarding formality, the more
complex is the respondent’s network picture.
Nature: Unidirectional and/or bidirectional ties.
Definitions: Unidirectional ties are associated with flows and influence activities between
actors with one sole direction; Bidirectional ties are about reciprocal flows and influences
between actors (Ford and Håkansson, 2006b; Henneberg et al., 2006),
Rationale: The larger the number of bidirectional ties identified, the more complex is the
respondent’s network picture.
Nature: Social and/or economic and/or technological ties.
Definitions: Social ties regard the atmosphere of the relationship; Economic ties are related
to the financial aspects of the relationship; Technological ties are about the
industrial/technical aspects that bound actors in a relationship (Easton, 1995; Halinen and
Tornroos, 2005).
Rationale: The larger the number of multidisciplinary aspects identified, the more complex
is the respondent’s network picture.
Nature: Actor bonds and/or resource ties and/or activity links.





Value 1 (minimum complexity)
when, (respectively for each subdimension), only one aspect that
allows characterizing the links
between actors is identified by the
respondent, e.g. only strong ties and
not weak ties;
Value 5 (maximum complexity)
when, (respectively for each subdimension), both strong and weak,
direct and indirect, formal and
informal, and unidirectional and
bidirectional
ties
are
comprehensively perceived by the
respondent;
Values 2, 3 and 4 (intermediate
complexity) for situations in-between
the two described above, with the
score increasing with complexity.
Value 1 (minimum complexity) when
none of the possible aspects of
relationships included in these subdimensions are mentioned;
Value 5 (maximum complexity)
when all of the possible aspects of
relationships included in these subdimensions
are
mentioned
IMM Manuscript 10-562RR
Variety in the substance of
relationships
Sub-dimension 2.2.7
Nature of the system of
relationships
Sub-Dimension 3.1
Changes
Sub-dimension 3.1.1
Changes in the
number/nature of actors
Sub-dimension 3.1.2
Changes in the
number/nature of
relationships
Sub-dimension 3.1.3
Changes in the
past/present/future
Definitions: Actor bonds reflect the atmosphere of a relationship; Activity links correspond
to the transfer or a transformation of resources between actors; Resource ties translate how
resources can be transferred or transformed in the network (Håkansson and Johanson,
1992).
Rationale: A respondent’s network picture covering relational aspects in terms of actors
bonds, resource ties, and activity links is assumed to be more complex than one that does
not include these three aspects.
Nature: Tightly coupled and/or loosely coupled systems.
Definitions: Tightly coupled systems are associated with linear systems, where the order of
sequences is invariant therefore not being place for flexibility or unpredictability; Loosely
coupled systems are associated with complex (chaotic) systems, where there is
unpredictability of events and associated consequences, and the order of sequences may be
changes without this affecting the system’s overall functioning (Perrow, 1984).
Rationale: A respondent’s network picture that can be described, as resembling a loosely
coupled system is considered more complex than one that is closer to a tightly coupled
system.
Nature: Large or small number of changes in the number/nature of actors
Definition: Changes in the number/nature of actors refer to all the changes that have
occurred on specific actors or groups of actors, whether business or non-business
counterparts (e.g. describing a restructuring that a client has gone through) (Halinen and
Tornroos, 1995; Medlin and Tornroos, 2007).
Rationale: The larger the number of changes in the number/nature of actors identified, the
more complex is the respondent’s network picture.
Nature: Large or small number of changes in the number/nature of relationships
Definition: Changes in the number/nature of relationships refer to all the changes that have
occurred on relationships between specific actors or between groups of actors (e.g.
describing how new relationships were created due to a change in legislation).
Rationale: The larger the number of changes in the number/nature of relationships
identified, the more complex is the respondent’s network picture (Hedaa and Tornroos,
2002).
Nature: Past and/or present and /or future
Definition: Changes in the past/present/future are about the period of time when the
changes that have been identified, referring to the number/nature of either actors or
relationships (Medlin and Tornroos, 2007).
Rationale: A respondent’s network picture covering changes that have taken place in more
periods of time (i.e. past, present and/or future) is assumed to be more complex than one
that includes fewer periods.







extensively;
Values 2, 3 and 4 (intermediate
complexity) for situations in-between
the two described above, with the
score increasing with complexity.
Value 1 (minimum complexity) when
a purely tightly coupled system is
described;
Value 5 (maximum complexity)
when a purely loosely coupled
system is described.
Values 2, 3 and 4 (intermediate
complexity) for situations in-between
the two described above, with the
score increasing with complexity.
Value 1 (minimum complexity) when
none of the possible aspects of
change included in these subdimensions are mentioned;
Value 5 (maximum complexity)
when most of the possible aspects of
change included in these subdimensions
are
mentioned
extensively;
Values 2, 3 and 4 (intermediate
complexity) for situations in-between
the two described above, with the
score increasing with complexity.
IMM Manuscript 10-562RR
Appendix 2
Illustration of a high complexity network picture
(Network picture complexity for a General Director in a company included in the project based network)
The visual representation
The textual description – quantification and illustrative quotes
IMM Manuscript 10-562RR
DIMENSIONS
COMPLEXITY
1.1.1. Number of actors
1.1.2. Number of groups
1.2.1. Business counterparts
1.2.12. Non-Business count.
AVERAGE
2.1.1. Number of Rel.
2.1.2. Number of groups Rel.
2.2.1. Strong and/or Weak ties
2.2.2. Direct and/or indirect
ties
.3. Formal and/or informal ties
2.2.4. Unidirectional and/or
bidirectional
2.2.5. Social and/or economic
and/or political and /or
technological ties
Quantification
Number of actors identified by this respondent: 20 (*)
Number of groups of actors identified by this respondent 28 (*)
Number of business counterparts identified by this respondent: 19 (*)
Number of non-business counterparts identified by this respondent: 29 (*)
Number of relationships identified by this respondent: 1 (*)
Number of groups of relationships identified by this respondent: 16 (*)
Quotes to Illustrate
Strong Ties - "This “owner of the building site” has an adjudicated which is a consortium called Chestnut. Here, a main relationship is originated but this
main relationship has several lateral tentacles."; "Regarding the consortium, and according to what I know, we have a group that is responsible for the civil
construction and engineering, a key aspect of the project."
Weak Ties - "...whoever is within the sector is able to understand the government political option; they know that this project is a measure to bring the
economy back to life."; "All the “game” with the external public and official entities has to be done by the sole client Oak; with our help, we are there, but
we can’t be getting in touch with all the entities… and it is also a way of getting efficiency..."
Direct Ties - "And therefore, Oak introduced this component of “urban insertion”, and thus the Councils began having an important role for in the
conception, construction and operations"; "For us, the main entity is the so called Oak, which has a contractual name in the legislation of “owner of the
building site.”
Indirect Ties - "...there is a series of little details that Oak and the inspection have to coordinate with us, since we only know and talk to Oak."; "...and some
sub-contracted other companies. For example, Hazel and Palm Tree worked a lot with big local companies."
Formal Ties - "This is all predicted in a thing called Consortium Agreement, which is an agreement where tasks division is defined..."; "It is completely
central; here, nothing is done within the project without Chestnut Consortium. Everything has the leadership of Chestnut, the entire project. "
Informal Ties - "There are informal contacts here, from the other companies. These general directors which are with me at Chestnut have a daily contact
with me..."; "I receive indications that are reciprocal, inputs from and to Palm Tree headquarters, (it is not here) inputs concerning their problems, [...] we
discuss and decide which strategies will be proposed. This takes place on an informal level."
Unidirectional Ties - "I have no doubt that the sole client Oak is the most important entity in this setting: it is the owner of the building site, the one paying
and the one giving us the money and everything."; "For example, the entity that does the homologation of the system: once everything is ready [...] to know
if it is ready to operate in terms of security and others, there is an official national entity that is called INTF that has the ways to do the homologation."
Bidirectional Ties - "Suggestions for implementing changes appear from several sides."; "But at my level I have a relation with the general directors [...]
two weeks meeting when I meet with every director in a consortium council; it is a relationship that has reciprocity and the same level of intensity and
quality."
Social Ties - "I feel that people, according to their profiles, and according to being positive or negative (the posture they assume towards life), react
differently and that has an impact in our relationships"; "that I would always defend the civil and engineering group, but what happens is that nowadays
these gentlemen have a an excellent relationship with me and total trust."
Economic Ties - "The commercial exploration with the public demands a maintenance of the network and this is carried out by an operator called Aspen.";
"who ever is within the sector is able to understand the government’s political options because everyone knows that this project is a measure to bring the
economy back to life. "
Political Ties - "...and the government understood that they can’t send this kind of work without studying the subject very well, to eliminate all doubts, to
define where it is going through [...] and other technical specifications that have to be pre-defined. "; "This building project was a huge “rehearsals
balloon” for every legislation that came out over the last 10 years, concerning security, environment and quality..."
Scale
(*)
3
4
2
5
3.5
1
3
3
3
3
3
4
IMM Manuscript 10-562RR
2.2.6. Actor Bonds and/or
Resource Ties and /or Activity
Links
2.2.7. Tightly coupled systems
and/or Loosely coupled
systems
Technological Ties - "Nowadays, there are very interesting security systems with very advanced software which is connected to the tram..."; "But this lead
to a series of technical changes, from the electric feeding, and then there would be two trams on the same line requiring electric feeding, signaling, etc..."
Actor Bonds - "If it wasn’t for this team spirit and mutual helping, and the problems exist and many times I have to moderate issues between the companies
(because they exist and what be weird would be if they did not exist) but if it wasn’t for this team spirit, this construction wouldn’t be done."; "The civil
construction has an excellent relationship with me, they trust me completely, it’s like the other groups, but there is one thing: sometimes, they are more
demanding because since they know that I am from the civil and engineering group, they feel more comfortable to be more demanding in what concerns
some things..."
Resource Ties - "Then we have the electro-mechanics which place the electrical feeding, the sub-stations, all electro-mechanic systems (all systems of radio
and voice, information to the public) which are done by two companies: Elm (essentially this one) and then there is also Eucalyptus, who does the signaling
part...”; "But this lead to a series of technical changes, from the electric feeding, and then there would be two trams on the same line needing electrical
feeding, signaling, etc and therefore, several adaptations had to be done to solve an issue that seems very easy but that is reality, it is not"
Activity Links - "That is to say, the building site has a big extension and it was possible to divide it by the two companies: the 10 stations were divided
among the 2 companies, 5 for each. They worked as sub-contracted companies of Consortium Willow doing their work."; "And how was this done? There
have always been several weekly meetings (forums) of the civil group with inspection and technical support; the electrical-mechanical group, from the
people from the trams, from the people of signaling, and therefore specific meetings of this issues with Oak."
Tightly Coupled system - “There are several external entities that have conditioned out daily life. For example, the entity that homologates the system: as I
have mentioned before, once everything is ready, although it has been already inspected by the civil construction, it has to go through INTF - an official
entity that has specific ways to homologate”.
Loosely Coupled System - “Therefore, it is a chain reaction that can arise from several sides."; “…and also due to the lack of clear definitions from the
Councils and from the government, also a consequence of these two wanting different things, Oak faced some hesitations and took some “back and
forward” steps, result of the inexistence of experienced structures on these issues.”
AVERAGE
3.1.1. In the number and/or
nature of actors
3.1.2. In the number and/or
nature of relationships
4
2
2.889
Quotes to Illustrate
Change in the number of actors - "This entity will disappear with the end of the adjudication. Consortium Chestnut got this adjudication and this contract
at the end of the 3 years of operations, finishes, independently from the new lines that these companies can run for."' ;"... and Dove Tree, a company that
nowadays almost does not have representation: it has 5% and it is no longer here in the project."
Change in the nature of actors- "Hazel is betting a lot in the internationalization"; "We are also investing in Cape Verde and Angola, and we have now
projects in Ireland and Greece but in fact, South America with Brazil for Palm Tree and the other countries for Cypress, is the predominant trend. This is
what the companies are trying to do so that they come overcome the space of time that exists until a new perspective arises."
Change in the number of relationships- "The foreign companies, like for example Elm, do not want to demobilize completely"; "...there was a new theme
very important and new and that were not predicted in the project and that were the “urban insertions". This led to an active participation of several City
Councils in the project’s design and implementation."
5
5
Change in the nature of relationships - "At the beginning, there was a leadership by Hazel [...] Afterwards, when Palm Tree takes the leadership, there
was a need to re-organize…”; "And it is predicted that as soon as the client finishes, every one leaves and Consortium Chestnut disappears. We do not want
to continue as Chestnut. Even if there are extensions of the work concerning Oak, we do not want Chestnut. We will want to create a new consortium."
3.1.3. In the past and/or
present and/or future
Change in the past - "For example, Palm Tree and Hazel used to work a lot with big local companies..."; "Not Eucalyptus since it had its own personnel
for signaling and the trams were built in the factory that was shut down in this country."
Change in the present - "Aspen is developing a huge national marketing campaign"; "And what is happening to that capacity: the small companies are
going bankrupt since they are always dependent on the sub-contracting for these big building sites and therefore, they are the first ones to shake and to go
down; and then the big companies also shake since they have big structures and therefore they begin having alternative programs"
4
IMM Manuscript 10-562RR
Change in the future - "With the new adjudication that it had now from Oak, it will supply some more trams but obviously they will have to build it
somewhere else since the plant in this country was shut down, but it does not matter."; "But that is the fact and these projects are a hope for the sector in the
sense of having once again the volume of work to occupy the installed capacity. For this period of 3 or 4 years, it is necessary to come up with solutions."
AVERAGE
4.667
FINAL AVERAGE
3.685
(*) To score these sub-dimensions, the minimum value of the scale (i.e. 1) was given to the lowest number of actors/groups of actors/business
counterparts/non-business counterparts/ relationships/number of relationships identified in the forty-seven network pictures. The distribution of complexity
on a scale of 1 to 5 is presented in the following table:
Sub-dimensions
1.1.1. Number of actors
1 (Minimum) to 47 (Maximum) actors
1
1 to 10
VALUE FOR THE 1-5 SCALE
2
3
4
11 to 19
20 to 29
30 to 39
1.1.2. Number of groups of actors
0 (Minimum) to 42(Maximum) groups of actors
0 to 8
9 to 16
17 to 25
26 to 33
34 to 42
1.2.1. Business counterparts
3 (Minimum) to 55 (Maximum) business counterparts
3 to 12
13 to 22
23 to 33
34 to 43
44 to 55
1.2.12. Non-Business counterparts
0 (Minimum) to 33 (Maximum) non-business counterparts
0 to 6
7 to 13
14 to 19
20 to 26
27 to 33
2.1.1. Number of relationships
0 (Minimum) to 16 (Maximum) number of relationships
0 to 3
4 to 6
7 to 10
11 to 13
14 to 16
2.1.2. Number of groups of relationships
0 (Minimum) to 36 (Maximum) number of groups of relationships
0 to 7
8 to 14
15 to 21
22 to 29
30 to 36
Minimum and Maximum Numbers
5
40 to 47
IMM Manuscript 10-562RR
Appendix 3
Illustration a low complexity network pictures
(Network picture complexity for a Production Director at a company included in the non-project based network)
The visual representation
IMM Manuscript 10-562RR
The textual description – quantification and illustrative quotes
DIMENSIONS
COMPLEXITY
1.1.1. Number of actors
1.1.2. Number of groups
1.2.1. Business counterparts
1.2.12. Non-Business count.
AVERAGE
2.1.1. Number of Rel.
2.1.2. Number of groups Rel.
2.2.1. Strong and/or Weak ties
2.2.2. Direct and/or indirect
ties
Quantification
Number of actors identified by this respondent: 3 (*)
Number of groups of actors identified by this respondent 2 (*)
Number of business counterparts identified by this respondent: 5 (*)
Number of non-business counterparts identified by this respondent: 0 (*)
Number of relationships identified by this respondent: 0 (*)
Number of groups of relationships identified by this respondent: 2 (*)
Quotes to Illustrate
Strong Ties – “We manage it conjointly, together with the in-house supplier. (Question: And this Alpha unit is only producing for your company or is it also
producing for other external companies?) No, it is producing solely for us. We have a very big producing capacity and therefore we need all their
production.”
Weak Ties – “I would have to go and check. I think it is Epsilon but I’m not sure… but we usually have two suppliers for each… although usually we use
one of them more frequently, we always have another one. But right now, for labels, I can’t remember. And for the lids… also don’t remember.”
Direct Ties – “Sigma works with several staff and suppliers, with a common aim which relies in satisfying the customers, the final customers and the middle
customers.”; “…it’s the people that I work with on a daily basis; there are some suppliers but with whom I have less contact.”
Indirect Ties – No illustration.
Scale
(*)
1
1
1
1
1
1
1
1
1
2.2.3. Formal and/or informal
ties
Formal Ties – “I can only speak about the production part since my relation is more directly with Alpha’s production area.”; “They can’t be on a short
term since they do a huge investment, and there is the machinery, the people, it could never be based on an annual contract.”
Informal Ties -
1
2.2.4. Unidirectional and/or
bidirectional
Unidirectional Ties – No illustration.
1
Bidirectional Ties – “We manage it conjointly, together with the in-house supplier.”
2.2.5. Social and/or economic
and/or political
and /or technological ties
2.2.6. Actor Bonds and/or
Resource Ties and /
or Activity Links
Social Ties – No illustration.
1
Economic Ties – “Potentially, I can talk about projects, concerning decrease of costs, modifications in the bottles or something like that.”
Political Ties – No illustration.
Technological Ties – “They have pre-forms, in which the tip of the final bottle is exactly the same as the tip of the pre-form; those pre-forms do not have a
shape, they have a certain weight (43 grams) and there is a machine with the plastic mold.”
Actor Bonds - “The relationship is very good. It’s good, very good. We are very close.”; “We are the ones in the leadership. We demand more from our
suppliers.”; “… and we always try to exceed the clients’ expectations.”
Resource Ties - "They can’t be on a short term since they do a huge investment, and there is the machinery, the people, it could never be based on an
annual contract.”; “The investment was done by Alpha. They installed here a production unit, we gave the space and they gave the resources.”
Activity Links - "The pre-form will be previously heated so that is can be shaped. And through a pressure process, it enters in the form and by pressure, the
pre-form gets the desired form. We then do the filling of the bottles. Then we have a storing room where they store the bottles and we then use the bottles
from that storage. We manage it conjointly, together with the in-house supplier.”; “They have a small production unit inside our facilities, they produce the
bottles here and we do the fillings with them; on the production process, on that stance I work directly with them.”
2
IMM Manuscript 10-562RR
2.2.7. Tightly coupled systems
Tightly Coupled system - No illustration.
and/or Loosely coupled
systems
AVERAGE
Loosely Coupled System - No illustration.
3.1.1. In the number and/or
nature of actors
3.1.2. In the number and/or
nature of relationships
1
1.111
Quotes to Illustrate
Change in the number of actors - No illustration.
1
Change in the nature of actors- No illustration.
Change in the number of relationships- No illustration.
1
Change in the nature of relationships - No illustration.
3.1.3. In the past and/or
present and/or future
Change in the past - No illustration.
Change in the present – No illustration.
Change in the future - No illustration.
AVERAGE
FINAL AVERAGE
(*) See Appendix 2 for the scores given to the sub-dimensions.
1
1
1.037
IMM Manuscript 10-562RR
IMM Manuscript 10-562RR
Table 1
Dimensions and sub-dimensions of network picture complexity
Dimensions of
network pictures complexity
DIMENSION 1.
Number and nature of actors
DIMENSION 2.
Number and nature of
relationships
DIMENSION 3.
Dynamism and flexibility
Sub-dimensions of
network pictures complexity
1.1. NUMBER OF ACTORS:
1.1.1. Number of actors
1.1.2. Number of groups of actors
1.2. NATURE OF ACTORS:
1.2.1. Business counterparts
1.2.2. Non-business counterparts
2.1. NUMBER OF RELATIONSHIPS:
2.1.1. Number of relationships
2.1.2. Number of groups of relationships
2.2. NATURE OF RELATIONSHIPS:
2.2.1. Variety in the strength of relationships
2.2.2. Variety in the immediateness of relationships
2.2.3. Variety in the formality of relationships
2.2.4. Variety in the directionality of relationships
2.2.5. Variety in the multidisciplinarity of
relationships
2.2.6. Variety in the substance of relationships
2.2.7. Nature of the system of relationships
3.1 CHANGES
3.1.1. Changes in the number/nature of actors
3.1.2. Changes in the number/nature of relationships
3.1.3. Changes in the past/present/future
Implications
for complexity
(+)
(+)
(+)
(+)
(+)
(+)
(+)
(+)
(+)
(+)
(+)
(+)
(+)
(+)
(+)
(+)
Note - in the column “Implications for complexity”, the symbol (+) indicates that complexity varies in the same direction
as the data value for its corresponding sub-dimension. For example, sub-dimension 1.1.1 (number of actors), the larger the
number of actors included in an individual’s representation and description, the greater the underlying complexity. More
detailed information is available in Appendix 1.
IMM Manuscript 10-562RR
Table 2
Average values for overall and dimensional network picture complexity: Including respondents from both
networks (i.e. project-based network and non-project-based network)
1.
1.1. Experience in Company
1 – Limited Experience
2 – Intermediate Experience
3 - Vast Experience
Individual Characteristics
Averages
Dimension 1
Dimension 2
Overall
Number and
Number and
complexity nature of actors
nature of
relationships
2.22
1.97
2.31
2.15
1.69
2.30
2.54
2.15
2.58
Overall
complexity
1.2. Function Orientation
Averages
Dimension 1
Dimension 2
Number and
Number and
nature of actors
nature of
relationships
Dimension 3
Dynamism
and flexibility
2.37
2.46
2.89
Dimension 3
Dynamism
and flexibility
1 - Consistently Internally Oriented
2- Currently Internally / Previously
externally Oriented
2.09
1.59
2.21
2.47
2.12
1.64
2.24
2.50
3- Consistently Externally Oriented
4- Currently Externally / Previously
Internally Oriented
5Simultaneously
Internally
and
Externally Oriented
2.37
2.46
2.32
2.33
2.53
2.31
2.56
2.72
2.37
2.08
2.08
2.54
2.
Company Characteristics
Overall
complexity
2.1. Number of years in Business
1 – Short Period of Time
2 – Intermediate Period of Time
3 – Long Period of Time
2.2. Complexity of the
Structure
1 – Low Complexity
2 – Intermediate Complexity
3 – High Complexity
Internal
2.24
2.42
2.08
Overall
complexity
1.84
2.17
2.54
Averages
Dimension 1
Dimension 2
Number and
Number and
nature of actors
nature of
relationships
1.97
2.31
2.00
2.37
1.78
2.10
Averages
Dimension 1
Dimension 2
Number and
Number and
nature of actors
nature of
relationships
1.44
1.92
1.88
2.39
2.14
2.55
Dimension 3
Dynamism
and flexibility
2.46
2.89
2.37
Dimension 3
Dynamism
and flexibility
2.17
2.25
2.93
IMM Manuscript 10-562RR
Table 3
Average values for overall and dimensional network picture complexity: Including solely respondents from the
non-project-based network
1.
1.1. Experience in Company
1 – Limited Experience
2 – Intermediate Experience
3 - Vast Experience
Individual Characteristics
Averages
Dimension 1
Dimension 2
Overall
Number and
Number and
complexity
nature of
nature of
actors
relationships
2.17
1.99
2.19
2.17
1.75
2.17
2.42
1.81
2.44
Overall
complexity
1.2. Function Orientation
Averages
Dimension 1
Dimension 2
Number and
Number and
nature of
nature of
actors
relationships
Dimension 3
Dynamism
and flexibility
2.33
2.58
3.00
Dimension 3
Dynamism
and flexibility
1 - Consistently Internally Oriented
2- Currently Internal
/ Previously
externally Oriented
2.01
1.57
2.07
2.39
2.03
1.59
2.09
2.40
3- Consistently Externally Oriented
4- Currently External / Previously
Internally Oriented
5Simultaneously
Internally
and
Externally Oriented
2.61
2.75
2.33
2.75
2.59
2.60
2.38
2.80
2.1. Number of years in Business
1 – Short Period of Time
2 – Intermediate Period of Time
3 – Long Period of Time
2.2. Complexity of Internal Structure
1 – Low Complexity
2 – Intermediate Complexity
3 – High Complexity
2.29
1.96
2.38
2. Company Characteristics
Averages
Dimension 1
Dimension 2
Overall
Number and
Number and
complexity
nature of actors
nature of
relationships
2.24
1.97
2.31
2.42
2.00
2.37
2.08
1.77
2.10
Averages
Dimension 1
Dimension 2
Overall
Number and
Number and
complexity
nature of
nature of
actors
relationships
1.84
1.43
1.90
2.16
2.07
2.22
2.45
2.04
2.42
2.52
Dimension 3
Dynamism
and flexibility
2.46
2.89
2.36
Dimension 3
Dynamism
and flexibility
2.19
2.19
2.89
IMM Manuscript 10-562RR
Table 4
Average values for overall and dimensional network picture complexity: Including solely respondents from the
project-based network
1.
1.1. Experience in Company
1 – Limited Experience
2 – Intermediate Experience
3 - Vast Experience
Individual Characteristics
Averages
Dimension 1
Dimension 2
Overall
Number and
Number and
complexity
nature of
nature of
actors
relationships
2.53
1.87
3.06
2.13
1.62
2.43
2.58
2.27
2.62
Overall
complexity
1.2. Function Orientation
Averages
Dimension 1
Dimension 2
Number and
Number and
nature of
nature of
actors
relationships
Dimension 3
Dynamism
and flexibility
2.67
2.33
2.84
Dimension 3
Dynamism
and flexibility
1 - Consistently Internally Oriented
2- Currently Internal
/ Previously
externally Oriented
2.23
1.62
2.46
2.61
2.29
1.72
2.49
2.67
3- Consistently Externally Oriented
4- Currently External / Previously
Internally Oriented
5Simultaneously
Internally
and
Externally Oriented
1.88
1.87
2.28
1.50
2.48
2.10
2.68
2.67
2.21
2.56
2.
2.2. Number of years in Business
2.1. Complexity of Internal Structure
1 – Low Complexity
2 – Intermediate Complexity
3 – High Complexity
2.46
2.21
Company Characteristics
Not measurable in this network
Averages
Dimension 1
Dimension 2
Overall
Number and
Number and
complexity
nature of
nature of
actors
relationships
1.83
1.50
2.00
2.18
1.72
2.52
2.64
2.25
2.70
Dimension 3
Dynamism
and flexibility
2.00
2.30
2.97
IMM Manuscript 10-562RR
Table 5
Summary table on the outcome of exploratory proposition assessment
Propositions
Proposition 1
There exists an inverted-U shape relationship between network picture complexity
and the number of years of an individual’s working experience
Proposition 2
Network pictures are more complex for individuals with more experience in
externally-oriented functions (e.g. marketing) than in internally-oriented ones (e.g.
production)
Proposition 3
There exists an inverted-U shape relationship between individuals’ network picture
complexity and the number of years that his/her company existed or operated in a
specific network/sector”.
Proposition 4
Network pictures are more complex for individuals in companies which have a
complex internal structure
Outcome of
exploratory analysis
Not supported
Supported
Supported
Supported
IMM Manuscript 10-562RR
Figure 1
The triadic epistemological dialogue of network pictures
ne
tw
of
th
e
g
ur
in
rk
o
tw
ne
pi
ct
n
ow
Ac
t
ct
pi
or
s’
s’
er
ch
ar
se
Re
or
k
NETWORK
g
in
ur
MANAGER
RESEARCHER
Researchers’ picturing of actors’ network pictures