Sunteți pe pagina 1din 23

Journal of Market-Focused Management, 5, 331 – 353, 2002

# 2004 Kluwer Academic Publishers, Boston. Manufactured in The Netherlands.

The Process of Business Relationship


Ending – Its Stages and Actors
JAANA TÄHTINEN jaana.tahtinen@oulu.fi
Faculty of Economics and Business Administration, University of Oulu, P.O. Box 4600, FIN-90014, Finland

Abstract

How to describe a process of business relationship ending, while considering both the
aspects of time and the multiplicity of actors involved? The process of ending consists
of actions performed by the actors, as well as their perceptions and decisions concerning
the future of the relationship and the consequences they have on the relationship. A
longitudinal, dyadic case study is used to empirically ground a process framework. The
model distinguishes six stages in the ending process and depicts different actor levels
involved. The process model helps to understand the events and actions that end the
relationship.
Keywords: business relationship, relationship ending process, processual research

Introduction

This research aims at increasing our knowledge of the business-to-business relationship


ending process. We have a fairly incomplete understanding of the ending phase as such
(e.g. Dwyer, Shurr and Oh, 1987; Sheth and Parvatiyar, 1995), although the ending of
business relationships is currently viewed as one of the essential elements of marketing
(e.g. Fiocca, 1982; Grönroos, 1997). The main interests of existing research have been
dissolution as a cognitive decision-making process (e.g. Heide and Weiss, 1995) and
the communication strategies used during the dissolution process (e.g. Alajoutsijärvi,
Möller and Tähtinen, 2000). Even though they have made significant contributions to
dissolution research, they have certainly not been enough to reveal all the aspects of the
phenomena.
How to describe a process of business relationship ending, while considering both the
aspects of time and the multiplicity of actors involved? As relationship ending is not an
event, but a process (see Pettigrew, 1990), the present study takes a processual approach.
Following a literature review, the paper builds and empirically illustrates a framework of
a business relationship ending process, incorporating both the time dimension (actions,
events, stages and process) and the multiplicity of the actor levels (individuals, both
partner companies, dyad, and network actors). The paper ends with a discussion of the
framework, presentation of propositions derived from the framework, and suggestions for
future research.
332 TÄHTINEN

Positioning of the Study

The theory of business relationship ending is only just evolving; therefore no termino-
logical conventions have been established. Tähtinen and Halinen (2002) suggest that
different terms should be used when referring to different kinds of endings, e.g. that ending
could be used as a general term, and that dissolution could refer to naturally ending
relationships. In this study the term ending is used because the model aims to cover all
kinds of endings.
The present study relates to the Interaction and Network Approach (see e.g. Ford, 1990;
Håkansson and Snehota, 1995), which sees relationships as developing in a continuous
process between companies embedded in a network. These interactive processes1 create
interdependence between the actors, which results in the development of the substance of
the relationship, i.e. activity links, resource ties and actor bonds connecting the parties2.
Thus a business relationship consists of interaction processes producing activity links,
resource ties and actor bonds between two companies (see also Tähtinen, 1999). Following
this definition, the process of ending disconnects the former partner companies from each
other by cutting the activity links, resource ties and actor bonds that have kept them
together.
The framework this study suggests will take in both the time dimension, and the
multiplicity of the actors involved in the process. To describe the complex process of
business relationship ending over time, more than just one stage is needed (Dwyer et al.,
1987). Thus, the framework will conceptualise the process through several stages. In
addition, business relationship ending process involves several actor levels in each
partner company (individuals, departments, business units, the companies as a whole) as
well as in the connected network of relationships (third parties, dyadic relationships with
them and broader networks) (see also Halinen and Tähtinen, 2002). Therefore, in order
to give a reasonably detailed description of the ending process both the different stages
and the different actor levels are incorporated in the framework.
Before describing the existing research on business relationship ending, it is necessary
to discuss the view of process that this study applies. Process is defined here as ‘‘a
sequence of events or activities that describes how things change over time’’ (Van de Ven,
1992). Thus, the meaning of process in this study is not a linear sequential model of
development, but a multiple progression of streams of activities that unfold as the
relationship develops3 over time. Multiple progression assumes that developmental
processes can follow more than a single possible path. The theories of change that
are used to explain how the change unfolds in relationship are teleology and dialectic
theories (Van de Ven and Poole, 1995). More specifically, teleology operates on single
entities and dialectical theories on multiple entities. Thus if we view a company and a
relationship as a single entity, teleology serves as a motor of change, but if we look
inside the entities, seeing the company as a collection of heterogeneous individuals and
a relationship as a dyad of two companies, dialectical theory serves as a motor of
change. The process of business relationship ending is described here from both points
of view, consisting of some multiple cumulative sequences of events moving the entity,
i.e. the relationship towards an envisioned end stage, as well as some repeating strings
THE PROCESS OF BUSINESS RELATIONSHIP ENDING 333

of events over time which are produced by the contradictory forces of two entities,
i.e. two companies.

Existing Research on Relationship Ending Processes

Relationship ending studies can be divided into four approaches (for a review, see
Tähtinen and Halinen, 2002), of which the business marketing approach is of interest
here. The decision to choose only the business marketing approach is a restriction made
on purpose. I fully acknowledge that the studies focusing on consumers’ switching
behaviour (e.g. Keaveney, 1995; Roos, 1999) contain valuable results and areas of interest
(e.g. emotions in the ending process) that can contribute to the research on business
relationship ending. However, interorganisational business relationship ending differs
from consumer switching behaviour, and the differences hinder a single model from
sufficiently describing both kinds of relationship ending.
The business marketing approach includes 10 studies that place their focus, as does this
study, on the ending process (Tähtinen and Halinen, 2002). These studies, as well as one
recent process study (Halinen and Tähtinen, 2002) were looked at to find suitable models
for the ending process as the basis of this research.
Of these eleven studies, those of Alajoutsijärvi et al. (2000) and Giller and Matear
(2001) focus clearly on a specific stage of the ending process, namely communicative
strategies based on Baxter (1985). Grønhaug et al. (1999) place their main focus on the
perception of warning signals, an important aspect of the ending process. None of the three
studies provide a comprehensive model of the ending process.
The study by Rosson (1986) focuses on changes in international channel rela-
tionships. It finds that a relationship can deteriorate in a fairly predictable way, but
also in an unpredicted manner. Gadde and Mattsson (1987) discover exit patterns taking
place as a company once used as a single source supplier was gradually eliminated from
the supplier base. However, the data did not allow Gadde and Mattsson to explain the
patterns, nor did they describe the actions of the parties within the different patterns or
exit processes. However, these two studies clearly conclude that different kinds of
ending processes take place, although they do not try to model them.
Havila (1996) and Havila and Wilkinson (2002) focus on the life after relationship
ending. Their conclusion is that relationship ending is a controversial issue, as relationship
energy can continue to exist within a personal level in social bonds, although the business
relationship has ceased. The studies bring forth the important issue of actor levels, which
this study takes into consideration.
The remaining literature focusing on the business relationship ending process consists
of four studies (Halinen and Tähtinen, 2002; Halinen and Tähtinen, 1999; Tähtinen, 1999;
Tähtinen and Halinen-Kaila, 1997), reporting different phases of a research project. All
these studies use not only the aforementioned studies as a source, but also consumer
research on customer exit and switching (e.g. Roos, 1999), channels research on rela-
tionship termination (e.g. Ping and Dwyer, 1992), economics (Hirschman, 1975), and social
psychology (e.g. Duck, 1982) to develop the theoretical process model of business
334 TÄHTINEN

relationship ending. Tähtinen and Halinen-Kaila (1997) build a model of triadic network
ending, and Tähtinen (1999) as well as Halinen and Tähtinen (2002) restrict their focus
to a single relationship ending.
Since Halinen and Tähtinen (2002) is the only model that covers more than a few stages
or aspects of any kind of ending process, in addition to including different actor levels, this
study has applied it as a tentative model. The model distinguishes seven stages in the
ending process and it also suggests different actor levels involved with the process. The
assessment and the decision-making stages are suggested to occur within one partner
company, dyadic communication and restoration in interaction between the parties, and
network communication at the network level. The aftermath stage involves all levels
identified.
The model (Halinen and Tähtinen, 2002) suggests that at the assessment stage the
relationship’s future and how it could be ended is assessed and decisions concerning exit
or voice (see Hirschman, 1975) are made during the decision-making stage. Voice strategy
allows both companies to restore and maintain their relationship. The decision to use
either exit or voice is communicated to the partner during the dyadic communication
stage. In the disengagement stage business exchange declines and resource ties weaken.
At the network communication stage, the partners deal with the consequences of
break-up, communicating the ending to other actors within the network and establishing
new relationships. This is suggested to happen at the same time as the assessment,
dyadic communication and aftermath stages. Finally, ending is mentally processed in the
aftermath stage.
This study empirically grounds the tentative process model (Halinen and Tähtinen
2002) through a dyadic and longitudinal case study of a business relationship ending.
Data were collected from both seller and buyer organisations as well as from important
network actors through interviews, and also from written sources in order to ensure
triangulation (Denzin, 1978). From each company, several individuals involved in the
focal relationship were interviewed. The context was tailored software development,
which involves a high number of persons and produces a large number of written data
(e.g. project memos, specifications, error lists etc.) providing rich data. The data
collection procedures as well as the full case story are reported in Tähtinen (2001).

The Empirically Grounded Framework of Relationship Ending Process

The grounding of the tentative model resulted in three major changes to the framework.
Firstly, by using the notion of actor level in the same way throughout the process,
the new framework has reduced the number of stages. Therefore, the framework
suggests only one communication stage, instead of dyadic and network communi-
cation stages. Secondly, the titles of the stages and their different contents were clarified
with the aid of the case. This also resulted in combining the original assessment and
decision-making stages into a single consideration stage and adding the sensemaking
behaviour to the aftermath stage. Thirdly, a new stage, namely an enabling stage,
emerged from the data and was thus added to the framework. In the following sections,
THE PROCESS OF BUSINESS RELATIONSHIP ENDING 335

Figure 1. Framing the business relationship ending process.

the framework is briefly described. Each stage is illustrated later on in more detail with
the case data.
The empirically grounded framework describes the ending process by dividing it into
six stages performed by the four actor levels as illustrated in Figure 1. The consideration
stage is composed of the actors’ total decision-making behaviour (including information
seeking and uncertainty reduction) concerning the question of continuing or ending
the relationship. The communication stage refers to all communication (be it within
one collective actor, the dyad or the network) that concerns the decision to end or to
continue the focal relationship. Adopting an exit strategy in the communication stage
means that the company aims to terminate the relationship, but the voice strategy allows
companies to take steps to restore the relationship. If a voice strategy is used, the
success of the restoration of relationship depends on the reactions of the partner. Unless
the parties agree on performing restoring actions, in the disengagement stage, the usual
business exchange declines and therefore the existing resource ties, actor bonds, and
activity links erode. The disengagement stage may need to be proceeded by the enabling
stage, which involves all actions aimed at making the ending of the relationship
possible by lowering the exit barriers or reducing the importance of the attenuating
factors. The sensemaking / aftermath stage refers to all actions that are performed in
order to explain the process and mentally safeguard the actors during it (sensemaking) or
after it (aftermath).
The framework is a collection of the possible stages that the ending process travels
through. In some endings, restoring actions, for example, may not be performed at all, and
336 TÄHTINEN

the order of the stages may also vary. However, the stage description helps to divide the
complex process into shorter time periods. By describing actions with the use of the
different actor levels, i.e. entities performing the actions, allows a more detailed
description of the complexity of the process. This also explains why the stages may take
place partly simultaneously because different actor levels are performing the actions. For
example, individuals in the buyer company may be considering ending the relationship
while the seller company is restoring it.
The actor levels, e.g. individuals, companies, the dyad itself and the network in which
the relationship is embedded are all included in the model as different levels of analysis,
since the change unfolds through actions performed by each of the actor levels. A
business relationship exists between two company actors, but one company is capable of
ending it through its own actions (Simmel, 1950). Therefore, at least two actor levels are
needed in order to study the dissolution process, namely the relationship and the company
level. Moreover, large companies consist of smaller units and within these units, single
managers can be powerful actors (see e.g. Alajoutsijärvi et al., 2000). In addition, other
network actors can influence the focal relationship and its actors, either towards or away
from the dissolution. Therefore, a network level is needed to shed light on the actions of
third actors.
Next we will use the case of a long and winding ending to illustrate the different stages
and the actor levels suggested by the dissolution process framework presented above.
First the story of the relationship ending is presented to give an overview of the process.
After that each stage is described with the help of illustrative quotations. Finally, the actor
levels involved in the dissolution process and the paths that the process followed are
discussed.

Empirical Illustration of the Process Framework

The Story of a Long and Winding End

The case is about the ending of a relationship between a large telecommunication


company (Telcom) and a small software company (Smallsoft). Telcom was already
dissatisfied with the performance of Smallsoft during the early days of the relationship,
because of several critical events. As the restoring actions did not satisfy Telcom, the
personal relationships between the Project Managers of the two companies ended up in a
conflict. The working atmosphere was far from co-operative, although there were
attempts by the companies’ higher management to resolve the situation.
The situation culminated in Telcom blaming Smallsoft for bugs in the software, thus
refusing to accept the delivery, while Smallsoft blamed Telcom for changing the specifi-
cations and not setting up the hardware needed. After higher management negotiations,
Smallsoft agreed to change the software, after which Telcom accepted the application.
However, the consultant who had developed the software left Smallsoft before the
application was fit for production use, and this slowed down the work considerably.
Telcom felt that they had to have the application in production use in one way or
THE PROCESS OF BUSINESS RELATIONSHIP ENDING 337

Figure 2. Long and winding end – process description at all actor levels.

another before the warranty period was over. This prompted Telcom to hire Smallsoft’s
partner company (Softpal) to complete the application. Smallsoft considered the hiring
of Softpal as a contract violation, which led to the invalidation of the warranty of the
software. As a result, Smallsoft contacted Telcom concern’s Head Office and a Reso-
lution group was set up to settle things between Telcom and Smallsoft. In spite of the
efforts, the group was not able to resolve the disagreements. Instead, the group made
a decision on the minimum corrections that they wanted Smallsoft to make to the
application (once again) and, after that, to end the project. However, Smallsoft’s efforts
in carrying out the agreed corrections did not satisfy the group and they made a formal
complaint to Smallsoft. This meant that Telcom would not do any further business with
Smallsoft until the complaint was resolved.
The dissatisfaction towards Smallsoft had reached a level where Telcom together with its
long-term software vendor (Solidware) started to evaluate other alternatives for data
warehouse development. The conclusion was a decision not to continue data warehouse
development with Smallsoft. The relationship with Smallsoft was officially closed in the
Resolution group’s last meeting.
After this Smallsoft reorganised its customer support unit, which resolved the remaining
problems in the software. This more or less resolved the official complaint, but it was too
late to restore the relationship with Telcom. Solidware was already developing a totally
new data warehouse system for Telcom and the software developed by Smallsoft was
never used.
Figure 2 provides an overview of all the stages the long and winding end travelled
through, using the framework presented above. The next section will concentrate on the
activities within each stage.

The Stages of the Long and Winding End

The Consideration Stage

The consideration stage involves actors’ decision-making behaviour evolving around the
question of continuing or ending the relationship. In both case examples, there were
338 TÄHTINEN

several reasons to end the relationship. However, the reasons as such did not end the
relationships. The managers evaluated both the problems in the relationship as reasons to
end it and the good things as reasons to continue the relationship. Based on this evaluation,
the managers made decisions to either end the relationship or to try to restore it.
The consideration stage can be separated from a normal, ongoing evaluation or
assessment of the relationship. Such assessment is part of every phase of a relation-
ship, and thus it is not only a stage in the dissolution phase (see e.g. Järvelin, 2001).
Moreover, as long as the outcome of the assessment is satisfactory, i.e. the evaluating
persons are satisfied with the relationship and thus wish to continue it, it is not part of
the dissolution process. If the outcome of the assessment turns out to be severely
unsatisfactory, the assessment turns into consideration as to what to do about it – either
to continue or to end the relationship. The consideration stage also includes aspects
such as seeking additional information about the partner from the network that is done
in connection with considering the ending and making decisions about it.
In the long and winding end, both companies considered the continuation of the
relationship in internal discussions during different periods as illustrated in the following
quotations. However, the consideration stage also involved people from the network
level, as a quotation from a member of the Resolution group reveals. In addition,
Telcom’s Project Manager also conversed with Smallsoft’s reference companies while
searching for an alternative supplier.

Project Manager, Smallsoft: Within Smallsoft we did have these talks with our
Project Manager that our people don’t want to visit the customer, talks on how
each person had seen difficult projects, but the line must be drawn somewhere,
that this is where it is. We have had some rough talks about it but the othe side [the
ones wanting to continue the relationship] won.

IT-expert, Telcom’s Head Office: Of course we discussed it passionately!


Everybody was absolutely fed up with Smallsoft’s actions. Personally I can’t
understand such an attitude: that no – how should I put it – customer satisfaction
was even sought after. — It wasn’t like they even tried, it left us with a feeling that
they didn’t even try their best, they were kings and that’s it, we should just accept
it. And then everything [that was wrong] was peachy inherent characteristics of
their product, which I just cannot get into my head or then again as the saying
goes; change the product, if it has these characteristics. These characteristics are
quite enough, thank you very much.

Resolution Group Member, Telcom: Then when the matter didn’t seem to
progress and the summer holidays were approaching, we had talks about how long
we were going to keep the meetings going on, how much money we wanted to
use, as preparing for meetings costs money. — We decided that we were going to
try and get the minimal repairs in tack, i.e. the hard codes and defects, and then
we’d call it quits. The project wasn’t expensive, I mean the part that we had agreed
to buy, or perhaps a bit expensive, but not any gigantic project, so it was useless to
THE PROCESS OF BUSINESS RELATIONSHIP ENDING 339

spend so much money if we weren’t going anywhere. Then we decided to file an


official complaint, which was done by the Purchasing Director and the lawyers
from the Head Office.

Project Manager, Telcom: At the end of the year I heard in seminars and also
from Smallsoft that they had a good reference, AB Corporation, and we visited
them couple of times to get to know their application. I asked around in AB who
had been working for them and how it had gone. AB said that it hadn’t gone all
OK and AB had switched. AB had hired [new people] and also a couple from
Softpal. Well, then after that I’ve heard quite a lot of information from other places
about failed projects.

The Restoration Stage

During the restoration stage, actions are directed towards saving the relationship and thus
continuing it in the future. As with any actions, their goal may not be achieved. This is also
the case in the long and winding end. As the following quotations reveal, many restoring
actions took place at different time periods. In spite of the restoring actions of both parties,
also involving network actors, the relationship ended.

Steering Committee Member, Telcom: First it just evolved into these accusa-
tions from both sides about where the fault lies. In the Steering Committee, we
just agreed that we’d try to get it [the application] into production use, that the
defects that were there would be removed so that the application would at least
work. And that the missing documentation would be done. To my recollection
we promised to split the costs. For the documentation Telcom paid more money
up to a point but the supplier corrected the application errors.

IT-Director, Telcom’s Head Office: It was many times that I was forced to
resolve rising problems and misunderstandings that arose. Very often there were
two messages, one from Telcom’s side and another from the other side. They were
often different issues, so they mixed apples and oranges in that entire hassle.

Sales Manager, Smallsoft: The discussions and the official complaint and
handling, it was done directly by our Consulting Manager and Managing Director,
as the matters were so serious. — I was aware of it, because, in a way, I had been
involved in getting our Managing Director to be part of the formal complaint
resolution. So it wouldn’t be left, for example, to the Consulting Unit, which was
already one party in the project where the personal relationships had become
pretty critical. So my goals [of saving the relationship] were filled by our
Managing Director being involved.

It has to be noted that an action or an event meant to restore the relationship is not
necessarily perceived as a restoring action by every actor. As the quotations reveal, in the
340 TÄHTINEN

particular case, setting up the Steering Committee and replacing Project Manager, were not
perceived as restoring actions by every individual in the companies, nor by a third
company involved in the relationship.

Head of the Steering Committee, Telcom: At the beginning we tried to get a grip
on it and we talked with them [Smallsoft] and talked with our [Telcom’s] Project
Manager, that she shouldn’t take such a strict attitude in relation to these [things].
And we talked with Smallsoft and all around tried to smooth things over. It only
started to aggravate from there. — We did discuss things with both Smallsoft’s
Project Manager and their Sales Manager but at that point when we had to really
start looking for a higher gear, then we went to Softcom to have a word with the
Managing Director.

Sales Manager, Smallsoft: The vagueness of the project team took shape in very
sore personal relationships. From our side there was a new person involved who
started to lead the project [Project Manager Jeremy]. But not even switching the
people changed the situation, like when on our side the very experienced Laura
replaced Jeremy.

Project Manager, Solidware: Laura — almost like immediately refused to


communicate with Telcom’s Project Manager. I don’t know why she took that kind
of attitude. Obviously she thought that Telcom’s Project Manager was like a skid
for the whole thing.

The Disengagement Stage

Disengagement is a stage where the actor bonds, activity links and resource ties start to
weaken and, in the end, break down altogether. This stage can be precipitated by a conflict
between the two companies or individuals representing the companies, but there can be
other reasons as well. Thus the deterioration of the interdependency between the
companies may be a result of network changes (e.g. new alternative partners entering
the network) or change of company strategy (e.g. closing down a production line).
In the case of the long and winding end the first actor bond that broke down was the
personal relationship between the two Project Managers. On an individual level, this
conflict resulted in a broken communication tie between the companies, leading to
decreased co-operation in the relationship.

Steering Group Member, Telcom: And then it became all personal when the
supplier announced that they don’t want to be involved with our Project Manager
and they want someone else doing the testing. — And then Smallsoft’s Project
Manager once went through the defects with ours. After that she announced that
she doesn’t want to be involved with our Project Manager and then just went
through the stuff with our IT Manager. It was kind of ugly the rest of the year,
going on like that.
THE PROCESS OF BUSINESS RELATIONSHIP ENDING 341

Project Secretary, Telcom: You could feel it in the meetings that the atmosphere
was real tight and that there was like this small poking and jibing going on on
both sides (laughter). Well you could see from it that there was a feeling that
they [Smallsoft] weren’t willing to co-operate anymore. You could sense it quite
easily.

In the long and winding end, Smallsoft was able to maintain a co-operative
relationship with Telcom’s Head Office. However, the series of discussions
Smallsoft had with Head Office concerning the relationship were perceived by
many at Telcom as violations of the rules of project work. The two main
communication links were 1) between the two Project Managers and 2) between
the Project Managers and the Steering Committee. The discussions Smallsoft was
having with Head Office and the Steering Committee did not involve Telcom’s
Project Manager. Therefore, from Telcom’s point of view Smallsoft broke both
main communication links.

Actor bonds, such as trust and commitment, have been found in many studies to be
strong predictors of voluntary decisions to continue a relationship (e.g. Hocut, 1998;
Morgan and Hunt, 1994). Therefore, loosing trust and commitment, as seen from the
following quotation, is a major change in the elements of the relationship reducing
considerably the interdependency that keeps the parties together.

IT-expert, Telcom’s Head Office: I found that the whole software wouldn’t
have worked in 1997 because they had hard-coded the year 1996 to the
software. — There were a couple of other hard codes there so it wouldn’t have
been very flexible anymore. And of course you lose trust, this is 1996 and all
that should have been excluded from operations long ago. — It sort of stuck in
my mind that damn it, do we need to write down every little detail with these
people [Smallsoft], that this and this and this is what we demand or want from
our co-operation. This hadn’t been needed before so we didn’t know how to
prepare for it.

During the disengagement stage not only do the actor bonds deteriorate, but also
activity links and resource ties are cut off. Legal contracts may need to be annulled
(see the quotation) after which the buyer stops placing new orders and the supplier no
longer plans its production of goods or services according to the needs of the focal
customer. In other words, all the joint planning and activities that used to connect the
parties to each other cease to exist, and therefore the actual content of the relationship
vanishes.

Member of the Resolution Group, Telcom’s Head Office: [During the final
negotiations] we started to read the contract, what had been promised, did we get
everything that was promised or could we claim compensation for breach of
contract. — But we had accepted the application and Softpal had changed the
342 TÄHTINEN

software during the warranty period, so the clauses of contract were on their
[Smallsoft’s] side.

The Sensemaking and Aftermath Stage

The sensemaking and aftermath stage includes all the actions through which the
individual and/or group actors make sense of what has happened in the relationship.
These actions include thinking about and explaining previous behaviour in the relation-
ship, one’s own as well as those of others. This self-reflection is not just internal;
sensemaking can also take place in conversation with an insider or an outsider. Explaining
an event to someone is not just about stating the actions as they took place; at the same
time, the speaker recreates the event, as she/he perceives it at the moment as vividly
illustrated in the following quotation.
In addition, the purpose of this behaviour is to mentally protect the actors them-
selves, to prove that they did not make mistakes, to enable them to consider what they
could have done differently, and to reduce their cognitive dissonance related to the pro-
cess. Therefore, ‘storytelling’, i.e. spreading the news and the actors telling their own
side of things to others in the network are important elements of this stage. The
speaker may have a need to present the event to the listener in such a way that it does
not highlight her/his own mistakes in it, at the very least. In addition, the speaker may
receive information from the listener that helps her/him in the sensemaking. The
information may convince her of her own story, or it can also make her doubt her own
explanations and cause her to revise the story. It is difficult to separate storytelling
from self-reflection because they often happen simultaneously. Telling the story also
changes it, and the speaker can realise something new, which again helps her/him in
sensemaking.

IT-expert, Telcom’s Head Office: Of course from our part there was the
weakness at that point, at the end of last year, that there was no knowledge
of Smallsoft’s tools, so you had to believe. . . Of course, if they say it’s a
characteristic [of the tools], then all right. Then later on, I talked with Softpal
and others so they said that there are [Smallsoft’s tools have] these, these, and
these options and of course you thought, well, didn’t they [Smallsoft] swindle us
nicely.

Sensemaking behaviour influences the managers’ decisions and actions during the
dissolution process, but sensemaking continues also after the business relationship has
ended. Thus sensemaking turns into aftermath once the actors are aware of the immediate
outcome of the event. In the case at hand, the sensemaking that took place once the
complaint negotiations ended can be labelled the aftermath. If the outcome of the
dissolution process had not been a dissolved relationship, the story of the dissolution
phase would be a story of a serious but temporary difficulty in the relationship. Moreover,
some of the actions that were explained as faults of the opposite partner or as their lack of
trying their best would be explained differently in that situation.
THE PROCESS OF BUSINESS RELATIONSHIP ENDING 343

Sales Manager, Smallsoft: I think that at that point we should have, as a supplier,
stopped everything, got a new approval for this [application specification]
and developed the rest of it so that it could be taken into operational use.
Something radical should have been done at that point, then we would have
reached the goal roughly in the target schedule and expense frame. — The
situation looked like this or it looks like this only afterwards, back then it was
quite confusing as there were lots of complaints and defect lists and complaints
from both parties, from our workers and from the customer that things were totally
screwed.

Project Manager, Smallsoft: It was hard on the personal level, as these kind of
‘dramas’ aren’t normally played. Usually the problem people are changed and the
work continues, i.e. it is strictly business.

Head of the Steering Committee, Telcom: If I did this, if I had this same
situation again, I wouldn’t do any freaking data warehouse. That’s a fact.

The Enabling Stage

The enabling stage consists of all the actions that aim at lowering the exit barriers or
minimising the attenuating factors. From the viewpoint of an individual disengager, an
internal barrier may be e.g. the strong relational bonds of other powerful individuals
within the company who want to continue the relationship. On the other hand, a contract
may provide reasons that justify ending the relationship, although the real reasons may
not be related to any contract violations. Enabling actions also include developing new
relationships and securing network positions; these make it possible for the disengager
company to end the relationship without large losses or disruptions in the company
functions. Thus the enabling stage consists of actions to destroy the barriers in order to
make the ending possible.

Member of the Steering Committee, Telcom: In [software] tool comparison it


was brought up that in the future, the tool is not necessarily Smallsoft’s. Solidware
was already involved at that point. So it was speculated that if we were to expand
to a bigger data warehouse as originally planned, how much of the pilot could be
re-used there. And it was verified that not much, so we might as well change the
tools.

Smallsoft’s management had a good personal relationship with Telcom’s Head


office Information Technology Director. Smallsoft was able to maintain commit-
ment and trust within the personal relationship. Thus IT Director wanted to
continue the business relationship between Telcom and Smallsoft and because of
his position in the Head Office, he was able to influence the course of the focal
business relationship. However, the IT Director changed jobs and was no longer
responsible of the information technology. After this, the Resolution Group was
344 TÄHTINEN

able to make decisions about the relationships future without consulting the former
IT Director. Thus this change lowered the internal exit barrier and enabled the
customer to end its relationship with Smallsoft.

The Communication Stage

The actions in the communication stage encompass communication about the ending or
the continuation of the focal relationship. Compared to the Halinen and Tähtinen (2002)
model suggesting two communication stages – a dyadic and a network stage – only
one communication stage is proposed here. There are two reasons for this. Firstly, the
actor levels are a sufficient means of determining whether communication, or any other
action, is performed at dyadic or at network level. Secondly, the aims of dyadic com-
munication and network communication are different. Dyadic communication is about
the future of the relationship, i.e. discussions about exit and/or voice, and/or informing
the other party about the decisions made. The label ‘dyadic communication’ is thus too
restricting to describe the actions and the actors at this stage for two reasons. Firstly,
it suggests that the actions involve only the two parties of the relationship. This is too
narrow an interpretation, because the disengager company may well use a network
actor to convey the message to the soon-to-be ex-partner. Secondly, a network actor
can intentionally or unintentionally send exit signals to the partner, even without the
disengager company’s knowledge of it, as shown in the following quotations. There-
fore, the stage is labelled communication stage, referring to the communication that is
about the ending or the continuation of the focal relationship, regardless of the sender
of the message.

Sales Manager, Smallsoft: This [changing the supplier] was a very secretive
issue. The customer had forbidden Softpal to tell us and did not itself inform us
about Softpal coming aboard. Officially we heard about it only when the customer
and Softpal had already signed a contract and we were asked to transfer
knowledge to Softpal.

In the long and winding end the communication concerning the ending of the
relationship was mostly indirect. However, the case also shows that during the ending
process, the actors may change the type of communication strategy, if the first choice does
not provide expected results. In this case there was a movement from an indirect strategy
to a direct communication strategy, from signalling to attributional conflict (see Alajout-
sijärvi et al., 2000), which was already evident when Softpal was hired, and continued
during the formal complaint period.

The negotiations with Smallsoft and the Resolution Group did not advance
smoothly, and the group decided to express Telcom’s disappointment in a more
official way. With the help of Head Office lawyers, they drew up an official
complaint and sent it to Smallsoft. This was the first direct message from Telcom
and also from their Head Office that they were seriously considering ending their
THE PROCESS OF BUSINESS RELATIONSHIP ENDING 345

Figure 3. The stages and actor levels of long and winding end.

relationship with Smallsoft. As Telcom also announced that the next projects
depended on this first project ending successfully, the complaint left Smallsoft in
no doubt as to what would happen.

The Actors Levels in the Long and Winding End

The theoretical model (Halinen and Tähtinen, 2002) already suggested that different actor
levels would help in understanding the dissolution process in more detail. The empirical
grounding also underlines the same issue of analysing the change in different levels. We
can see the actor levels as nesting of entities into large entities (e.g. individuals forming a
company, two companies forming a relationship, several relationships forming a network)
creating a hierarchical system of levels (see Van de Ven and Poole, 1995). Change can go
on at all the levels, the outcome being an extinct relationship. The suggested actor levels
in business relationship ending process are individual, company, dyadic, and network
level.
Figure 3 shows the levels that in the long and winding end were active in each of
the process stages. The individual level was active in two stages, namely consideration
and sensemaking and aftermath stage. This refers to change in the individuals’ per-
ceptions of the relationship and of the partner company as well. When considering what
the future of the relationship would be, it is no longer automatically perceived as
continuing, i.e. the certainty of the relationship’s existence has changed in the mind of
the individual.
During sensemaking and aftermath, each individual thinks about the events and actions
in the relationship and while making sense of it, perceives them differently. Thus the two
stages are related to the individual’s perceptions of the relationship (which may be also
shared among the company, the relationship, and the network) and the changes in the
perceptions which also influence the decisions made at all stages.
The enabling stage differs from all the other ending stages since it is the single stage that
was not active on the dyadic level. This relates to the fact that in the case at hand, one
company acted as a disengager, i.e. it actively sought the goal of ending the relationship.
Certain events within the company (e.g. IT Director changing jobs) reduced the internal
346 TÄHTINEN

Figure 4. The actor levels in the disengagement stage of the long and winding end.

attenuating factors. Moreover, the customer company developed relationships with


alternative suppliers in order to reduce the network related attenuating factors. As the
ending was not a joint wish shared by both companies, no enabling events took place at
the dyadic level.
As for the rest of the suggested stages, disengagement, restoration, and communication
took place at company, dyad and network levels. Figure 4 takes a closer look at one stage
within the process, namely the disengagement stage and the events that took place during
that stage. During disengagement, events took place and actions were taken inside both
companies, together between the companies as well as involving third actors from the
network.

The Process of the Long and Winding End

Although all the entities mentioned above act during an ending process, not all of the
actions and events are equally important and influential in the change. Some actions within
the company, such as Telcom’s Project Manager trying to convince her superiors to end the
relationships already before Smallsoft changed its Project Manager, do not influence the
next organisational level, i.e. the company level. Therefore, to depict the path of change
more clearly, the following discussion takes a look at the critical events that changed at
least the company level.
THE PROCESS OF BUSINESS RELATIONSHIP ENDING 347

Figure 5. The long and winding end – main events in company, dyad, and network stages.

The process theories applied in explaining the change from an ongoing relationship to
an extinct one are teleological and dialectic. Thus in Figure 5, which shows the critical
events at the dyadic, relationship and network levels, both multiple progression, i.e. stages
going on simultaneously, and repeating strings of events appear. Although the change
is described by using stages, business relationship ending is not to be understood as
occurring as a succession of stages as in life-cycle theories (see Van de Ven, 1992).
However, the stages of the dissolution process explain what happens during the process,
although different paths may occur in different relationships.
One action and/or event may indicate more than one stage within the dissolution process.
For example, companies may decide to replace their contact persons to restore the
relationship (indicating restoration stage), but at the same time, this ends a personal
relationship between the companies and disconnects an important communication link
(indicating disengagement stage). Moreover, different actor levels may perceive the same
event differently, thus indicating different dissolution stages.
Although different paths of ending may occur, something can be said about the likely
order of the stages. It is logical to assume that the start of the process is an event or activity
during the consideration or disengagement stage, as they indicate a change in the
perception of the relationship or in the substance of the relationship. However, the trigger
event can also be a part of the communication stage if e.g. an influential network actor
changes its perception of the focal relationship and communicates this change to the
partner(s). In the long and winding end the communication and the enabling stages were
the last ones to start during the dissolution process. However, the stage that ended the
dissolution process was the sensemaking and aftermath stage, as the theoretical model
(Halinen and Tähtinen, 2002) also suggested.
348 TÄHTINEN

Propositions Derived from the Framework

Based on the empirical grounding of the framework and the discussion concerning the
elements of the framework, the following section will bring forward a set of propositions
related to the different stages of the ending process. The propositions are derived from
the framework to encourage future research but also to portray the connections of the
elements e.g. in situations where management wishes to save a business relationship
from ending.

Consideration stage related propositions:

As seen in the long and winding end, individual level considerations and attempts to end
the relationship failed, but network level considerations led to the ending of the
relationship. However, if the individual had been e.g. the CEO of the customer company,
her/his position could have made a difference. Thus three propositions arise:

P1: Individuals with strong power within their own organisation can decide solely the
future of a particular business relationship.
P2: Individuals lacking power within their own organisation can influence others in the
organisation to consider ending a particular business relationship.
P3: The higher the number of individuals within an organisation who consider a
particular relationship as a candidate for ending, the higher the likelihood of the
ending.

Restoration stage and disengagement stage related propositions:

The case illustrated that unless both sides of the relationship perceive certain events or
actions as restoring, they will not produce results. Joint communication about the aims of
the restoring actions may produce such a joint perception, if the actions as such can be
understood in many ways. A common understanding within the companies also helps to
avoid misinterpretations. An example from the case at hand was Smallsoft’s discussions
with Telcom’s IT Director, during which Smallsoft tried to resolve the tension in the
relationship, but which were perceived by a number of people in Telcom as overriding
them. All in all, in order to produce results, restoring actions also seem to need a certain
level of trust and commitment between the companies to be perceived as positive
actions.

P4: If there are conflicting views within an organisation of its commitment, the restoring
actions in the relationship are less successful.
P4: If there are conflicting views within an organisation of its trust, the restoring actions
in the relationship are less successful.
P5: If either of the organisations loses its commitment, restoring actions are not
successful.
P6: If either of the organisations loses its trust, restoring actions are not successful.
THE PROCESS OF BUSINESS RELATIONSHIP ENDING 349

P7: If the communication links between the two organisations in a business relationship
are not open, the restoring actions are less successful.

Sensemaking and aftermath stage related propositions:

In the long and winding end, during the sensemaking and aftermath stages several events
took place at the network level. This was partly due to the fact that several network actors
were closely connected to the focal relationship. In addition, the relationship was not free
from problems, and there was an increased need to make sense of the ‘mess’ with the
help of other actors from the network. Van de Ven and Poole (1995) state that persons
undergoing a constructive mode of change may experience a high degree of uncertainty,
as the change is unpredicted, and therefore the need to make sense of the changes
increases. Moreover, the sensemaking and aftermath stage was reasonably long, shown
by the fact that the individuals interviewed had no problems in recalling the case,
although they were involved in several relationships and software projects both before
and after this one.

P8: The more network actors there are directly connected to the relationship, the more
intense is the sensemaking and aftermath stage.
P9: The more dissatisfied the individuals within the relationship are, the more intense is
the sensemaking and aftermath stage.
P10: The more dissatisfied the individuals within the relationship are, the longer the
sensemaking and aftermath stage takes.

The enabling stage related propositions:

The enabling stage is a new stage that emerged from the empirical data. In the long and
winding end the process was not quick and simple. This was due to strong internal and
external attenuating factors, which were perceived by managers as hindering the
relationship ending. However, it is difficult to imagine a situation with no attenuating
factors present in a business relationship, since the links, ties and bonds as such can be
seen as attenuating the relationship’s ending. Thus it is logical to propose that:

P11: The more attenuating factors there are, the more intense is the enabling stage.
P12: The more attenuating factors there are, the longer the enabling stage takes.

The communication stage related propositions:

As for the communication stage, the two studies focusing on it (Alajoutsijärvi et al.,
2000; Giller and Matear, 2001) present reflections concerning the circumstances which
increase the importance of managing the communication stage beautifully. However, the
network level as the sender of messages concerning the future of a focal relationship
saw the light in this study. For management this means that it is not enough to pay
attention to your own actions and those of your partner, but also of the network actors
350 TÄHTINEN

towards your partner. Following the logic in sensemaking and aftermath stage related
proposition no 8;

P13: The more network actors there are directly connected to the relationship, the more
probable it is that they may be active during the communication stage.

Discussion

The stages of business relationship ending process provide a detailed view of the dynamics
of relationship ending. Instead of having to talk about the ending phase as a black box we
can refer to the stages, i.e. to different action periods within the phase. Moreover, the model
gives support to the notion that the end of a relationship is not similar to the beginning of
the relationship, but needs its own conceptual language, if it is to be understood.
The process framework of relationship ending presented here is one of the few models to
have applied a processual perspective. The compiling of stages includes individual,
department, company, dyadic, and network levels, thus reflecting the view of structural
embeddedness (see Halinen and Törnroos, 1998) or nesting (see Van de Ven and Poole,
1995) underlying the process. The model also acknowledges temporal embeddedness,
i.e. the different time periods that can be used as a tool to analyse the process, or any
process. By analysing both the different periods (actions, events, stages and phases) and the
different actor levels, the model strives for high specificity.
The ending process model highlights actors’ actions and the fact that these actions drive
the process, but may also stop or stall it. Restoring actions are as important in the process
as the terminating ones. It is also emphasised that not every ending travels through each
stage and certainly not in a prescribed order. The case strongly suggests that multiple paths
are possible, some stages may go on at the same time and the process may also contain
loops backwards. Thus the model is flexible enough to allow for the complexity of these
processes to be discovered.
This study creates an understanding of how relationships end and how they can be
ended, and also how they could be restored. Knowledge about the process and its stages
helps management to look for early and vague signals of the ending process. If these are
detected early enough, decisions can be made as to whether to try to restore the
relationship or to prepare the company for its ending. Moreover, a company may wish
to terminate certain unwanted relationships. In any of these situations, proper management
is needed in order to avoid negative consequences to any of the actors involved (see also
Alajoutsijärvi et al., 2000).
Even if the relationship ending is unwanted, something good may come out of it. The
aftermath stage, if used consciously, forms an ideal occasion to turn even the most
unpleasant ending into a useful learning experience for the whole company. There is often
as much to learn from failures as from success stories. At best, the lesson learned will help
the company manage other relationships and potential ending processes.
As this research presents an empirically grounded framework of business relationship
dissolution in software business, the obvious avenue for future research is an examination
THE PROCESS OF BUSINESS RELATIONSHIP ENDING 351

in other industries. I do, however, assume that the model would need to be modified if it
were applied in a different context, as context and action are interwoven.
The different stages of the dissolution process; i.e. the communication, consideration,
disengagement, enabling, restoration, and the sensemaking and aftermath stages, all de-
serve dedicated research in their own right. The communication stage, for example, is one
of the turning points in the dissolution process, and therefore it merits extra attention.

Acknowledgements

The authors wishes to thank the reviewers for their valuable guidance as well as the
financial support of the LIIKE Research Programme funded by the Academy of Finland
and the National Technology Agency Tekes, Finland.

Notes

1. Business or economic exchange, communication or social exchange, coordination and adaptation have
proved central to business interaction (e.g. Möller and Wilson, 1995).
2. The substance of a business relationship can be identified in three different layers; activity, resource, and actor
layers (Håkansson and Snehota, 1995). In the activity layer the various internal activities (e.g. production,
R&D activities) of the focal companies are connected or linked. The resource layer connects various resource
elements needed and controlled by the focal companies (e.g. knowledge, technology), and thus the resource
ties are being developed. The actors are connected with bonds (e.g. trust, commitment, personal relationships),
affecting the way the actors perceive, evaluate and treat each other.
3. Development is seen as a neutral term, thus meaning both positive and negative developments, growing and
declining etc.

References

Alajoutsijärvi, Kimmo, Möller, Kristian, and Tähtinen, Jaana (2000), ‘‘Beautiful Exit: How to Leave Your
Business Partner,’’ European Journal of Marketing, vol. 34, no. 11/12, pp. 1270 – 90.
Baxter, Leslie A. (1985), ‘‘Accomplishing Relationship Disengagement.’’ In S. Duck and D. Perlman (eds.),
Understanding Personal Relationships. An Interdisciplinary Approach, London: Sage, pp. 243 – 65.
Denzin, Norman K. (1978), The Research Act: A Theoretical Introduction to Sociological Methods, New York:
McGraw-Hill.
Duck, Steve (1982), ‘‘A Topography of Relationship Disengagement and Dissolution.’’ In: S. Duck (ed.),
Personal Relationships 4: Dissolving Personal Relationships, 1 – 30. London: Academic Press.
Dwyer, F. Robert, Schurr, Paul H., and Oh, Sejo (1987), ‘‘Developing Buyer-Seller Relationships,’’ Journal of
Marketing, vol. 51, no. 2, pp. 11 – 27.
Fiocca, Renato (1982), ‘‘Account Portfolio Analysis for Strategy Development,’’ Industrial Marketing
Management, vol. 11, no. 1, pp. 53 – 62.
Ford, David (ed.) (1990), Understanding Business Markets. Interaction, Relationships, Networks, London:
Academic Press.
Gadde, Lars-Erik, and Mattsson, Lars-Gunnar (1987), ‘‘Stability and Change in Network Relationships,’’
International Journal of Research in Marketing, vol. 4, no. 1, pp. 29 – 41.
352 TÄHTINEN

Giller, Caroline, and Matear, Sheelagh (2001), ‘‘The Termination of Interfirm Relationships,’’ Journal of Business
& Industrial Marketing, vol. 16, no. 2, pp. 92 – 112.
Grønhaug, Kjell, Henjesand, Inge Jan, and Koveland, Anita (1999), ‘‘Fading Relationships in Business Markets:
and Exploratory Study,’’ Journal of Strategic Marketing, vol. 7, no. 3, pp. 175 – 90.
Grönroos, Christian (1997), ‘‘Value-Driven Relational Marketing: from Products to Resources and Competen-
cies,’’ Journal of Marketing Management, vol. 13, no. 5, pp. 407 – 19.
Halinen, Aino, and Tähtinen, Jaana (1999), Towards a Process Theory of Relationship Dissolution, Publications
of the Turku School of Economics and Business Administration, Discussion and Working Papers 9:1999.
Halinen, Aino, and Tähtinen, Jaana (2002), ‘‘A Process Theory of Relationship Ending,’’ International Journal of
Service Industry Management, vol. 13, no. 2, pp. 163 – 80.
Halinen, Aino, and Törnroos, Jan-Åke (1998), ‘‘The Role of Embeddedness in the Evolution of Business
Networks,’’ Scandinavian Journal of Management, vol. 14, no. 3, pp. 187 – 205.
Havila, Virpi (1996), ‘‘When Does an International Business Relationship Become Non-Existent?’’ In
Proceedings of the 25th EMAC Conference, Budapest, Hungary.
Havila, Virpi, and Wilkinson, Ian (2002), ‘‘The Principle of the Conservation of Relationship Energy: or Many
Kinds of New Beginnings,’’ Industrial Marketing Management, vol. 31, no. 3, pp. 191 – 203.
Heide, Jan B., and Weiss, Allen M. (1995), ‘‘Vendor Consideration and Switching Behavior for Buyers in
High-Technology Markets,’’ Journal of Marketing, vol. 59, July, pp. 30 – 43.
Hirschman, Albert O. (1975), Exit, Voice and Loyalty. Responses to Decline in Firms, Organizations, and States.
4th reprint, Cambridge: Harward University press.
Hocutt, Mary Ann (1998), ‘‘Relationship Dissolution Model: Antecedents of Relationship Commitment and
the Likelihood of Dissolving a Relationship,’’ International Journal of Service Industry Management, vol. 9,
no. 2, pp. 189 – 200.
Håkansson, Håkan, and Snehota, Ivan (eds.) (1995), Developing Relationships in Business Networks, London:
Routledge.
Järvelin, Anne-Mari (2001), Evaluation of Relationship Quality in Business Relationships. University of
Tampere, PhD thesis, Acta Universitatis Tamperensis 794.
Keaveney, Susan M. (1995), ‘‘Customer Switching Behavior in Services Industries: an Exploratory Study,’’
Journal of Marketing, vol. 58, April, pp. 71 – 82.
Möller, Kristian, and Wilson, David D. T. (eds.) (1995), Business Marketing: An Interaction and Network
Perspective, Boston: Kluwer.
Morgan, Robert M., and Hunt, Shelby D. (1994), ‘‘The Commitment-Trust Theory of Relationship Marketing,’’
Journal of Marketing, vol. 58, no. 3, pp. 20 – 38.
Pettigrew, Andrew M. (1990), ‘‘Longitudinal Field Research on Change: Theory and Practice,’’ Organization
Science, vol. 1, no. 3, pp. 267 – 92.
Ping, Robert, and Dwyer, F. Robert (1992), ‘‘A Preliminary Model of Relationship Termination in Marketing
Channels.’’ In G. L. Frazier (ed.), Advances in Distribution Channel Research, vol. 1, Greenwich: Jai Press,
pp. 215 – 33.
Roos, Inger (1999), ‘‘Switching Processes in Customer Relationships,’’ Journal of Services Research, vol. 2,
no. 1, pp. 68 – 85.
Rosson, P. J. (1986), ‘‘Time Passages: The Changing Nature of Manufacturer-Overseas Distributor Relations in
Exporting,’’ Industrial Marketing and Purchasing, vol. 1, no. 2, pp. 48 – 64.
Sheth, Jagdish N., and Parvatiyar, Atul (1995), ‘‘The Evolution of Relationship Marketing,’’ International
Business Review, vol. 4, no. 4, pp. 397 – 418.
Simmel, Georg (1950), ‘‘Qualitative Aspects of the Group. A translation of Chapter II of Soziologie,’’
In K. H. Wolf (ed.), The Sociology of George Simmel, Glengoe: Free Press of Glengoe, pp. 87 – 177.
Tähtinen, Jaana (1999), The existence and the dissolution of a business relationship in tailored software business.
A theoretical framework. Department of Economics, University of Oulu, Research reports No. 39. Oulu. Finland.
Tähtinen, Jaana (2001), The Dissolution process of a business relationship. A case study from tailored software
business. Department of Marketing, University of Oulu, Acta Universitatis Ouluensis, G 10.
Tähtinen, Jaana, and Halinen-Kaila, Aino (1997), ‘‘The Death of Business Triads. The Dissolution Process of a
Net of Companies.’’ In Proceedings of the 13th IMP Conference, Lyon, France, pp. 553 – 90.
THE PROCESS OF BUSINESS RELATIONSHIP ENDING 353

Tähtinen, Jaana, and Halinen, Aino (2002), ‘‘Research on Ending Exchange Relationships: A Categorization,
Assessment and Outlook,’’ Marketing Theory, vol. 2, no. 2, pp. 165 – 88.
Van de Ven, Andrew H. (1992), ‘‘Suggestions for Studying Strategy Process: A Research Note,’’ Strategic
Management Journal, vol. 13, summer, pp. 169 – 88.
Van de Ven, Andrew H., and Poole, Marshall Scott (1995), ‘‘Explaining development and change in organisa-
tions,’’ Academy of Management Review, vol. 20, no. 3, pp. 510 – 40.

Jaana Tähtinen is Acting Professor at the University of Oulu, Finland, where she also received her Ph.D in 2001.
Her thesis deals with the dissolution process of a business relationship in the tailored software business. Before
joining the faculty she worked within the industry and service sector as well as a director in a local science centre.
She has published in the European Journal of Marketing, Industrial Marketing Management, International Journal
of Service Industry Management, and Marketing Theory.

S-ar putea să vă placă și