2 pgs-2-

Investigating the role of business processes and knowledge management systems on performance: A multi-case study approach

Qing Cao*, Mark A. Thompson and Jason Triche

Texas Tech University, Rawls College of Business, Lubbock, United States

(Received 5 February 2012; final version received 20 March 2013)

In the ever-changing and competitive market place, organisations continuously need to improve their competitive advan- tage. One method to accomplish this is to form collaborative networks. Both knowledge management (KM) and KM systems play a pivotal role in the success of collaborative networks since information sharing and knowledge assets are so critical to the network. There has been a vast amount of research on KM systems but very little is known about how it affects individual and organisational performance. Drawing on the task–technology fit theory in this study, we explore the fit or alignment between business process (task) and KM systems (technology) and its impact on KM systems utilisa- tion based on multiple case studies. Subsequently, we investigate the impacts of both the task–technology fit and KM systems utilisation on individual and business performance. This paper contributes to the collaborative network/KM liter- ature in several ways. First, it extends the task–technology fit theory to an important context of collaborative network/ KM. Second, it replaces task with business process, which has the potential to help explain KM systems’ success on business performance. Third, the paper explores the positive impact of task–technology fit on KM system utilisation and business performance. Fourth and finally, the study provides insight into the future development of KM systems and how to better align them with managerial purposes.

Keywords: knowledge management; collaborative network; task–technology fit; knowledge management systems; business process; business performance

1. Introduction

Complex business activities in organisations may lead to the formation of a collaborative network to minimise the impacts of market fluctuation and dynamic operational behaviour by effective inter-organisational collaboration and coordination (Jagdev and Thoben 2001). Both knowledge management (KM) and KM systems play a pivotal role in the success of a collaborative network since information sharing and knowledge assets are two main pillars of the network.

Knowledge is a powerful resource in helping organisations as well as individuals preserve their identity, culture, best practices, and core competencies. Managing this knowledge has become an important topic to both industry and aca- demics. The topic of KM has been around for two decades, while the practice of KM is still evolving and changing for both practitioners and researchers alike. The importance and influence of KM is evident by the plethora of research arti- cles dedicated to the topic (see Alavi and Leidner 2001; Schultze and Leidner 2002; Tanriverdi 2005; Gunasekaran and Ngai 2007; Nachiappan, Gunasekaran, and Jawahar 2007; Wang, Klein, and Jiang 2007). KM systems, on the other hand, are the information technology used by an organisation to capture, represent, and apply knowledge to itself and its collaborative network (Dhaliwal and Benbasat 1996). In other words, KM systems are viable tools for achieving effective knowledge management.

Substantial investments have been made to KM system technologies and in 2007 KM software was a $73 billion market (AMR Research 2007). Despite the high expenditures in KM, some researchers note the failure rate of KM pro- jects at around 50% (Peyman, Jafari, and Fathian 2005). KM approaches fail when they do not integrate human resources, processes, and technology regardless of how much money companies spend. The most common error in KM implementation is failing to coordinate efforts between information technology and human resources (Ambrosio 2000). The KM effort should not be a technology issue or a people issue, but instead should be a joint effort.

Although there is a general consensus of the organisational settings in which KM systems can be applied, very little is known about how to enhance business processes or how to measure KM systems’ impacts on business performance (Kulkarni, Ravindran, and Freeze 2007). Such knowledge is crucial, however, for a realistic determination of the

*Corresponding author. Email: [email protected]

International Journal of Production Research, 2013 Vol. 51, No. 18, 5565–5575, http://dx.doi.org/10.1080/00207543.2013.789145

� 2013 Taylor & Francis

opportunities and requirements presented by KM systems. We endeavour to fill the void from both managerial and academic perspectives in the KM systems arena.

This paper contributes to the collaborative network/KM literature in several ways. First, it draws on the task–tech- nology fit theory (Goodhue and Thompson 1995) to explore the alignment between business process (task) and KM systems (technology) and its impact on KM system utilisation. Subsequently, we investigate the impacts of both the task–technology fit and KM system utilisation on individual and business performance via multiple case studies. By extending the task–technology fit theory to an important context of collaborative networks/KM, we replace task with business process to help explain KM systems’ success on business performance. Thus, the study builds contingency theory in collaborative networks/KM by moving beyond the question of mere tool usage (KM systems) to the match between KM tool and business process. As such it presents a viable framework for practitioners to effectively implement KM systems. Finally, the study provides insight into the future development of KM systems on how to better align them with managerial purposes. The development of more appropriate and useful KM tools would seem to be a critical enabler of improved business performance.

The paper proceeds as follows. In the next section, we review the literature on collaborative networks, KM, and KM systems. In the theoretical background section, we propose a theoretical approach that is appropriate for the study of KM systems, namely, the task–technology fit framework. Then, we explain our research model based on this framework, develop propositions, and describe our case studies and analytical procedures for the study. Finally, our results section presents support for each of our propositions. The lessons learned section highlights the contributions that we make by providing a framework for research into the collaborative networks/KM systems arena and discusses possible limitations to our study along with several avenues for future research.

2. Literature review

In the ever-changing and competitive market place, organisations continuously need to improve their competitive advantage. One method to accomplish this is to form collaborative networks both within the organisation and along an organisation’s supply chain. ‘A collaborative network is constituted by a variety of entities (e.g., organisations and peo- ple) that are largely autonomous, geographically distributed, and heterogeneous in terms of their: operating environment, culture, social capital, and goals.’ (Camarinha-Matos and Afsarmanesh 2005, 439). The study of collaborative networks is relatively new and it draws on multiple disciplines covering computer science, economics, management, operations management, and information systems. The formation, operation, and success of collaborative networks depend on a common base among its members, such as common goals, common IT infrastructures and supporting services, real-time information sharing, common standards, and common views of business processes (Camarinha-Matos and Afsarmanesh 2003, 2005). Since real-time information sharing and the needed baseline of business processes are crucial in collabora- tive networks, both KM and subsequently KM systems play a vital role in achieving an effective collaborative network. Effective KM systems are critical to the success of a collaborative network.

There are a number of different definitions of KM and KM systems and these concepts are difficult to define (Earl 2001). According to Schultze and Leidner (2002, 218) KM is defined as the ‘generation, representation, storage, transfer, transfor- mation, application, embedding, and protecting of organisational knowledge.’ A more concise definition of KM is to identify and leverage the collective knowledge in an organisation to help the organisation compete (Alavi and Leidner 2001).

The practice of KM in itself is broad and covers many topics including business processes, business practices, concepts, frameworks, methodologies, tools, and architecture. For example, the methodologies and frameworks include knowledge creation (Nonaka and von Krogh 2009), knowledge assets (Wiig 1997; Wilkins, van Wegen, and De Hoog 1997), intellectual capital (Liebowitz and Wright 1999), strategy management (Drew 1999; Hendriks and Vriens 1999), systems thinking (Rubenstein-Montano et al. 2001), artificial intelligence (Liebowitz 2001), and knowledge inertia (Liao 2002). Likewise, KM is also used across all different industries including oil and gas (Preece et al. 2001), manufacturing (Paiva, Roth, and Fensterseifer 2002), government (Liebowitz 1999), agriculture (Kristjanson et al. 2009), high-tech firms (Collins and Smith 2006), and health care (Bose 2003). In fact, a recent survey of 342 managers on whether or not they are knowledgeable about their company’s usage of knowledge management and collaborative technologies indicated that 47% of companies have formal knowledge management initiatives or are planning them (Currier 2010). Given the breadth of different methodologies, frameworks and industries that utilise KM, research topics are vast and numerous.

However, KM systems, much like KM, are also difficult to define and there are a number of different definitions. One definition of KM systems is that it captures, represents, and applies expert knowledge to an organisation (adopted from Dhaliwal and Benbasat 1996). Basically KM systems refer to a class of information systems that manage organisational knowledge (Alavi and Leidner 2001). As mentioned before, KM systems are employed across multiple industries, in multiple ways, by a majority of companies in today’s ever competitive marketplace.

5566 Q. Cao et al.

The current KM system literature can be grouped into four knowledge management processes: (1) knowledge creation, (2) storage/retrieval, (3) transfer, and (4) application/use (Alavi and Leidner 2001). Our research focus is to study and analyse the application and use of KM systems on utilisation and individual and organisational business per- formance. We draw on the task–technology fit theory (Goodhue and Thompson 1995) and extend it to the collaborative network/KM systems domain. To the best of our knowledge, this is the first attempt to measure collaborative network/ KM systems using task–technology fit. We employ a multiple-case study approach using three different organisations in three different industries.

3. Theoretical background

3.1 Task–technology fit

Goodhue and Thompson (1995) propose a model where tasks (defined as actions carried out by individuals in turning inputs into outputs) and technology (defined as tools used by individuals in carrying out these tasks) predict a construct called the task–technology fit. The task–technology fit is defined as the degree to which technology assists in performing the respective tasks. This construct, combined with whether an individual utilises the technology, determines the impact on an individual’s performance. Utilisation of the technology may be either voluntary or mandatory for the individual.

Task–technology fit theories are contingency theories that argue that the use of technology may result in different outcomes depending on the task that it is used for (Goodhue and Thompson 1995). This theory proposes that if a technology is utilised and it is a good fit with the task it supports, then the technology will have a positive impact on individual performance. While this model has been extended by Zigurs and Buckland (1998) to test the effect of different types of tasks and technology on group performance, the theory has also been applied to several other areas of information system research over the last 15 years. For example, the theory has been used to study how virtual teams can match communication technologies to different types of interpersonal interactions (Maruping and Agarwal 2004) as well as to study the effects, adoption, and impacts of mobile commerce (Gebauer and Shaw 2004; Lee, Cheng, and Cheng 2007). Other ways the theory has been studied is in the use of simulation training for the military (Cane, McCarthy, and Halawi 2010), the use of the web as an information source for international travel (D’Ambra and Wilson 2004), the ease and use of user interfaces (Mathieson and Keil 1998), and the study of information technology in mana- gerial decision making (Ferratt and Vlahos 1998).

3.2 Proposition development

We extend the task–technology fit framework by Goodhue and Thompson (1995), where business process represents tasks and KM systems represent technology. As such, fit is defined as the degree to which KM systems assist an indi- vidual or organisation in performing their business process. The business process–KM systems fit construct combined with whether an individual utilises a KM system determines the impact on an individual or organisation’s performance. Our research model is described in Figure 1.

KM systems represent the technology aspect for our model. As defined by Goodhue and Thompson (1995), technol- ogy comprises tools used by individuals in carrying out their tasks. We posit that a KM system is a tool that manages organisational knowledge and is used by individuals and organisations to perform and facilitate tasks. As such, a KM system facilitates several elements of a collaborative network, that is, the ability to integrate data for various users and to search for content (Camarinha-Matos and Afsarmanesh 2005).

Business process

KM systems

Business process-KM system Fit

Utilisation

Individual and organisational performance

Figure 1. Research framework.

International Journal of Production Research 5567

A business process is an activity or set of activities that will accomplish a specific organisational goal. Business processes are defined as any activity or group of activities that takes one or more inputs, transforms them, and provides one or more outputs for its customers (Krajewski, Ritzman, and Malahortra 2010). In the task–technology fit model, Goodhue and Thompson (1995) define tasks as actions carried out by individuals in turning inputs into outputs. We claim that a business process contains one or more tasks as defined by Goodhue and Thompson (1995). As such, in our research model, we substitute business process in place of task.

In Figure 1, business process consists of three characteristics: non-routineness, interdependence, and job title. A business process is non-routine if there are a large number of exceptions and search is not logical or analytical (Perrow 1967; Thompson 1967; Goodhue 1995; Goodhue and Thompson 1995). Therefore, a non-routine business process is one where individuals deal with ill-defined business problems or ad hoc business problems. In most organisations the knowledge exists, either explicit or tacit, on how to deal with such non-routine business processes but an individual will need a way to access that knowledge for it to be beneficial. Interdependence is defined as dependence with other organi- sational units (Perrow 1967; Thompson 1967; Goodhue 1995; Goodhue and Thompson 1995). Some business processes involve multiple organisational units and knowledge exists across these different units. In such cases, a KM system can aid an individual in finding cross-organisational knowledge. The last characteristic of the business process construct is job title, which is a pragmatic proxy to capture the differences of employee levels ranging from clerical users to high- level managers (Goodhue and Thompson 1995). Different levels of an organisation use a KM system to find content for different business processes. For example, a clerical employee may use a KM system to find content for ordering sup- plies from a specific supplier and a high-level manager may use a KM system to find content on how to conduct vendor selections. While both employees are using the KM system in the same way, they are looking for different answers.

Another construct in Figure 1 is KM systems which consist of two characteristics: systems used and departments (Goodhue and Thompson 1995). These characteristics allow us to analyse the data from the case study by grouping departments and number of systems used. In order to measure the effect of a KM system, we need to understand how many other systems an individual uses. For the department characteristic, we use it as a proxy measure to capture the potentially different levels of attention paid by IS departments (Goodhue and Thompson 1995).

In our research model, we align the business process construct and the KM system construct to form the business process–KM system fit. The business process–KM system fit is defined as the degree to which KM systems assist an individual or organisation in performing business processes. The business process–KM system fit construct consists of eight different factors adopted from Goodhue and Thompson (1995) with modifications to fit our case: data quality, loc- atability of data, authorisation, compatibility between systems, production timeliness, training and ease of use, system reliability, and relationship with user. The first five factors focus on using content in facilitating business processes. For example, data quality is measured by the detail and currency of the content (Goodhue and Thompson 1995). The con- tent in a KM system must be kept up to date and old content must be systemically purged. The content must also be at the right level of detail as too much detail can complicate an issue and not enough detail can obscure an issue. Locat- ability of content refers to the ease of determining what content is available and the ease of determining what the con- tent means (Goodhue and Thompson 1995). From a KM system perspective, locatability is a measurement for how easy it is to find content in a KM system even on issues that rarely occur. Authorisation is defined as access to content that is necessary for a participant to do his or her job (Goodhue and Thompson 1995). Depending on the employee level, content may be restricted in a KM system. External customers may have access to only a small percentage of an organi- sation’s content, clerical employees may have access to a little more content, and so forth. This measurement verifies if restricted content in a KM system is given to individuals or groups who need it in order for them to carry out their job responsibilities. Compatibility is defined as the degree to which content from different IT systems can be consolidated or compared without inconsistencies (Goodhue and Thompson 1995). Although organisations strive to consolidate con- tent into one KM system, as seen from our case study, content exists in multiple systems across an organisation. Com- patibility verifies if the content in the different systems are consistent. The fifth factor that focuses on using content in facilitating business processes, ease of use and proper training, is defined as the ease of using the KM system and access to the proper amount of training to use the system (Goodhue and Thompson 1995). Like any IT system, a KM system must be easy to use and, if not, adequate training should be provided to use the KM system.

The next two factors, production timeliness and system reliability, focus on meeting day-to-day operations and the last one, relationship, focuses on responding to changing business needs (Goodhue and Thompson 1995). Timeliness is defined as the degree to which the IT department meets its pre-defined production turnaround schedules (Goodhue and Thompson 1995). A KM system is supported by either an internal IT department or by specialised vendors and these groups set a maintenance schedule for upgrades and/or updates to the KM system. Timeliness measures how well these groups meet their maintenance schedule, while reliability is defined as the dependability and consistency of access and uptime of a KM system (Goodhue and Thompson 1995). As is true with any IT system, in order to use a KM system,

5568 Q. Cao et al.

the system must be up with little downtime, dependable, and free of major errors and defects. The last factor, relationship with user, is defined as how well the IT department understands the business customer’s day-to-day opera- tions, supports the business customer, and provides adequate turnaround to the business customer’s needs (Goodhue and Thompson 1995). This definition also applies to a specialised vendor if they host the KM system. These eight compo- nents of the fit construct are used to measure its effectiveness. Therefore our first proposition is as follows:

Proposition 1: The business process–KM system fit will be affected by business process characteristics or KM system characteristics.

The fit construct combined with whether an individual utilises the KM system determines the impact on an individual or an organisation’s performance. Utilisation of the technology may be voluntary or mandatory for the individual or group. Utilisation is measured by a user’s perceived dependence on the KM system and finally the (perceived) individ- ual and organisational performance impact.

Individual and organisational performance impact relates to the accomplishment of a portfolio of tasks by an individ- ual or group (Goodhue and Thompson 1995). As demonstrated in Goodhue and Thompson’s (1995) research, performance impacts are a function of both task–technology fit and utilisation and not just utilisation. In the spirit of this research, we posit the fit of business process and KM systems will influence the effectiveness of collaborative networks on individual and organisational performance. Therefore, adopting the task–technology fit model proposed by Goodhue and Thompson (1995), our second and third propositions are as follows:

Proposition 2: User evaluations of business process–KM system fit will positively influence the utilisation of KM systems by individuals.

Proposition 3: User evaluations of business process–KM system fit and utilisation will positively influence perceived individ- ual and organisational performance impacts respectively.

In the next section, we describe the case study methodology that we employed.

3.3 Methodology

We pursue an explanatory case study method as opposed to what is commonly referred to as an exploratory or descrip- tive study. Since we are investigating causal explanations rather than initial descriptions of some occurrence or an event, we developed theoretical propositions before collecting data (Yin 2009). One advantage to our case study approach is it allows us to understand a complex organisational phenomenon or processes within a real-world setting by identifying how and why particular phenomena or events take place (Yin 2009). Alternatively, other methods such as surveys may provide greater control but may also be limited in their context or ability to find alternative explanations (Yin 2009).

We chose to conduct a multiple-site case study with the organisation as the primary unit of analysis. Since multiple sites were selected, our findings may be more robust than if we had chosen just a single case study (Yin 2009). As such, we collected data from a major business communication provider, a networking infrastructure solutions provider, and a GPS technology provider (referred to hereafter as BCP, NIS, and GPS, respectively).

BCP is a global organisation that has received numerous leadership, customer service, and communication awards. They provide products, services, and support to a global market either directly or through partner channels. Their client base is vast and ranges across multiple markets. At BCP, we interviewed two employees, an IT manager/solution archi- tect and a global business owner of web and knowledge management. The two interviewees use the KM system but are from different departments, have different job responsibilities, and are located in different parts of the country.

NIS is a global organisation that has received numerous awards including product of the year, best practice, hottest growth, and fastest growing company, to name a few. They provide products through resellers, distributors, and original equipment manufacturer channels. At NIS, we interviewed a senior manager of technical support in the global services division. The interviewee uses a KM system to perform some of his duties.

The third company, GPS, is a global leader in GPS technology, products, and services. GPS has also won many best product and innovation awards. Their clients range from multinational companies to individual consumers. At GPS, we interviewed one employee, the outbound logistics manager.

We selected these specific organisations for several reasons. First, these three companies are competing in different markets, with different products, services, and customers. This allows us to draw comparisons between different compa- nies and their KM systems post-adoption and performance. Second, these three companies use a KM system to support their business strategy. The KM system is an integral part of business operations and this allows us to study the KM

International Journal of Production Research 5569

system in routine and non-routine business processes. Lastly, these three organisations were chosen because we had access to managers within these companies that are actual users of the KM system. The different managers allowed us to study the KM system post-adoption across different business processes. That is, these managers are actual users of the system as opposed to executives who do not use the system but are concerned with how the KM systems align with their strategic goals. In addition, another advantage to using these managers is that they often represent departments or areas where there are multiple users or analysts. As such, our interviews represent problems that are user specific but also contain elements of multiple users. In particular, we interviewed one employee each from NIS and GPS and two employees from BCP, where we asked general questions about job responsibilities and length of employment at their respective organisations. We then followed up with more specific questions about KM system use, finding KM solutions, relationships with the IT department, and handling customer problems. Interviews were conducted on a confidential basis but were recorded with the employee’s permission. Afterwards, we transcribed the recording and analysed each of the interviews. Finally, we went through each transcript and sought to identify the theoretical concepts from our propositions in the raw data.

Throughout this process, we use the definitions of the dimensions of business process–KM system fit from prior lit- erature on task–technology fit (Goodhue and Thompson 1995). Using similar definitions from prior literature helps ensure that we will be able to accurately identify the relevant information from our case studies. One question that often arises in case studies is whether one’s findings are applicable outside of the study context. Another question with respect to case studies has to deal with reliability. Reliability refers to the stability, accuracy, and precision of measurement. As such, we have documented our procedures and developed a case study protocol so that this work can be replicated. To further increase our study’s reliability, we developed a case study database that includes interview recordings, transcripts, and previous literature that was used to develop our constructs and interview questions.

In the next section, we discuss our case study findings in the context of the propositions from our business process–KM system research framework.

4. Results and discussion

In the previous section, we described the methodology employed to examine our propositions about the business process–KM system fit. This section explains how we used our propositions on business process and KM system characteristics as well as business process–KM system fit to seek evidence related to these propositions. We identify direct quotations from our interviews that align with our propositions. The case studies are presented in line with the different dimensions of our propositions.

Proposition 1 is related to how business process–KM system fit is affected by both business process characteristics and KM system characteristics. Business process characteristics have been broken down into three dimensions (Goodhue and Thompson 1995): non-routineness, interdependence, and job title. For example, our informants from BCP, NIS, and GPS discussed several issues with respect to non-routineness within the business process. In the case of BCP, one exam- ple of a non-routine issue that arose was customer service having to support a business customer that was on a unique technology platform that was not supported by BCP. There were support documents in the KM system for other technology platforms, but not for this specific business customer. This caused the customer service agent at BCP to spend additional time in assisting the business customer. As a result, the alignment between business process and their KM system was affected by this business process characteristic.

To no surprise, all three organisations provided examples that demonstrated non-routineness in their business processes. With respect to questions on the frequency of dealing with customer problems that involve more than one business function (i.e., interdependence dimensions), in most cases the occurrence of these problems was relatively low. BCP indicated that were of an interdependent nature occurred about 20% of the time, whereas the other cases occurred less than 10% of the time. Along with business process characteristics, KM system characteristics are expected to affect the business process–KM system fit. The technology characteristics can be broken out into two dimensions: system used and department. Our informants indicated the types of systems used and by what departments. For example, BCP mostly uses email and the web but is involved in KM implementation. NIS also uses multiple systems including networking, a knowledge base tool, and call tracking. Likewise, GPS uses various collaboration tools such as email, calendaring, and an Oracle CRM system, to name a few. As such, we conclude that our informants are aware of the systems used and the respective department(s).

The business process–KM system fit consists of eight dimensions following previous research on task–technology fit (Goodhue and Thompson 1995). These dimensions are data quality, locatability, authorisation, compatibility, ease/train- ing, timeliness, reliability, and relationship. As previously mentioned, the first five dimensions focus on using content in

5570 Q. Cao et al.

facilitating business processes, while the next two dimensions focus on day-to-day operations, and the last one focuses on responding to changing business needs (Goodhue and Thompson 1995).

Data quality is measured by the relevancy and currency of the content. Our informants make a variety of statements that indicate the importance of data or content quality with respect to business process–KM system fit and its perceived performance impact. For example, all three companies experience data quality issues but for different reasons. BCP experiences data quality issues due to the KM system tool, whereas NIS and GPS experience data quality issues because of business process issues. This is consistent with the task–technology fit literature with regards to non-routineness and data quality (Goodhue and Thompson 1995). All three companies have a relatively high degree of non-routineness and data quality issues.

While data quality focuses on the relevancy and currency of the content, locatability of content focuses on the ease with which content is available and determining what the content means. Our informants were asked if the content was easy to find in their respective KM systems.

Both BCP and NIS experience locatability issues with regard to KM system content, which is also consistent with the task–technology fit literature (Goodhue and Thompson 1995). Both BCP and NIS have high non-routineness and low locatability. GPS also experiences locatability issues with tacit knowledge but not explicit knowledge. An example of tacit knowledge is the shared impressions of a vendor during a vendor selection process. The vendor may look acceptable on paper, but after conference calls and face-to-face meetings it was determined that the vendor was not acceptable (e.g., not trustworthy). This type of tacit knowledge is hard to document but is important for the organisation to capture. Task–technology fit literature also shows a positive relationship between the number of systems used (a KM system dimension) and locatability. Both informants at BCP and NIS report that they use three different systems to perform their respective jobs and have low locatability, whereas GPS uses multiple systems with a higher degree of locatability.

While the authorisation dimension is defined as access to content necessary to do a job, our informants at BCP and NIS responded by saying that they had access to the necessary content they needed to do their jobs or were not aware of not having access to the necessary content. However, the outbound logistics manager at GPS indicated he could assess the general KM system, but he needed to apply for clearance to assess specialised KM databases.

Likewise, the compatibility dimension depends of the respective organisation. For example, we asked each informant if all the content was in one KM system or in different systems and, if different systems, whether or not the content between systems was consistent. For BCP and NIS, there are different systems and there may not be consistency between the systems. This is consistent with the task–technology fit literature with regards to non-routineness and com- patibility (Goodhue and Thompson 1995). Statements made by (middle-level management) informants from BCP and NIS indicate a high degree of non-routineness and low compatibility. For GPS, there are several different KM systems in the outbound logistics department but mostly a CRM-based KM system and a production-based KM system are used. According to the informant, ‘these two KM systems are very different…KM systems in my department are not fully compatible with other KM systems in GPS.’

The next dimension that focuses on using content in facilitating business processes is ease of use and proper training. Our informants from BCP, NIS, and GPS all indicated that the KM system was easy to use and that there was adequate training. With respect to day-to-day dimensions, we asked each informant whether they felt that the IT depart- ment met its service level agreements for the KM system (timeliness) and whether they could count on the KM system (reliability). This appears to be consistent with the task–technology fit literature with regards to systems used and reli- ability (Goodhue and Thompson 1995). Both BCP and NIS experience reliability issues with the KM system, while GPS appears to have reliability issues with the content in the KM system. The informants from BCP and NIS describe using three different systems with low reliability, whereas GPS informants use several systems with high reliability in terms of the KM system technology but may lack some consistency in terms of content. The final dimension with respect to business process–KM system fit is relationship with user, which focuses on being able to respond to changing needs. Our informants were asked whether they deal with the IT group directly, whether they think the IT group under- stands their specific day-to-day processes, whether they take problems seriously, and whether problems get addressed in a timely manner. In all three cases there appear to be some serious issues or concerns with respect to this dimension. The business owners from both BCP and NIS indicate that they deal with the IT group directly, but there are some issues with resolving problems in a timely manner. GPS indicates that IT is very responsive to fixing bugs, but they are not equipped to deal with information in the systems as they are not the domain experts of KM system content. Accord- ing to task–technology fit theory, this dimension helps align the business process with the IT department (i.e., technol- ogy). Unfortunately, we find marginal support for this dimension.

There appears to be considerable evidence in support of proposition 1. Evidence from our case study links the dimensions of the business process construct such as non-routineness, interdependence, and job title to the business

International Journal of Production Research 5571

process–KM system fit construct. Specifically the dimensions of the business process demonstrate a relationship with quality of data, compatibility, and locatability of the business process–KM system fit. Likewise, the dimension of systems used, which is a proxy for the KM system construct, shows a relationship with locatability and reliability in the business process–KM system fit. Therefore, a relationship exists between business process and KM system characteris- tics to business process–KM system fit, which is consistent with the task–technology fit literature (Goodhue and Thompson 1995). Table 1 summarises the evidence we found between business process characteristics and KM system characteristics to business process–KM system fit among the three case studies.

Proposition 2 relates to whether business process–KM system fit can predict utilisation. We measure utilisation by perceived dependence. Our informants were asked to what degree they are dependent on the KM system. We show a mixed reaction to utilisation, which may have more to do with job title than the business process–KM system fit. The business managers described scenarios where they (and their customers) were dependent, whereas the IT manager showed no dependence but described how customers were dependent. From a business manager’s standpoint there is evidence that business process–KM system fit influences utilisation but not so from the IT manager’s perspective. This may be due to the IT manager’s job, where he is to maintain the KM system as opposed to using the systems for day- to-day activities. Table 2 summarises the evidence we found between business process–KM system fit and utilisation among the three case studies.

Proposition 3 is related to business process–KM system fit and utilisation to perceived individual and organisational performance. Since we are interested in what influence the business process–KM system fit has on perceived performance, we asked our informants whether the KM system has a positive impact on effectiveness and organisational productivity and whether it is important and valuable to them in their job. Our informants indicate that the KM system does have a positive impact and is important to their performance as well as to the organisation. That is, along with our findings from the case studies and the literature, business process–KM system fit and utilisation will directly influence perceived performance and the effectiveness of collaborative networks. For example, the business owner at NIS indi- cated that when the KM system is working properly (e.g., in alignment with business processes), then it is quite helpful. In addition, the logistics manager at GPS described how the KM system helps with their business process and increases

Table 1. Evidence of proposition 1 – BP characteristics and KMS characteristics to BP–KMS fit characteristics.

BP–KMS fit characteristics

Fit characteristic BCP NIS GPS

Data quality U U U Locatability U U Authorisation U Compatibility U U U Ease of use/

training U U U

Timeliness U U U Reliability U U U Relationship

with user U U U

Key quotes: Because of the limitations of the tool, we are not able to publish as we need to. I would say a good percentage of the content is stale or partly outdated. I would say it could be as high as 20%.If you ask the business, No. There is a constant struggle with relevancy and I would say it is not easy to find [the content] for the most part. I have counterparts who deal with the KM implementation who spend more than 50% of their time on it [ill-defined customer problems].

Complexity in trying to access your content – not within the tool. Business process issues – not technology issues.

The challenge for us is how to use KM to catch the illusive tacit knowledge and how to find them quickly when we need them. In general we can easily find basic CRM content in the KM system, but not all the tacit content. In a given day, our department normally spends a bit more than half of the time to deal with non- routine issues. KM system in my department is not fully compatible with other KM systems in GPS. We often rely on KM to handle non- repetitive tasks.

Note: The checkmarks indicate evidence from the interviews of a fit characteristic that was affected by either a business process characteristic or a KM system characteristic.

5572 Q. Cao et al.

productivity. Likewise, at BCP, the business process–KM system fit and the utilisation of the system together led to decreased call-handling time for the agents as non-routine content was easy to find within the KM system. There appears to be considerable evidence in support of proposition 3. Table 3 summarises the evidence we found between business process–KM system fit and utilisation to individual and organisational performance.

5. Lessons learned

This study identifies the fit between business process and KM systems amongst five of the eight dimensions. We do find inconsistent results on whether the fit between business process and the KM system affects utilisation; however, it appears the results may be dependent on job title. In addition, this research explores and discovers that both business process–KM system fit and the KM system utilisation positively influence business performance. The ultimate goal of any organisation is to improve its performance as well as its employees’ performance. The business process–KM system fit and KM system usage positively lead to increases in performance. That is, the knowledge stored and used in the KM system between individuals, departments, and organisations within the value chain can play a role in increasing the per- formance and effectiveness of a collaborative network. This section discusses the study’s implications and insights for researchers and managers and then its limitations and future research.

5.1 Implications

To the best of our knowledge, this study is the first to apply the task–technology fit theory exploring the business process–KM system fit and its impact on business performance in the KM research domain. From the theoretical per- spective, this study underscores the importance of the fit between business process and knowledge management systems in achieving a successful collaborative network. Without exploring the impact of business process–KM system fit on KM system utilisation and impacts of business process–KM system fit and KM system utilisation on business perfor- mance, the salient critical success factors of a KM system would not have been revealed. As a result, this study adds to the KM system literature by demonstrating the significance of a conceptual model derived from the task–technology fit theory.

Table 2. Evidence of proposition 2 – BP–KMS fit to utilisation.

Utilisation

Job title BCP NIS GPS

IT manager No — — Business

owner Yes Yes —

Outbound logistics manger

— — Yes

Key quote ‘For my job I’m not dependent on it all, but for my customers they are very dependent.’

‘My initial reaction is very. Everything we are doing relies on the knowledge system.’

‘[The] KM system is a part of many decision making arsenals we use every day.’

Table 3. Evidence of proposition 3 – BP–KMS fit and utilisation to individual and organisational performance impacts.

Individual and organisational performance Impacts

Job title BCP NIS GPS

IT manager Yes — — Business owner Yes Yes — Outbound logistics manger — — Yes Key quote ‘Yes! It is critical to our strategy

and pivotal to our success. All of strategy is based around knowledge sharing culture and our KM tool provides that.’

‘If working properly, yes… Yes, I believe until we have a working system we didn’t realise how much of an aid it is.’

‘Positive! KM system helps us improve our business process and increase productivity. It also helps creating a knowledge sharing culture our company is pursuing.’

International Journal of Production Research 5573

From a managerial perspective, Chief Information Officers (CIOs) and other business administrators can use this case study as a framework to understand elements of the fit between business process and KM systems, which will lead to higher perceived individual and organisational performance. For example, our case study demonstrated that non-routine business processes have a negative relationship with the data quality of the content. This is a troubling trend since non- routine business processes should have higher quality content as employees may need more assistance in the process. As such, administrators should spend more time and effort on increasing the data quality of content for non-routine business processes. Another finding from our case study is the relationship between non-routine business processes and compati- bility and locatability. This is another example where administrators can spend more time and effort on increasing the compatibility of content in different systems and the ability to locate content for non-routine business processes. These elements combined with other fit elements such as reliability and ease of use/training can ultimately lead to higher per- ceived individual and organisational performance. Increased business performance using a KM system can lead to a more effective collaborative network, thus improving an organisation’s competitive advantage.

5.2 Limitations and future research

There are some limitations of this study that warrant further discussion and need to be kept in mind when interpreting the results. First, this study employs a multiple-case study approach on three high-tech companies in the United States. While multiple-case studies allow for cross-case analysis (Benbasat, Goldstein, and Mead 1987), caution needs to be taken when generalising the results of this research to other industrial settings or applying the framework generated from this study to companies in other cultural contexts. As such, one avenue of future research could increase the number of cases, along with expanding the case study approach to either individual categories of industries or to a broader collec- tion of industries. Second, the KM system application in the organisations we studied was limited in scope (e.g. specific functional areas and not enterprise wide). As our participants highlighted during the interviews, the limited scope of the KM system project prevented them from seeing the value/benefits of enterprise-wide KM system applications and the impact they have on the effectiveness of collaborative networks. A follow-up study could examine the differences in end-users’ reactions to the holistic applications of KM systems.

References

Alavi, M., and D. E. Leidner. 2001. “Review: Knowledge Management and Knowledge Management Systems: Conceptual Founda- tions and Research Issues.” MIS Quarterly 25: 107–136.

Ambrosio, J., 2000. “Knowledge Management Mistakes: Experts Reveal Five Pitfalls to Avoid When Starting Down the Knowledge Man- agement Path.” [Online]. (Accessed 8/29/2011) http://www.computerworld.com/s/article/46693/Knowledge_Management_Mistakes.

AMR Research. 2007. “The Knowledge Management Spending Report, 2007–2008.” http://www.gartner.com/technology/supply- chain-professionals.jsp?pmillid=20768.

Benbasat, I., D. K. Goldstein, and M. Mead. 1987. “The Case Research Strategy in Studies of Information Systems.” MIS Quarterly 11: 369–386.

Bose, R. 2003. “Knowledge Management-Enabled Health Care Management Systems: Capabilities, Infrastructure, and Decision- Support.” Expert Systems with Applications 24: 59–71.

Camarinha-Matos, L. M., and H. Afsarmanesh. 2003. “Elements of a Base VE Infrastructure.” Computers in Industry 51: 139–163. Camarinha-Matos, L. M., and H. Afsarmanesh. 2005. “Collaborative Networks: A New Scientific Discipline.” Journal of Intelligent

Manufacturing 16: 439–452. Cane, S., R. McCarthy, and L. Halawi. 2010. “Ready for Battle? A Phenomenological Study of Military Simulation Systems” Journal

of Computer Information Systems 50: 33–40. Collins, C. J., and K. G. Smith. 2006. “Knowledge Exchange and Combination: The Role Of Human Resource Practices in the

Performance of High Technology Firms.” Academy of Management Journal 49: 544–560. Currier, G. 2010. Knowledge Management and Collaboration Create Knowledge Sharing. Baseline: Ziff Davis Enterprise. D’Ambra, J., and C. S. Wilson. 2004. “Use of the World Wide Web for International Travel: Integrating the Construct of Uncertainty

in Information Seeking and the Task-Technology Fit (TTF) Model.” Journal of the American Society for Information Science and Technology 55: 731–742.

Dhaliwal, J. S., and I. Benbasat. 1996. “The Use and Effects of Knowledge-Based System Explanations: Theoretical Foundations and a Framework for Empirical Evaluation.” Information Systems Research 7: 342–362.

Drew, S. 1999. “Building Knowledge Management into Strategy: Making Sense of a New Perspective.” Long Range Planning 32: 130–136.

Earl, M. 2001. “Knowledge Management Strategies: Toward a Taxonomy.” Journal of Management Information Systems 18: 215–233.

Ferratt, T. W., and G. E. Vlahos. 1998. “An Investigation of Task-Technology Fit for Managers in Greece and the US.” European Journal of Information Systems 7: 123–136.

5574 Q. Cao et al.

Gebauer, J., and M. J. Shaw. 2004. “Success Factors and Impacts of Mobile Business Applications: Results from a Mobile E-Procure- ment Study.” International Journal of Electronic Commerce 8: 19–41.

Goodhue, D. L. 1995. “Understanding User Evaluations of Information Systems.” Management Science 41: 1827–1844. Goodhue, D. L., and R. L. Thompson. 1995. “Task-Technology Fit and Individual Performance.” MIS Quarterly 19: 213–236. Gunasekaran, A., and E. W. T. Ngai. 2007. “Knowledge Management in 21st Century Manufacturing.” International Journal of Pro-

duction Research 45: 2391–2418. Hendriks, P. H. J., and D. J. Vriens. 1999. “Knowledge-Based Systems and Knowledge Management: Friends or Foes?” Information

& Management 35: 113–125. Jagdev, H. S., and K. D. Thoben. 2001. “Anatomy of Enterprise Collaborations.” Production Planning & Control 12: 437–451. Krajewski, L., L. Ritzman, and M. Malahortra. 2010. Operations Management: Processes & Supply Chains. Upper Saddle River,

New Jersey: Prentice Hall. Kristjanson, P., R. S. Reid, N. Dickson, W. C. Clark, D. Romney, R. Puskur, S. MacMillan, and D. Grace. 2009. “Linking Interna-

tional Agricultural Research Knowledge with Action for Sustainable Development.” Proceedings of the National Academy of Sciences 106: 5047–5052.

Kulkarni, U. R., S. Ravindran, and R. Freeze. 2007. “A Knowledge Management Success Model: Theoretical Development and Empirical Validation.” Journal of Management Information Systems 23: 309.

Lee, C.-C., H. K. Cheng, and H.-H. Cheng. 2007. “An Empirical Study of Mobile Commerce in Insurance Industry: Task–Technology Fit and Individual Differences.” Decision Support Systems 43: 95–110.

Liao, S.-H. 2002. “Problem Solving and Knowledge Inertia.” Expert Systems with Applications 22: 21–31. Liebowitz, J. 1999. Knowledge Management Handbook. Boca Raton, CA: CRC Press. Liebowitz, J. 2001. “Knowledge Management and Its Link to Artificial Intelligence.” Expert Systems with Applications 20: 1–6. Liebowitz, J., and K. Wright. 1999. “Does Measuring Knowledge Make “Cents”?” Expert Systems with Applications 17: 99–103. Maruping, L. M., and R. Agarwal. 2004. “Managing Team Interpersonal Processes through Technology: A Task-Technology Fit Per-

spective.” Journal of Applied Psychology 89: 975–990. Mathieson, K., and M. Keil. 1998. “Beyond the Interface: Ease of Use and Task/Technology Fit.” Information & Management 34:

221–230. Nachiappan, S. P., A. Gunasekaran, and N. Jawahar. 2007. “Knowledge Management System for Operating Parameters in Two-Eche-

lon VMI Supply Chains.” International Journal of Production Research 45: 2479–2505. Nonaka, I., and G. Von Krogh. 2009. “Perspective–Tacit Knowledge and Knowledge Conversion: Controversy and Advancement in

Organizational Knowledge Creation Theory.” Organization Science 20: 635–652. Paiva, E. L., A. V. Roth, and J. E. Fensterseifer. 2002. “Focusing Information in Manufacturing: A Knowledge Management Perspec-

tive.” Industrial Management & Data Systems 102: 381–389. Perrow, C. 1967. “A Framework for the Comparative Analysis of Organizations.” American Sociological Review 32: 194–208. Peyman, A., M. Jafari, and M. Fathian. 2005. “Exploring Failure Factors of Implementing Knowledge Management in Organizations.”

Journal of Knowledge Management Practice. Available online at http://www.tlainc.com/articl85.htm Preece, A., A. Flett, D. Sleeman, D. Curry, N. Meany, and P. Perry. 2001. “Better Knowledge Management through Knowledge Engi-

neering.” IEEE Intelligent Systems 16: 36–43. Rubenstein-Montano, B., J. Liebowitz, J. Buchwalter, D. McCaw, B. Newman, and K. Rebeck. 2001. “A Systems Thinking Frame-

work for Knowledge Management.” Decision Support Systems 31: 5–16. Schultze, U., and D. E. Leidner. 2002. “Studying Knowledge Management in Information Systems Research: Discourses and Theoret-

ical Assumptions.” MIS Quarterly 26: 213–242. Tanriverdi, H. 2005. “Information Technology Relatedness, Knowledge Management Capability, and Performance of Multibusiness

Firms.” MIS Quarterly 29: 311–334. Thompson, J. D. 1967. Organizations in Action. New York: McGraw-Hill. Wang, E., G. Klein, and J. J. Jiang. 2007. “IT Support in Manufacturing Firms for a Knowledge Management Dynamic Capability

Link to Performance.” International Journal of Production Research 45: 2419–2434. Wiig, K. M. 1997. “Knowledge Management: Where Did It Come from and Where Will It Go?” Expert Systems with Applications

13: 1–14. Wilkins, J., B. Van Wegen, and R. De Hoog. 1997. “Understanding and Valuing Knowledge Assets: Overview and Method.” Expert

Systems with Applications 13: 55–72. Yin, R. K. 2009. Case Study Research: Design and Methods. Thousand Oaks, CA: Sage. Zigurs, I., and B. K. Buckland. 1998. “A Theory of Task/Technology Fit and Group Support Systems Effectiveness.” MIS Quarterly

22: 313–334.

International Journal of Production Research 5575

Copyright of International Journal of Production Research is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.