Industrial Marketing Management 39 (2010) 437–449

Contents lists available at ScienceDirect

Industrial Marketing Management

Mapping the values in B2B relationships: A systemic, knowledge-based perspective

J.H. Powell a,⁎, J. Swart b
a
Cardiff University Business School, Colum Drive, Cardiff, CF2 3SX, UK
b
School of Management, University of Bath, Bath, BA2 7AY, UK

a r t i c l e i n f o a b s t r a c t

Article history: The management of an enduring relationship between provider and supplier has at its heart an implicit
Received 1 March 2006 interaction between the valuation systems of the counterparts. We take the view that this interaction is
Received in revised form 1 May 2007 conveniently understood through the lens of knowledge management. Knowledge management informs
Accepted 1 November 2008
our treatment of business to business relationships through two mechanisms. It helps us manage better the
Available online 12 June 2009
dissemination and co-creation of knowledge in an organisation and new work in the mapping of knowledge
Keywords:
allows us to represent the knowledge aspects of a relationship in a way which allows us to manage it better.
Marketing We present, therefore, an approach to allow the specific representation of these valuation systems and their
Systems thinking interaction, using a case study of the marketing of a nuclear submarine programme to a government. A
Knowledge management conclusion is that structures which support the co-creation of knowledge between the companies is critical
Co-production to winning the contract and we indicate how this co-creation can be engendered.
© 2009 Elsevier Inc. All rights reserved.

1. Introduction and perspective
Marketing is a member of a large class of human activities which
seek to match the output of a provider with a predicted, perceived
demand. Business-to-business (from now on B2B) marketing has, in
particular, the characteristic that, with few exceptions, the provider
seeks to establish an enduring relationship with the buyer (Ford,
1997; Håkansson, Johanson, & Wootz, 1976; Turnbull & Valla, 1986).
Certainly it imbues more cooperation and collaboration than the more
general B2C case and hence we adhere to the term ‘provider’ as more
accurate here than ‘customer’ (Ritter & Ford, 2004). Marketing then
becomes very distinct from the act of selling (Gummesson, 2002),
since its focus of attention is on the satisfaction felt by the buyer (and
indeed by the provider) in continuing and enacting the relationship
(Vargo & Lusch, 2004).
Within this almost ubiquitous framing of B2B marketing as a
relationship issue, this paper assumes, specifically, the network under-
standing of marketing where the effects and implications of dyadic
company-to-company relationships (for example technology exchange)
occurs in a distributed, wider multi-agent structure characterised as a
network. These dyadic relationships between firms, then, expressed
within the wider network are characterised by being multi-agent,
episodic in nature, interactive, not standardised and distributed in nature
(Ford, Gadde, Håkansson, & Snehota, 2003 — pp 6 and 227–228). It
will be seen later that these characteristics place particular demands
upon the supporting tools and schemata used by managers of these
⁎ Corresponding author.
E-mail addresses: powelljh@cardiff.ac.uk (J.H. Powell), mnsjas@bath.ac.uk (J. Swart).
networks. It is the purpose of this paper to present a practical way of
representing, and hence aiding the understanding of the specific rela-
tionship of a firm with its encompassing network in order to satisfy these
particular demands.
The second assumption made in this paper is that the emergent
behaviour of the network as a whole (as a result of management
within it, particularly of the constituent inter-firm dyads), is a matter
of and the outcome of knowledge management (hereafter KM). This
does not imply that no other lens is valid; merely that such a per-
spective, whereby the relationship is conveniently and effectively
understood to be enacted and reified by the knowledge conveyed in its
discourse, is an important, unifying and productive one. When firms
interact in an attempt to reconcile, jointly and severally, the needs
implied by their valuation systems, knowledge is exchanged and,
indeed co-created about the other's valuations. Moreover, particularly
in a technological industry, the exchange of knowledge is the most
significant bearer of the commitment so central to the social network
model which informs our understanding of B2B network relationships
(Cook & Emerson, 1978).
The view taken here is that the act of marketing is, at heart, the
interaction between two valuation systems. We mean by this the set of
attributes emergent from the relationship which are valued by each
party and the nature of the mutual interaction of these attributes.
For example, one firm may value greatly the transfer or co-creation
of technical knowledge relevant to a developing product line while
the dyadic partner may value short term cash flow and reflected
reputation gained by working with a more strongly branded partner.
This is in contradistinction with a second, common interpretation of
‘value system’ as the totality of the inherent, underlying values of
the firm and its constituents. We make no claim to throw light on

0019-8501/$ – see front matter © 2009 Elsevier Inc. All rights reserved.
doi:10.1016/j.indmarman.2008.11.011
438 J.H. Powell, J. Swart / Industrial Marketing Management 39 (2010) 437–449
these moral and ethical underpinnings of the members of the firm;
here ‘valuation system’ is to be understood as the embodiment in the
moment of the permanent set of ‘things we stand for’ which are
underpinned by the ‘value system’ in this second sense (Glynn, 2007;
Stryker, 1980, 2000).
For example, one firm may, because of its commercial and market
circumstances, be motivated to place great and continuing value on its
skills and competence, its ability to create innovative technologies,
processes and products. The valuation system of such a firm would
then constitute a listing of those attributes valued together with some
representation of how they interact. For example, there are clear
connections between competencies, skills, training investment, ability
to innovate and differentiation in the market place. The firm needs to
know and, further, needs to be able to express the interconnections of
its valuation system, because it is the performance of this very system
which is the focus of the whole of the management of the firm, since
what are we managing if it is not that which we value (Burt, 1992)?
This idea of dyadic interaction being a matter of interaction of valua-
tion systems is inherent in much of the network literature (Brusoni,
Prencipe, & Pavitt, 2001).
After a brief discussion of the implications of this network and
knowledge perspective on management of B2B relations the paper
illustrates the use of a particular method, deriving from the systems
and KM fields and then, through the medium of a case study, the
appropriateness of the technique (known as System-based Knowl-
edge Management (Powell & Swart, 2006), or SBKM) is indicated.
We then make observations on the knowledge management issues
illuminated by the case study and in particular on the rôle of co-
creation mechanisms in creating network knowledge.
2. Managerial challenges
The nature of the relationships within the context of a network of
interactions is, as has been observed, that they are not to be under-
stood solely as dyadic interactions. They are episodic and dynamic
in nature, distributed, interactive, and not standardised (Ford et al.,
2003 — pp 227–228). These characteristics place special demands on
the management of the series of dyadic relationships which constitute
the marketing activity of the firm. The challenge of managers and
indeed of the approach presented here is to understand the respective
valuation systems and their interactions.
Firstly, because they are of a distributed and dynamic nature, the
sense-making of managers aspirant to conditioning (if not control-
ling) the network is particularly challenged (Ford & Redwood, 2005).
It is no simple matter even to understand the internal systems of a
firm (Simon, 1972; Tsoukas, 1996; Tsoukas & Vladimirou, 2001; von
Krogh, Roos, & Kleine, 1998a); expecting managers to understand and
empathise with the valuation systems of others in the network is an
additional demand, made essential by the recognition that what is
done in the dyadic system surrounding the firm will have referred
and changing effects elsewhere in the network. Action is dependent
upon knowing (Brusoni et al., 2001; Baumard, 1999; Cook & Seely
Brown, 1999; Davenport & Prusak, 1998; Polanyi, 1966; Swart &
Powell, 2006) and hence on sense-making (Tsoukas, 1996; Tsoukas &
Vladimirou, 2001; Weick, 1979). Managers of these networks, then,
seeking to influence the network as a whole through the limited
communication and action channels of their immediate dyadic set,
require a representation method which will allow them at least to
capture their misconceptions about others in the network (Kahaner,
1996). At best, of course, these misconceptions are challenged and
resolved through the recognition of mutual inconsistency by the
structuring which a system representation provides (Checkland &
Scholes, 1990; Coyle & Exelby, 2000; Davenport & Prusak, 1998; Eden,
1989; Flood & Jackson, 1991; Powell & Bradford, 1998).
Of particular importance is the observation that the nature of
the relationship of the firm with its surrounding network is both
systemic and (critically) non-standardised — it is specific to the firm,
its context, its contemporaneity. The implication of this upon the
support needed by managers of these networks, then, is that any
sense-making representation method has to speak to the specifics of
the firm. Generalised models are inadequate. There is an extensive
body of literature which deals with this type of specific dynamic
modelling (Checkland & Scholes, 1990; Coyle & Exelby, 2000; Flood &
Jackson, 1991; Coyle, 1977, 1998; Forrester, 1961; Richardson & Pugh,
1981; Sterman, 2000; Tustin, 1953; Wolstenholme, 1990) which is
able to represent the specifics of a firm's valuation systems and con-
text and upon which we draw in our subsequent representations.
Lastly, the assertion that the knowledge component of these net-
works is critical to their understanding and enactment leads us to
specify that whatever representation method is used to assist man-
agers in their action planning and sense-making, it has to have a
capacity to represent the knowledge inherent in the transactional
relationships of the dyads and hence the network (Powell & Bradford,
1998).
The purpose of this paper from this point onwards is to present a
practical way of representing the specific interactions between valuation
systems which complies with the implied requirements identified in
this section.
3. General model
If we are to address the managerial issues surrounding the nego-
tiated valuation systems as we propose it is highly desirable to have
some disciplined, preferably commonly held and auditable way of
representing the interaction (Checkland & Scholes, 1990; Coyle, 2000;
Powell & Bradford, 1998; Quinn, Mills, & Friesen, 1992; Wolstenholme,
1990). Fig. 1 indicates the general architecture which we should
expect to see in such a system model supporting action-directed
managerial sense-making and analysis.
One should expect to see each party's ‘business model’ interacting
with the mechanisms of the market (Håkansson & Ford, 2002). This
would be true even if the parties were non-communicating com-
petitors; there would still be a dialogue but one mediated only by
the mechanism of the market. In point of fact there will be other
mechanisms of intercommunication which are not directly driven by
the economic, short term issues of the market. These are referred to in
Fig. 1 as ‘indirect’ connections. Such indirect communications can be
quite explicit. For example, as we shall see in the case example cited
below, two high technology firms can and usually will co-operate over,
say, joint technology development or research agenda definition even
if the market as such does not recognise the direct benefit. An example
of such a market is aerospace and defense. Here states frequently
adopt policies either of arguing for a particular procurement on the
Fig. 1. Interaction of participants' valuation systems within market context.
 J.H. Powell, J. Swart / Industrial Marketing Management 39 (2010) 437–449
basis of Strategic Defense Industrial Base (SDIB), where the continuing
existence of a source of supply is deemed more important than value
for money in any particular procurement, or for myopic Value For
Money (VFM) procurement. In both cases the purchase is instru-
mental in the sense that procurement officers are driven to adopt rules
which stress the outcome of the industrial system rather than ac-
cepting and accommodating its behavioural characteristics. No value
judgement upon such an approach is made here, but the observation
is offered that such an instrumental market tends to strengthen the
relative importance of the indirect communications between nego-
tiating/co-operating firms since they become responsible jointly for
those aspects which they find strategically desirable but which are not
reflected in procurement policy.
The behavior of firms reflects precisely what should be expected
from the study and practice of KM. As firms engage in these indirect
communications (enacted, of course, through the interactions of
individuals and groups of individuals) KM predicts the appearance of
co-created explicit and tacit knowledge emerging from and supported
by social and organisational structures supporting the knowledge
generating mechanisms (Burt, 1992; Cook & Seely Brown, 1999;
Davenport & Prusak, 1998; Granovetter, 1973; Leana & van Beuren,
1999; Nonaka, 1994; Quinn et al., 1992; Scarbrough, Swan, & Preston,
1999; Swart & Kinnie, 2003; Swart, 2006; von Krogh et al., 1998b;
Weick, 1979). Thus we see in the case of high technology companies
(both for pure prime-contractor/supplier relationships and more
equipotent B2B relationships), the setting up of joint project teams
from which knowledge about complex projects are co-created.
Examples abound world-wide, and include JSF, Eurofighter, Common
New Generation NATO frigate and the example discussed later in this
paper, the nuclear submarine programmes of the United Kingdom. In
all these cases we see both the purposive generation of knowledge,
altering the explicit technical valuation systems of participants and
the non-purposive, deeply tacit co-experiencing of an emergent
requirement which inexorably, and in just as powerful a way as that of
the explicit mechanisms, alters the valuation of the participating
organisations (Tsoukas, 1996; Swart & Powell, 2006; Swart & Kinnie,
2003; Krogh et al., 1998b).
It will be seen, then, that a system model which purports to
represent the engagement between co-operating firms in an essential
dyadic relationship must also, reflect the third, silent party, namely
the surrounding network context (Powell & Bradford, 1998). In the
case of the nuclear submarine joint programme (to which we will
soon turn) we see a rich engagement by the provider of the nuclear
prime mover (the so-called nuclear steam raising plant or NSRP) and
the prime contractor who, conventionally, is the hull provider. At first
glance this looks like a one-to-one B2B relationship of conventional
form; the NSRP provider seeks to sell a plant to the prime contractor.
Fig. 1 illustrates that in fact the negotiated relationship is triadic in that
each party is, in addition to this straightforward one-to-one engage-
ment, experiencing interactions with the market, here, narrowly, the
monopsonic procurer of a strategic national defense asset (the nuclear
submarine) representing, unusually, the whole surrounding network
context. As a result the case represents a simplified, canonical example
of a B2B network of the form generalized in Fig. 1. The effect of the
network of suppliers is not ignored here, not least because there are of
the order of 3 million separate components in the product, but the
triad analyzed here is the dominant feature of the project landscape.
In addition to the overt triadic relationship driven by explicit
market demands we should expect our system model to show evi-
dence of co-creation of knowledge through the social mechanisms of
interaction among workers on the project and between them and
client experts (Morris, 2000; Reagans & Zuckerman, 2001; Swart &
Kinnie 2003). If we can generate a system model which exhibits these
features and which aligns with the characteristics of the interaction
indicated above we have a basis for identifying managerial action
which is system-based, respectful of the knowledge dimension of
439
analysis and aimed at managing the relationship between both sides
of the B2B relationship.
4. Research and analysis design
The toolset we choose to represent these interactive valuation
systems was developed particularly for the representation of knowl-
edge systems. It is known as Systems Based Knowledge Management
(SBKM) (Powell & Swart, 2006; Swart & Powell, 2006) and derives
from the qualitative variant of the ubiquitous Systems Dynamics
family of tools (Coyle, 1998, 2000; Eden, 1989; Sterman, 2000). As
such it is essentially directed towards the representation of the com-
plex valuation system and hence to action aimed at the conditioning
of the behaviour system in focus and at allowing the rehearsal of
system changes. Thus it is both descriptive and normative in its output
(Coyle, 1977, 1998; Richardson & Pugh, 1981). There are other
approaches. Cognitive mapping (Eden, 1989), for example, can be
used to represent interactive systems but suffers from the disadvan-
tage that, while it is flexible, it lacks the grammatical stability nec-
essary for any resulting analysis to be transparent and well-founded
(Coyle, 1998, 2000; Liddell & Powell, 2004; Powell & Coyle, 2005;
Sterman, 2000). Network pictures, to some extent, circumvent these
difficulties but lack the required specificity of system representation
(Ramos & Ford, 2006) The large body of numerically based modelling
methods is rejected also, on the argument that many of the important
variables present in interactive valuation systems deriving from
human activity systems are, at best, arbitrarily rendered by a simple
number (Coyle, 2000; Powell & Swart, 2006; Powell & Coyle, 2005);
one would find it difficult to justify attaching a simple number to the
concept of trust or reputation in a business network, for example.
In essence SBKM represents the interaction of the protagonists'
valuation systems through the following procedure…
• Generate a causal loop or influence diagram which captures the
causal relationships between descriptive variables of a system in
which the protagonists are acting to enhance their valuations.
• Identify which actors within the firm control the strength of
connection between each pair of variables. These actors are not
synonymous with those in the AAR model, but may well include
some or all of these.
• Identify the tacit and explicit knowledge used at each connection by
each actor in pursuit of her rôle in that part of the system.
• With reference to the objectives of the system client, identify what
actions are appropriate in order to manipulate the objectives of each
actor at each point in the system and hence to derive the knowledge
required for each actor's enactment within the system.
We illustrate its use in a simplified version of a real life case study
of the triadic relationship between a UK nuclear submarine builder, its
NSRP provider and the monopsonic market in which that specific B2B
relationship is enacted.
5. Case setting — co-operation in a major defense program
The case now described has been constructed from public sources
and from a series of conversations and interviews with principal
players, notably the Rolls Royce and Associates' Chief Engineer on the
program together with the Chief System Engineer. These interviews
were conducted over a period of two years from January 1990 to
December 1991, primarily in the context of the project team's sites,
being in Bath, (where the procurement authority was located) and
Barrow-in-Furness (where the design team was located). Data was
also included deriving from contemporary personal conversations
between one of the researchers and senior members of both the firms
involved and with the Director General in charge of the procurement
at the time. The most important contributor to the case and the one
440 J.H. Powell, J. Swart / Industrial Marketing Management 39 (2010) 437–449
whose viewpoint the case takes is that of a senior member of a third
company, BAe Systems, who were contemplating, at the time, entering
this strongly competitive market by offering their prime-contractor-
ship skills. Thus the informant was well-informed of the market
dynamics and characteristics of the business network but was, at the
same, not burdened by too great an asymmetry of knowledge among
the participants, as would have been the case had a senior member of
any of the three protagonists been the informant. As part of the joint
project team, the informant was seconded to the team by agreement
with the other parties. Thus the viewpoint of the analysis is from the
position of a member of the joint project team but not, conveniently,
from any of the viewpoints of the three explicit protagonists. The
model is constructed from contemporary private papers and notes,
using the SBKM grammar (Powell & Swart, 2006) post facto to re-
present their content. The viewpoints taken are, successively and
separately those of the informants capturing the valuation systems of
each protagonist. The overall system model then emerges from the
modelers' linking (in this case, the three) informants' views, checking
that linking variables (i.e. those used by two informants) are similarly
understood.
6. Background
In the early 1990s the procurement of nuclear submarines in the
UK was problematic. These military platforms are amongst the most
complex objects ever built by man, containing over 3 million separate
identifiable components. From an engineering point of view the sys-
tem integration problem is immensely difficult, with the most sig-
nificant challenges being the engineering of the noise environment of
the submarine. In summary terms the sonar of the submarine is
affected not just by its own inherent effectiveness and sensitivity but
also by the noise created by the hull moving through the water, often
at relatively high speed and by the noise generated by the systems
inside the boat, particularly that of the machinery of the prime mover
plant, known as the nuclear steam raising plant, or NSRP. It can be
imagined that there are other huge integration issues surrounding
avoiding interference of one system with another and even in achiev-
ing a workable spatial layout of a submarine.
These submarines cost hundreds of millions of pounds and togeth-
er with their operating costs command substantial fractions of the
defense budget even of a developed economy. Of the unit procurement
cost some 60% accrues to the pressure hull, some 15% to the NSRP and
the remainder to weapon systems and integration. The risk, however,
is not in proportion to the cost since, although at first glance one would
expect the vitally important pressure hull to contain substantial risk, it
does not relative to the weapons systems and, above all, the integration
of the systems into a working warship. This is because the pressure hull
even for a new design, will be little changed in critical respects from
previous ones. For example, both ends of the cigar-tube shaped
pressure hull might remain unchanged, so that that difficult part of
the stress design would not need to be revisited and the construction
arrangements would be essentially unaltered. The weapons system,
however, invariably deploys revolutionary new technologies and pres-
ents a disproportionately high fraction of the risk.
In the UK the only supplier of NSRP was (and remains) Rolls Royce
and Associates (RRA) who, under the terms of a technology transfer
agreement with the US government dating from the 1950s (and
known as ‘the 1954 Agreement’) are allowed to implement the design
principles of a specific type of NSRP particularly suitable for under-
water use. The type is known as a Pressurized Water Reactor or PWR. At
the time of the case the only hull manufacturer was Vickers Ship
Engineering Limited (VSEL). Other competent manufacturers in France
and the United States were effectively disbarred from any competition
because of SDIB considerations.
Thus, when a proposal to build a new class of submarine was
requested by the UK government, the two manufacturers of hull and
NSRP respectively came together in cooperation. Their relationship
was a complex one. VSEL was protective of its knowledge of pressure
hulls and of the integration of plant into those hulls. RRA was in-
different as such as to whether it held knowledge about the pressure
hull itself, but it was a substantive issue for the company whether it
could contribute strongly to design discussions about the integration
aspects and about the NSR itself. Were it to lose a place at that design
table, its ability to be seen to be a part of the risk reduction regime of
the project would decline and, ultimately, its very place as the sole UK
supplier of the relevant nuclear technology under the 1954 agreement
might decline also. RRA was, however, very definitely a supplier of
nuclear technology equipment, design knowledge and integration
know-how while VSEL was the main contractor, a position to which
RRA did not aspire. Others such as GEC and BAe did, however, and
ultimately GEC took over VSEL and BAe merged with GEC, leaving RRA
still in its monopoly design position.
7. Method
The influence diagrams on which the method is based consist of
networks of causal relationships between variables which informants
declare as important or relevant to describing their problem. It can be
thought of as a causal relationship diagram which connects descrip-
tors chosen such that knowledge of their behavior would constitute
valuable knowledge as viewed by the informant. The arrows capture a
causal relationship (as distinct from mere correlation) and the signs
attached to the arrows indicate whether that relationship is direct
or inverse. In the latter case as the variable at the tail of the arrow
increases, the one at the head decreases and vice versa. Such an
inverse relationship is indicated by a − sign associated with the arrow.
For clarity, arrows with negative signs are also shown dashed in
figures. A direct relationship (often indicated by a + sign, but here
suppressed) is where the variable at the tail and at the head rise and
fall together. Relationships are not necessarily linear and the strength
of correlation will vary, not least as the system is managed to produce
the effects desired.
The method used here derives from the powerful and widely-used
approach of system dynamics which, being an example of a system
approach is concerned with the overall behavior of an interconnected
set of components or factors. System dynamics (Sterman, 2000) in its
quantitative form is particularly well-suited to the representation and
study of those systems whose components can be represented in a
purely numerical way, but in the type of system exemplified by B2B
marketing, where the components or factors are less easily and less
satisfactorily represented in a purely numerical way, the qualitative
form of system dynamics (Coyle, 1998) is more appropriate.
Influence diagrams are a common feature of both quantitative and
qualitative system dynamics and are well-documented in the liter-
ature (Montibeller & Belton, 2008; Sterman, 2000). Diagrams are built
up step-by-step by eliciting from informants the causal relationships
between variables which serve to describe the system behavior. These
relationships are not necessarily linear, and indeed can be either
strong or weak in their effect. The method requires variables to be
well-defined (by the informants) and to be capable of expression on a
scale (although not necessarily to be measurable in practice). There
are well-established rules of praxis (Coyle, 1998; Sterman, 2000)]
which ensure structural integrity of the diagram, for example in en-
suring that a consistent level of analysis is used throughout. The main
analysis approach is to examine the influence diagram for resonant
structures, known as loops, which have the capacity, once activated, to
continue to produce managerial benefits. The emergent diagrams are
authenticated by the informants where possible and achieve a high
degree of contemporary valuation and hence authentication by these
informants.
Here, we interviewed experienced individuals from all three firms
involved in the procurement in question, working with them to create
 J.H. Powell, J. Swart / Industrial Marketing Management 39 (2010) 437–449
an influence diagram which each felt was appropriate. The diagrams
were then checked for consistency and validity of variables and re-
presented to the informants who then checked them for validity and
relevance.
8. Context of supply (see Fig. 2)
The procurement officers will be more willing to place a contract
(here necessarily to a home supplier) the greater the predicted value
for money (VFM). VFM can accrue from many sources but a sub-
stantial component of it in these types of contract derives from
effective risk management. If the procurement officers judge that the
risk remaining with the state (known in the UK as The Crown) can be
reduced, the effective VFM is increased. Thus we see a negative sign
attached to the causal relationship residual risk to crown → perceived
VFM. Separately the procurement officers will judge residual risk as
undesirable of itself and hence a high residual risk (if the contractor
cannot carry out the task or cannot bear a liability) will lead to a
reduced desire to procure at all.
We see in Fig. 2 how these primary issues are connected with the
competences of the potential contractors and in particular the overall
project capability, the ability of VSEL and RRA to bear their share of the
risk (ability to bear risk V and ability to bear risk R, respectively). We
see in Fig. 2 the first inklings of a KM perspective, too, since the
willingness of the procurement officers to grant contracts will affect
the track record of VSEL and RRA jointly and severally and will hence
increase their experience, knowledge and competence to carry out
these complex design and implementation tasks. The joint working
arrangements between the firms will prove critical. The working ar-
rangements followed a standard contemporary practice, with a joint,
co-located project team concentrating on overall project design and
management together with separate in-house working teams design-
ing and planning specific hull and NSRP issues.
In many respects the B2B marketing task of supplying a NSRP is a
trivial one in that RRA are a monopolistic supplier. The design of the
PWR is essentially a stable, proven one. We have a monopsonic buyer
(the UK government), a single main contractor (VSEL) and a monop-
olistic supplier of NSRP. It is neither ‘offer for sale’ nor ‘selling’ as
described in the earlier part of this paper. Because the replacement
submarine represents the embodiment of many years of experience
of in-fleet operation and because nuclear technology, like all high
technology contexts, is a changing one, there is a true convergence of
valuation system throughout the project life. The specification of the
NSRP, although necessarily based on the PWR basic design, has many
options which will affect the way in which NSRP, hull and weapons
Fig. 2. Connections with participant firms' competences. (V = VSEL, R = Rolls Royce
and Associates).
441
Fig. 3. VSEL's valuation system (s/m mfr = submarine manufacturer, V = VSEL, R = RRA).
system work together to set the weapons system sonar performance,
for example. There are many other considerations than the noise. For
example, the fuel life of the submarine is an important through-life-
cost issue which affects both the hull design and the NSRP design not
least because of the exigencies of refueling a nuclear reactor by means
of opening a pressure hull designed to go to great depths without
failing. The fewer times this is done the better.
9. Effect of competence on marketability
The marketability of this complex product is the ability of VSEL
(in interaction with RRA) to satisfy the needs of the buyer. If we
consider the internal valuation system of each of the contractors, VSEL
and RRA we begin to see how the competences of each affect that
marketability. Fig. 3 shows how the VSEL top-team see the interaction
between such important attributes of the firm as its ability to bear risk,
its financial security and the valuation placed on those attributes by
the governmental buyer. We shall see later that the VSEL team takes
into account the contribution of RRA towards that overall project
competence as well as the VSEL and RRA risk-bearing capabilities in
their own respects. Thus we will see VSEL working within the context
of the valuation system of the governmental buyer and manipulating
the joint valuation system of itself and RRA to satisfy the market
demand.
VSEL, of course, is concerned in the long term with their own
security as submarine manufacturer (security of position as s/m mfr
(V)) that security stems from a number of interrelated attributes (as it
happens all of them positively correlated for cause and effect). An
important relationship exists between the overall technical knowl-
edge of integration discussed above (technical knowledge integration
(V)) and the individual technical competences of knowledge of hull,
NSRP and weapon system components. Each of these are bolstered by
the track record of VSEL (track record V) in the sense that the longer
the track record (the more business won) the more opportunity there
has been for knowledge of this integration aspect of the project to be
co-created by VSEL and its partners and hence to be accessible to VSEL.
From a KM perspective, VSEL's marketing imperative is to maintain
that engine of co-creation by participating in projects. As Noel Davies,
the Chief Executive put it in conversation with his MD, ‘The most
442 J.H. Powell, J. Swart / Industrial Marketing Management 39 (2010) 437–449

Fig. 4. Rolls Royce and Associates' valuation system.

important thing is to be there [in the project]. That lets us learn more
and learn more and faster than [competitor 1].’
10. RRA's valuation system
We see in Fig. 4 the very similar cause and effect diagram of RRA's
competence dynamics and their interrelation with the desired char-
acteristics of the firm. In a similar way to that of VSEL in Fig. 5, we see
the important loop of track record R → technical knowledge (NSRP) R →
security of position as NSRP provider → track record R. This loop, once
kicked off in a positive direction will tend to self-reinforce, thus
making RRA's position as monopolistic NSRP provider even more
difficult to supplant because any outsider will have no access to the
RRA knowledge. More subtly, this loop in RRA's valuation strengthens

Fig. 5. Combined triadic system diagram.

 J.H. Powell, J. Swart / Industrial Marketing Management 39 (2010) 437–449
VSEL's hand as the incumbent main contractor, since any newcomer
attempting to supplant them would have to have access to that NSRP
component of the co-created knowledge needed to raise the overall
project capability (Fig. 5). Observe that the integrity of the 1954
agreement appears in the RRA valuation cause and effect diagram. This
agreement between the US and the UK in essence ‘licenses’ RRA to be
the UK's repository of knowledge and provider of the relevant nuclear
technology, Clearly other firms, all in the US, which have that capa-
bility would clearly and understandably prefer to be the supplier and
hence there is inherent pressure on the governments only to maintain
the 1954 agreement in place so long as RRA has a demonstrable ability
to discharge their duties. In short, the firm's technical and financial
competence is essential to attenuate any desire by the US nuclear
lobby to remove its UK monopolistic licensed status.
11. Triadic system
Clearly the two supplier firms' valuation systems diagrams interact
with those of the buyer. We can see this even at the most obvious
level, because certain variables (such as track record V) appear in more
than one of Figs. 2–4. At a practical level, too, it is clear that we should
expect the competence of each of the contractors to appear as a
conditioning factor in their long-term attractiveness as contractor to a
long-term buyer such as a national government. Fig. 5 combines these
three valuation systems by concatenating the diagrams through the
identification and absorption into one valuation system diagram of
any variable held in common with another. Certain minor duplications
have been removed but no other topological changes are made to the
combined figure at this point.
Fig. 5 reflects the essential structure shown in Fig. 1. We have yet to
add the other connections between the two firms, those which we
called the indirect mechanisms previously.
Fig. 5 reflects the realities of the project in a very satisfying way
(i.e. in a way which resonates with the experiences and understanding
of the informants). It can be seen that it falls into three main parts,
namely the VSEL valuation system in the top left, the MOD customer
top centre and the RRA valuation system bottom right. While there are
implicit inter-company connections for example overall project
capability and ability to contribute to project R, most of the connections
between VSEL and RRA pass though the mediating mechanisms of the
MOD. For example, RRA's ability to bear nuclear risk affects, through
a long chain, VSEL's track record but only through the MOD's will-
ingness to place home contracts (ability to bear risk R →residual risk to
crown → willingness to place home contracts → track record V). This is
most accurate; VSEL and RRA are not the same company; they do not
come together in joint projects because of an inherent joint interest
but because the buyer values their joint product and because they
cannot produce that product in isolation. We see later, as we add
further components to the direct interactions between VSEL and RRA,
that the knowledge sharing underpinning this MOD-mediated joint
motivation is an essential part of the project relationship.
12. Knowledge management considerations
We turn now to the counterpoint of this paper, namely the way in
which knowledge and its management are central to the competitive
advantage and survivability of these two firms. Fig. 6 shows some of
the key specific KM mechanisms elucidated by the informants. They
are evidenced in an extensive literature as surveyed by Swart (2006)
and before integrating it into the larger influence diagram we discuss
briefly the KM praxis issues which underlie the informants' links
shown in Fig. 6.
Fig. 6 is not a full causal loop diagram (in fact there are no loops in
it). It is, rather, a summary of a handful of key mechanisms evidenced
in the KM literature which address the practicalities of engendering
the co-creation of knowledge in an organization (here the complex
443
Fig. 6. Knowledge co-creation mechanisms.
interrelationship of the UK government scientists, VSEL, RRA and the
joint project team).
Studies of the way in which knowledge is created between and
within professional groups show that there are a number of con-
ditioning factors which support that co-creation. We see three of
those conditioning factors emerging here from the actions of the
industrial partners in setting up their project teams to respond to the
MOD's requirement.
• A collocated joint project team was set up near the MOD authority. The
objective of this team was very clear: by collocating subject experts
with different professional skills and knowledge (heterogeneity of
project team in Fig. 9) knowledge was ‘worth sharing’ because it was
obvious to team members that their peers both knew a great deal
about something relevant to the design and that what was known was
not known by others. Thus RRA put in an expert PWR NSRP designer
who could be seen to be a subject expert in those matters by the VSEL
submarine hull designers.
• The two industrial partners, however also set up an in-firm team of
experts to look after their own firm's interest. Clearly there was a
decision to be made about the extent of collocation vis-à-vis the in-
house teams (see degree of collocation of project teams) and this
was in part a function of the expressed (and therefore actionable)
expression of mutuality by the two firms. In other words the more
they believed and moreover publicly stated that they were engaged
in a common endeavor, the more they had to weight the total
staffing towards collocation and away from single firm teams.
• Being a member of a collocated project team, however is not just a
matter of working together but of socializing, building trust and, in
general, building social capital (Ambrosini & Bowman, 2001; Argyris
& Schön, 1996; Leonard & Straus, 1998; Swart & Kinnie, 2003;
Wenger, 2000). Moreover the working arrangements of the joint
team contribute towards the organizational social capital compo-
nent of the co-creation context. We summarize these by the variable
common experiences of the project team members which tends to
create homogeneity of view and hence to reduce heterogeneity of
project team. Conversely, however, the common experiences build a
common currency of communication, a lingua franca through which
the trust, reputation and other attributes emerging from the social
capital are expressed (Ambrosini & Bowman, 2001; Brusoni et al.,
2001; Cook & Emerson, 1978; Ford, 1997; Gummesson, 2002;
Håkansson & Snehota, 1995). The term lingua franca here is not to
be understood in the strict linguistic sense, but as shorthand for the
whole set of carriers of the social capital, linguistic, meta-linguistic,
behavioral etc.
Thus three well-understood knowledge co-creation mechanisms
can be seen operating in Fig. 6, namely, heterogeneity of teams
444 J.H. Powell, J. Swart / Industrial Marketing Management 39 (2010) 437–449
(leading to a perception of the worth of communicating in pursuit
of co-creating knowledge) homogeneity of teams (expressed by the
informants as common experiences of project team members and lead-
ing to a means of communicating and an inherent trust in the mu-
tuality of co-creation) and the clarity of expression of mutuality
(leading to initially externally-derived and later internalized valua-
tions of the worth of new co-create knowledge).
Fig. 7 shows the complete causal loop diagram including the
indirect connections just discussed and is the specific form of Fig. 1 for
the case in hand.
In Fig. 7 (towards the bottom of the figure) we see the effect
of the contributions of the two firms’ designers to heterogeneity.
Each firm brings its own specialized knowledge, so that if the other
brings that knowledge too, heterogeneity is reduced, Thus we see
that VSEL's adding of hull knowledge technical knowledge (hull) V is
of opposite sign to that of RRA's (lesser) knowledge of submarine
hulls. This is because, given VSEL's existing knowledge, the appear-
ance of RRA's similar knowledge reduces the heterogeneity of the
project. The more RRA designers know about what VSEL ‘brings to
the party’ the less the heterogeneity of knowledge in the joint team.
This is not true of knowledge about the integration aspects, how-
ever, since the more that is known about the system under design
as a whole the less the knowledge heterogeneity in the team. Thus
Fig. 7. Full loop diagram including indirect (firm to firm) connections.
we see in Fig. 7 that in both firms' cases knowledge of integration
aspects reduces heterogeneity.
We can also see in Fig. 7 that the propensity to share (more strictly
to co-create) knowledge leads to an increase of each firm's knowledge
of the other's expertise and hence to knowledge of the all-important
integrative aspects of the project.
This completes the representation of the interactive system be-
tween firms and market. We turn now to the use of the represen-
tation in identifying action in managing the relationships which it
captures.
13. Actual events
During the gestation period of the nuclear submarine project in
question, both these approaches, direct and indirect, could be seen. At
one point (early 1990) HM government made a strong statement that
for the new replacement for the Trafalgar Class submarines, (a project
known as SSN20), offshore purchase using the national dockyards
of France was a new and distinct possibility. The effect was to cause
the two protagonists here, VSEL and RRA, to make clear public state-
ments of mutuality and common cause, claiming a joint track record in
submarine design and invoking a movement towards the establish-
ment of a common project team. The loop of Fig. 8 can be seen to be
 J.H. Powell, J. Swart / Industrial Marketing Management 39 (2010) 437–449
Fig. 8. Effect of knowledge sharing on integration knowledge.
operating with the variable willingness to place home orders as the
initiating factor in the change of valuations of the contractors (and
eventually the buyer).
After some time the expense of SSN21 caused HM government to
suspend the new build project and instate a redesign of the existing
platform, the Trafalgar Class submarine. This new project, known as
Batch 2 Trafalgar Class (B2TC) eventually produced the Astute class of
UK nuclear submarines. It commenced in early 1993.
Before any statement of offshore purchase could be made, the cor-
porate players acted decisively. They immediately set up a connected
set of in-house and joint project teams, the latter split between a
requirements team placed near the buyer and a large design team sited
at the dockyard where the submarines would be built. The teams
immediately reacted to this clear statement of mutual intent and com-
menced a fast-moving requirements activity which took the design
initiative away from the MOD customer. Because of the collocation,
the expression of mutual endeavor and the organizational structural
arrangements (such as the funding arrangements, reporting chains and
spread of responsibilities of project posts among the companies, which
now included GEC and BAe Systems) the project team was highly
motivated to work together and the design progressed rapidly.
This effective initiative on the part of the companies indicates
clearly that the marketing of the project need not be enacted in direct
response to the buyer's valuation but, through a system approach,
indirectly, using the system mechanisms exposed by the causal loop
diagram methods of System Dynamics. Secondly, it illustrates that the
KM mechanisms are central to this example of accommodation of
valuation systems between provider and buyer. It is the setting up and
effective establishment of a system for co-creating the knowledge
necessary to convince the UK government of the competence of
the joint project which cut off at the pass any potential renewal of an
offshore purchase policy.
445
14. Discussion
Fig. 8 further illustrates the connectivity between the market
issues and the valuation system of one of the participating firms
through the KM mechanisms.
We see that as the MOD withdraws support from the home market,
the experience of RRA in the long term declines, since their designers
become less au fait with the needs of the complex system. Thus they
have less to bring to the conversation within a project. In turn this leads
to an INCREASE in the heterogeneity of the team: the VSEL designers
have much integration expertise but the RRA designers are tending to
bring only NSRP knowledge. Hence the demarcation within the project
is more visible and, perhaps counter-intuitively, the worth of sharing the
disparate knowledge increases. Hence the VSEL designers take more
from the knowledge environment for their part and their ability to
govern the project risk thereby goes up. In short, over a long period of
time the effect is to stabilize the relationship between the two firms; as
cooperation operates, heterogeneity decreases and it becomes less
worthwhile to work together — we see the platform supplier becoming
more confident in its relationship with the MOD customer, but as the
RRA specialists work separately they establish greater heterogeneity and
it became once more, worthwhile for VSEL to work with them. In terms
of relationship strategy the conclusion is obvious — RRA should, to the
extent that it sees a joint submarine product emerging from a long term
relationship with VSEL, ensure that in periods when it is in contact with
VSEL its engineers are absorbing the maximum amount of submarine
knowledge while, when the two firms are not working together, RRA
should embrace enthusiastically the specialist expertise of nuclear
engineering in order the better to be valued by VSEL in the later and
inevitable cooperative stage. We see again that the effect of a market
statement by the customer affects in a very direct way those variables
which represent the valuation system of the two contracting firms.
446 J.H. Powell, J. Swart / Industrial Marketing Management 39 (2010) 437–449
SBKM, however, allows more than a generalized approach to be
taken to the management of knowledge between firms in such B2B
relationships, since it allows a more precise, procedural approach to
the KM aspects of the system.
Fig. 9 shows how actors can be attached to Fig. 8 in order to show
which agents are controlling the strength of connection between the
important causal links of Fig. 8. Actors in the procurement authority or
national government are given Greek letters while actors in the two
firms carry Roman letters.
We see that the key controlling influences over the extent to which
a reduction in the MOD's willingness to place contracts induces
mutuality between the two contractors are, naturally, the boards of
the two firms, V and R. Similarly we can observe towards the top of
the diagram how the project director, D, the project staff S, the MOD
staff, σ, and the MOD risk team, π, all act together to control the extent
to which the main contractor's ability to bear risk is brought into effect
in an actual reduction in risk to the crown. The full SBKM procedure
requires this to be done for all appropriate loops; here we illustrate the
method with only the one loop.
Having identified the relevant agents, we proceed to establish
what knowledge they are using in the specific rôle associated with
each link. Note that the knowledge and indeed the rôle of each agent
will in general be different as each of their positions in the system is
examined. For example, the rôle of and knowledge employed by the
Project Director, D, in engendering a rich tacit learning environment in
Fig. 9. A key loop with agents attached to causal links.
the causal connection propensity to share knowledge → overall project
capability will be very different from his rôle and knowledge in the
link overall project capability → perceived VFM. In the former rôle he is
concerned with knowing the way in which knowledge is co-created
in a professional team whereas in the second rôle he is much more
directed towards the specific tasking of that team towards a specific
set of design aims.
Table 1 illustrates the nature of the SBKM output in indicating
the knowledge, both tacit and explicit employed by actors in such a
system diagram. Table 1 shows the knowledge deployed by the agents
in only two links, namely the connections willingness to place home
orders → clarity of expression of mutuality and overall project capa-
bility → perceived VFM. These links are representative of the output
of the method and are selected here to indicate specific marketing
lessons. The entries in the boxes marked explicit and tacit (knowl-
edge) are indicative of the knowledge deployed and do not constitute
a full description of that knowledge.
Table 1 indicates the way in which the SBKM method can support
the marketing policy by connecting the agents who have control
over the system behavior (through the loop behavior) and the knowl-
edge that they use. Marketing is then seen as a process of engagement
with that knowledge as deployed by agents in the context of the
system.
For example, consider Fig. 8. We see that in order to achieve the
desired result of improving the propensity of the MOD to place home
 J.H. Powell, J. Swart / Industrial Marketing Management 39 (2010) 437–449 447

Table 1
Tacit and explicit knowledge deployed in submarine project.

Link Agent Initial motivation Knowledge deployed

Willingness to place home orders → V Low
clarity of expression of mutuality
R Low
μ Indifferent
Overall project capability → perceived VFM D Very high
S High
σ Indifferent
τ Negative
Explicit
Appropriateness of channels of communication
MOD organization and relationships
Expressed values of stakeholders
Expressed values of MOD …
Tacit
Appropriateness of communication and language
‘Which strings to pull’ …
Explicit
Appropriateness of channels of communication
MOD organization and relationships
Expressed values of stakeholders
Expressed values of MOD …
Tacit
Appropriateness of communication and language
‘Which strings to pull’ …
Explicit
Contact details of V and R CEOs
V and R expressed policy …
Tacit‘How to talk their language’
Personal experience of V and R CEOs
‘How to talk their language’
Explicit
Design linkages leading to VFM
What constitutes VFM
Tacit
Flexibilities in VFM definition
How to ‘sell’ the design
How to motivate for design excellence
Explicit
Engineering knowledge
Tacit
Experience-based knowledge of how submarine systems interact
Explicit
Design book definitions of VFM attributes…
Tacit
Flexibility in meaning of VFM…
Explicit
Rule-based investment appraisal knowledge
Tacit
Cultural expectations of Treasury mandarins

contracts, we should be considering such things as the rôles of the two
CEOs and the Minister of Defense. Conventionally one would consider
how the CEOs might act directly upon the position of the Minister
in order directly to alter his propensity to place home orders. With
the system viewpoint, however, we see that the CEOs should be
preemptively setting out an agenda of mutuality and we can help to
bring this about by ensuring that the appropriate knowledge shown n
Table 1 is, in fact, available to the protagonists. For example, should the
Minister change due to a reshuffle or change of government, the loop
of Fig. 8 will become very weak in its effect because of the weakening
of the knowledge shown in Table 1. In particular the tacit under-
standing of the social norms and expectations of a new minister would
be an absolute priority in order to ensure the renewed strength of the
critical mechanisms of success.
15. Managerial implications and practice
The application of the SBKM method to the case study illustrates
some of the benefits to practice and the academic perspective of
business networks.
15.1. Practicalities
In practical terms the method indicates that it is possible to move
from the somewhat abstracted view of business networks necessary
to advance their study as an intellectual schema towards that which
is needed to form the basis for action-centered managerial action.
Abstraction is necessary, desirable and welcome in providing general-
izations upon which academic progress is founded, but the practi-
tioner requires support in identifying potential actions to manage the
business network and, ideally, some prediction of the consequences
of that action. SBKM as a representation of the specifics of a firm's
situation is very powerful in the first respect. Its natural product
(illustrated above through the knowledge based action analysis) is to
generate lists of actions which, while they still have to be reconciled
one with the other through the ubiquitous political process of decision
making in firms, can be guaranteed to represent ‘good’ things to do.
We mean by this that each action can be associated with a specific and
visible system mechanism to achieve a result favored by the client/
manager. While it has less to offer in terms of its predictive ability
(since it is qualitative) it does allow arguments of causality to be
presented, which, nonetheless, can be very powerful in advocacy for
resources.
15.2. Benefits to the study of networks
The ability to represent a specific situation is of inestimable
value to the academic study of business networks, not least because
it provides a valuable research tool to capture and place under a
generalizing theory the experience, challenges and context of a firm.
Since SBKM derives from what is now a highly standardized family
of methods (namely Qualitative System Dynamics) the method has
448 J.H. Powell, J. Swart / Industrial Marketing Management 39 (2010) 437–449
the potential of forming of itself a standardized and therefore uni-
fying representation method for the study of the dynamics of
business networks.
15.3. Future work
It is anticipated that the knowledge structure view of marketing
is applicable not only to the B2B case, but to the B2C case, where the
value relationship with the customer is one-to-many rather than
one-to-few and we anticipate the application of the knowledge a
management view to this, more complicated requirements capture
process as part of the ongoing product design process of B2C,
particularly in such interactive products as automobiles and
electronic goods where the need to capture and condition emergent
valuations on the part of potential customers is critical.
An exciting but emergent field of study is that of epistemetrics,
the measurement of knowledge. Recent work (Powell & Swart,
2008) indicates that it may now be possible to measure the flow of
knowledge per se around a system. Past work has concentrated on
the value of knowledge and its effects and social embodiments.
While these remain vital viewpoints, the absence of a credible,
consistent measure of how much is known at a node of a network
has constrained the representation of these systems in respect, for
example, of normative advice in resource allocation aimed at the co-
creation of knowledge in the firm. The existence of a knowledge
scaling metric should allow a substantial increase in our ability to
represent the dynamic features of knowledge flow around a busi-
ness network.
The focus of this present initiative in furthering a KM perspective
on marketing, however, remains the extension of practice in the
further application of the SBKM method within marketing to allow the
practical exercise of KM perspectives in improving the effectiveness of
marketing, both in B2B and wider applications.
References
Ambrosini, V., & Bowman, C. (2001). Tacit knowledge: some suggestions for operationaliza-
tion. Journal of Management Studies, 38(6), 811−829.
Argyris, C., & Schön, D. (1996). Organizational learning II: Theory, method and practice.
New York: Addison-Wesley.
Baumard, P. (1999). Tacit knowledge in organisations. London: Sage.
Brusoni, S., Prencipe, A., & Pavitt, K. (2001). Knowledge specialization, organizational
coupling and the boundaries of the firm: Why do firms know more than they make?
Administrative Science Quarterly, 46, 597−621.
Burt, R. (1992). Structural holes: The social structure of competition. Cambridge, MA:
Harvard UP.
Checkland, P., & Scholes, J. (1990). Soft systems methodology in action. Chichester: John
Wiley.
Cook, K., & Emerson, R. (1978). Power, equity and commitment in exchange networks.
American Sociological Review, 721−739.
Cook, S. D. N., & Seely Brown, J. (1999). Bridging epistemologies: The generative dance
between organisational knowledge and organisational knowing. Organization Science,
10(4), 381−400.
Coyle, R. G. (1977). Management system dynamics. Chichester: John Wiley.
Coyle, R. G. (1998). The practice of system dynamics: Milestones, lessons and ideas from
30 years experience. System Dynamics Review, 14(4), 343−365.
Coyle, R. G. (2000). Qualitative and quantitative modelling in system dynamics: Some
research questions. System Dynamics Review, 16(3) (Fall), 225−240.
Coyle, R. G., & Exelby, D. (2000). The validation of commercial system dynamics models.
System Dynamics Review, 16(Spring) (1), 27−41.
Davenport, T. H., & Prusak, L. (1998). Working knowledge: How organizations manage
what they know. Boston, MA: Harvard Business School Press.
Eden, C. (1989). Using cognitive mapping for strategic options development and analysis
(SODA). In J. Rosenhead (Ed.), Rational analysis for a problematical world Chichester:
John Wiley.
Flood, R., & Jackson, M. C. (1991). Creative problem solving: Total systems intervention.
Chichester: John Wiley.
Ford, D. (1997). Understanding business markets, 2nd ed. London: Dryden.
Ford, D., Gadde, L., Håkansson, H., & Snehota, I. (2003). Managing business relationships,
2/e. Chichester: John Wiley.
Ford, D., & Redwood, M. (2005). Making sense of network dynamics through network
pictures: A longitudinal case study. Industrial Marketing Management, 34(7), 648−657.
Forrester, J. W. (1961). Industrial dynamics. Harvard, MA: MIT Press.
Glynn, M. (2007). Beyond constraint: How institutions enable identities. In R. Greenwood, C.
Oliver, R. Suddaby & K. Sahlin-Andersson (Eds.), Handbook of organizational institution-
alism London: Sage.
Granovetter, M. (1973). The strength of weak ties. American Journal of Sociology, 78,
1360−1380.
Gummesson, E. (2002). Relationship marketing and a new economy: It's time for
deprogramming. Journal of Services Marketing, 16(7), 585−589.
Håkansson, H., & Ford, D. (2002). How should companies interact in business networks?
Journal of Business Research, 55(2), 133−139.
Håkansson, H., Johanson, J., & Wootz, B. (1976). Influence tactics in buyer–seller processes.
Industrial Marketing Management, 4(6), 319−322.
Håkansson, H., & Snehota, I. (1995). Developing business relationships in business networks.
London: Thomson.
Kahaner, L. (1996). Competitive intelligence: From black ops to boardroom — How businesses
gather, analyze and use information to succeed in the global marketplace (pp. 20−35).
New York: Simon and Schuster.
Leana, C. R., & van Beuren, H. J. (1999). Organizational social capital and employment
practices. Academy of Management Review, 24(3), 538−555.
Leonard, D., & Straus, S. (1998). Putting your company's whole brain to work. Har-
vard Business Review on Knowledge Management. Boston: Harvard Business School
Press.
Liddell, W., & Powell, J. H. (2004). Agreeing access policy in a general medical practice:
A case study using QPID. System Dynamics Review, 20(1), 49−73.
Montibeller, G., & Belton, V. (2008). Causal maps and the evaluation of decision options—
A review. Journal of the Operational Research Society, 57, 779−791.
Morris, T. (2000). Promotion policies and knowledge bases in the professional
services firm. In M. Peiperl, M. Arthur, R. Goffee & T. Morris (Eds.), Career
frontiers (pp. 138−152). Oxford: Oxford University Press.
Nonaka, I. (1994). A dynamic theory of organizational knowledge creation. Organization
Science, 5(1), 14−35.
Polanyi, M. (1966). The tacit dimension. London: Routledge and Kegan Paul.
Powell, J., & Bradford, J. (1998). The security–strategy interface: Using qualitative
process models to relate the security function to business dynamics. Security
Journal, 10, 151−160.
Powell, J., & Coyle, R. (2005). Identifying strategic action in highly politicised contexts using
agent-based qualitative system dynamics. Journal of Operational Research Society, 56,
787−798.
Powell, J., & Swart, J. (2006). Men and measures: Using qualitative system modelling
to map knowledge in firms. Journal of the Operational Research Society, 57,
10−21.
Powell, J., & Swart, J. (2008). Scaling knowledge: How knowledge accrues in organisations.
Journal of the Operational Research Society, 59, 1633−1643.
Quinn, J., Mills, D., & Friesen, B. (1992). The learning organisation. European Manage-
ment Journal, 10, 146−156.
Ramos, C., & Ford, D. (2006). Network pictures: Conceptual development and pre-
liminary findings. Proceedings of the 2nd IMP Conference, Milan: December 2006.
Reagans, R., & Zuckerman, E. W. (2001). Networks, diversity and productivity: The social
capital of corporate R&D teams. Organization Science, 12(4), 502−517.
Richardson, G. P., & Pugh, A. (1981). Introduction to system dynamics modeling with
DYNAMO. Harvard, MA: MIT Press.
Ritter, T., & Ford, D. (2004). Interaction between producer and user in business markets.
In D. Håkansson & D.A. Harrison (Eds.), Towards a new approach to marketing
Chichester: John Wiley.
Scarbrough, H., Swan, J., & Preston, J. (1999). Knowledge management: A literature review.
London: Chartered Institute of Personnel and Development (CIPD).
Simon, H. (1972). Theories of bounded rationality. In R. McGuire (Ed.), Decisions and
organisation Amsterdam: North Holland.
Sterman, J. D. (2000). Business dynamics. Boston, MA: McGraw-Hill.
Stryker, S. (1980). Symbolic interactionism: A social structural version. Menlo Park, CA:
Benjamin/Cummings.
Stryker, R. (2000). Legitimacy processes as institutional politics: Implications for theory
and research in the sociology of organizations. Organizational Politics: Research in
the Sociology of Organizations, 17, 179−223.
Swart, J. (2006). Intellectual capital: Disentangling an enigmatic concept. Journal of
Intellectual Capital, 7(2), 136−159.
Swart, J., & Kinnie, N. (2003). Knowledge-intensive firms: The influence of the client on
HR systems. Human Resource Management Journal, 13(3), 37−55.
Swart, J., & Powell, J. H. (2006). This is what the fuss is about: Systemic modelling for
organisational knowing. Journal of Knowledge Management, 9(2), 45−58.
Tsoukas, H. (1996). The firm as a distributed knowledge system: A constructionist
approach. Strategic Management Journal, 17, 11−25.
Tsoukas, H., & Vladimirou, E. (2001). What is organizational knowledge? Journal of
Management Studies, 38(7), 973−994.
Turnbull, P., & Valla, J. P. (1986). Strategies for international industrial marketing. London:
Croom-Helm.
Tustin, A. (1953). The mechanism of economic systems. London: Heinemann.
Vargo, S., & Lusch, R. (2004). Evolving to a new dominant logic for marketing. Journal of
Marketing, 68(January), 1−17.
von Krogh, G., Roos, J., & Kleine, D. (1998a). Knowing in firms. London: Sage.
von Krogh, G., Roos, J., & Kleine, D. (Eds.). (1998b). Knowing in firms:Understanding,
Managing and Measuring Knowledge Pub: Sage ISBN-13: 978-0761960133.
Weick, K. (1979). The social psychology of organising. Reading, MA: Addison-Wesley.
Wenger, E. (2000). Communities of practice and social learning systems. Organization,
7(2), 225−246.
Wolstenholme, E. F. (1990). System enquiry. Chichester: John Wiley.
 J.H. Powell, J. Swart / Industrial Marketing Management 39 (2010) 437–449 449

John Powell is Professor of Strategy at Cardiff University, having held similar
appointments at Bath University and Southampton after a career at board level in the
European Defense industry. His main research area is in the application of
mathematical methods, and particularly those of Operations Research and Systems
Theory to the study of knowledge and strategy. He holds a PhD from Cranfield
University together with HM the Queen's Gold Medal for academic excellence and
the OR Society's President's Medal.
Dr. Juani Swart specializes in Knowledge management and the management of Knowledge
workers at the University of Bath where she is Director of the Work and Employment
research Centre (WERC). Her early research focused on collective tacit knowledge her
current work develops an understanding of knowledge and its renewal in social networks
(for which she has received a best paper award). Juani has published in the areas of
people management in knowledge intensive firms, intellectual capital structures, systems
approaches to knowledge management and network influences on strategic choice.