Capital budgeting is the process within organisations where prospect investments are screened, evaluated and reviewed. Butler et al. (1998) describe the capital budgeting as a process where organizational capital is employed in relation to their future estimated gains. Within many large organisations the capital budgeting process can be seen as ideas stemming from lower levels of the organisations, later to be reviewed by divisional management (Slagmulder, Bruggeman and van Wassenhove, 1995). The process can be portrayed as several joint stages. Anthony, Dearden and Bedford (1984) describe the capital budgeting process as eight steps, starting with the identifying of a project need, and ending with post-implementation audits to assess cost and benefits realisations. One of these steps is individual projects being appraised and revised if necessary, which is the main focus of this paper. In a similar manner, Mukherjee (1987) also classified the capital budgeting process as containing eight steps, starting with strategic planning where defining the strategy and hence deciding how to allocate capital, and the final step being a post-implementation audit and project review. Just as Anthony, Dearden and Bedford (1984), the financial appraisal of screened projects is positioned in the middle of the process. Mintzberg, Raisinghani and Theoret (1976) propose a somewhat simpler four-stage model, where investment appraisal is included in the stage of selecting a project.
The investment appraisal can be seen as similar to ex-ante evaluation or as a part of the same (Ababneh, Zeglat and Shrafat, 2017), in contrast to ex-post evaluation which could be seen as the final review of the project, post-implementation. The investment appraisals’ purpose in the capital budgeting process is declared by both Irani and Love (2002) as well as Farbey, Land and Targett (1992). Reasons include the need to compare projects, rank projects, act as a control mechanism of the following costs, benefits and implementation as well as creating a framework for organizational learning (Irani and Love, 2002). Farbey, Land and Targett (1992) use similar reasons for why the investment appraisal is needed, but however adds one reason, namely the ability to justify the investment. Investment appraisal can be seen as a process conducted prior to the beginning of each project in order to support the approval of the business case, while the ex-post evaluation occurs at project closure in order to identify project success or failure (Zwikael and Smyrk, 2012). Ex-ante evaluation of IT investments is mostly dependent on financial estimates. It is normally performed using financial criteria such as NPV, payback and IRR. Its purpose is to support the justification of the investment (Ababneh, Zeglat and Shrafat, 2017). However, Christensen, Kaufman and Shih (2008) criticized these financial estimates by coining the DCF trap expression. The DCF trap refers to the miss guiding comparison of cash flows from innovation against the default scenario of doing nothing. Christensen, Kaufman and Shih (2008) argue that it is incorrect to assume that the present health of the company will persist indefinitely if the investment is not carried through. However, the standard scenario should according to Christensen, Kaufman and Shih (2008) be declining sales and cash flows, given the changing environment, since technology advances and competitors most likely will advance.
Finally, despite not being all available methods within the area, Irani and Love (2002) compiled a framework of investment appraisals techniques used within the IT/IS investment area, which it considered relevant. This model is presented in a simplified form in figure 1.0 below.
Business Intelligence systems
Luhn (1958) coined one of the first definitions of BI system accordingly, “a comprehensive system may be assembled to accommodate all information problems of an organization. We call this a Business Intelligence System”. Chee et al. (2009) reviewed existing definitions of BI and made a distinction between the technological aspect, the process perspective and the product. The technological aspect refers to the BI system, the process is the implementation of the system and finally, the product is the result of implementation and outcome generated by the system.
The definition of BI can be perceived as multifaceted due to the nature of possibilities within these systems. The term is according to Vitt, Luckevich and Misner (2002) used in a very broad manner and has a different meaning for different stakeholders. Likewise, Arnott and Gibson (2005) argue that it depends on who is defining it, and vendors can easily frame it differently depending on their needs, which can be seen in appendix. 2.
Negash (2004) explains how BI systems support decision makers by offering actionable information in the correct format at the right time. It argues that the most important aspect is that the information is delivered quicker than by a standard DSS, which enables decision makers to act proactively. Dedić and Stanier (2017) describe BI as systems that enable companies to extract and present data from internal and external sources by running queries, which results in beneficial reports helpful to streamline the daily work within operations and decision making. Olszak and Ziemba (2007) highlight following components and argue that they constitute the fundamental basis of BI systems:
Extracting tools – Tools to extract and load data i.e. Extract Transform Load (ETL), mainly concentrating on transforming data from transaction systems and Internet to data warehouses.
Data warehouses – Offers room for storing aggregated and analysed data.
Analytics tools – Enables users to access, analyse and model business problems, to be able to distribute information that is stored in the data warehouses, e.g. OLAP.
Data mining tools – Makes it possible to discover patterns, generalizations, regularities and rules in data resources.
Tools for reporting and ad hoc inquiring – Creates and utilities synthetic reports.
Presentation layer – Includes graphics and multimedia interfaces, which offer users information in a comfortable and manageable form.
BI systems were initially mostly used by the IT departments, given the low amount of expertise within the area among other departments. However, the usage among different departments increased as technology and user-friendlier programs were being developed. BI is today used by many different departments and the number of users will increase even further, which implies large volumes of future investments (Moore, 2017; Hawking and Sellitto, 2010). Hawking and Sellitto (2010) and Negash (2004) argue that BI has significant impacts on companies’ performance and is consequently perceived as a high priority amongst managers. However, Hawking and Sellito (2010) highlight the challenges with BI investments, which are foremost related to computing the anticipated return on the investment. It is difficult due to the large costs up front and the fact that the efficiency savings are only a small portion of the payoff, coupled with many intangible benefits. Also, Negash (2004) state that it is uncommon for a BI system to pay off itself strictly through cost reductions.
Negash (2004) made a definition of a BI system as “… a system that combines data gathering, data storage, and knowledge management with analytical tools to present complex and competitive information to planners and decision-makers.”, which is an appropriate classification for this paper and shall, therefore, be the basis going forward. The definition by Negash (2004) is well suited for this paper, given that it contains several of the notions discussed as key factors during the literature review and problem definition.
Following section cover the specific and relevant theories and literature given the research question and purpose of the study.
Literature review approach
The literature review was conducted with a narrative approach. Trieu (2017) acknowledged that the area of precedent literature within BI systems was thin, especially regarding ex-ante evaluation, which implied a need to extend the review to include articles regarding ex-ante IT/IS evaluations. Appendix. 3 displays how the literature review was conducted, it highlights what search engines and keywords that were used in order to find relevant journals and articles.
It has since long been acknowledged that IT/IS investment appraisals are difficult given the significant amount of intangible benefits provided. Kaplan (1986) discusses an early example of investments in computer integrated manufacturing (CIM), where it is clarified that there is a conflict between the financial justification (i.e. appraisal in our meaning) and strategic justification. It argues that it necessarily must not be the case, given that there is no underlying issue with the DCF, and hence conclude that practitioners must learn to apply the DCF more appropriately, and adjust it to be more “sensitive to the realities and special attributes of CIM.” (Kaplan, 1986). Though, Kaplan (1986) endorses the fact that some intangibles are very difficult to assess a cash flow value to, but nevertheless argue that practitioners may be too conservative not giving these benefits a value at all and therefore leaving the investment to faith alone. However, as will be presented in the following literature review, many researchers argue that new appraisal methods are needed to perform the ex-ante evaluation (Willcocks, 1992; Hochstrasser, 1990; Gibson, Arnott and Jagielska, 2005).
Klein and Beck (1987) discuss a similar issue as Kaplan (1986) presents, namely that the current evaluation methodologies used when considering a IS ignore the need to choose between qualitative factors or require a numerical value to be attached to the qualitative attributes. Qualitative factors should be included in assessing investment appropriateness since they in these types of assets are important, but the lack of ability to assess a quantifiable value to it hinder their value to be shown in current decision-making models (Klein and Beck, 1987). Drawing on preference theory, Klein and Beck (1987) discuss a method based on the decision maker choosing between presented attributes, however, the model does not present much aid in evaluating a single alternative and is hence limited in appraisals concerning a single investment opportunity.
Similarly as many of the articles presented in the literature review, Parker, Benson and Trainor (1990) and Maskell (1991) explore the issue of companies using traditional evaluation methods to justify IT/IS investments. Parker, Benson and Trainor (1990) argue that most managers feel responsible for making appraisals when investing in IT/IS, despite traditional investment appraisals not being sufficient to capture the many intangible benefits.
Hochstrasser (1990) refers to the Kobler Unit study and highlights the result, which is that most practitioners tend to use methods that emphasise the contribution to the bottom line. These methods may result in the quantification of benefits easy to measure rather than evaluating what is important, which is in line with Kaplan (1986) claiming that IT/IS investments should not be left to faith alone. However, the arguments behind this claim are in contrast to those of Kaplan (1986). It argues that there is a need for methods focusing on intangible benefits. Finally, Hochstrasser (1990) argues that IT/IS investments cannot be justified by one single appraisal procedure due to the dynamic factors inherent in IT/IS investments.
Thereafter, Farbey, Land and Targett (1992) set out to understand how organisations actually do while appraising whether or not to go ahead with an IT/IS investment or not and also what role the appraisal itself plays. The research is undertaken since evaluating the potential benefits of these investments are considered a major issue amongst managers in general. In line with Hochstrasser (1990) they argue that the previous dominance of a few appraisal techniques has led to the search for a single optimal appraisal technique for IT/IS investments, which most likely will not be successful (Farbey, Land and Targett, 1992). In order to understand how organisation appraise investments, Farbey, Land and Targett (1992) presents the foundational reasons to why they do appraisals, namely (1) justification of the investment, (2) compare investments, (3) allow benchmarking to project performance, and lastly (4) compare actual outcome for organisational learning. Looking at 16 large IT investments, Farbey, Land and Targett (1992) found a mixture of approaches, ranging from standard procedures, ad-hoc justification or even no justification at all. The article concludes that very few of the available, by academia presented methods, were used in practice, and as a response draws on a framework for matching appraisal techniques with IT/IS investments, depending on several variables.
In line with the above-presented uncertainties regarding what methods are deemed usable, Powell (1992) highlights the issues regarding IT/IS investment appraisals and reviews existing as well as new methods to be able to consider if IT/IS investments differ from other investments. Powell (1992) cannot decide whether IT/IS investments differs from other investments but concludes that scholars should focus on improving existing methodologies rather than creating new ones since the field is already crowded. However, the findings suggest that the major technique at the time were NPV and IRR. This consequently led to many non-quantifiable benefits being left out, making the justification of IT/IS investment difficult. These difficulties resulted in many organizations not making ex-ante evaluations at all.
Willcocks (1992) defines the problems within the IT/IS investments as a Catch 22, referring to the situation where companies need to invest in IT/IS for competitive reasons but the appraisals do not justify it. Also, it argues that the current evaluation techniques are not suitable. Furthermore, Willcocks (1992) describes how most of the current techniques have been developed and pushed forward by experts, vendors and consultant instead of the profit center managers, which entails biased and inappropriate techniques. Finally, Willcocks (1992) concludes in line with Powell (1992) that traditional techniques cannot be relied upon and highlights the large range of modern techniques, which means that organizations need to shape the process of conducting evaluations.
Similarly as Parker, Benson and Trainor (1990), Hirschheim and Smithson (1998) argue that IT/IS appraisal is a “necessary evil”, which is demanding and complex but still very important to enable a justification to senior management. Hirschheim and Smithson (1988) analyzed the methods at the time by dint of an old framework and were able to draw the conclusion that the development of both business and IT/IS itself has made the evaluation process more complex, mainly due to the shift in the nature of benefits related to the investments.
Just as Farbey, Land and Targett (1992) and Hochstrasser (1990), Ballantine and Stray (1998) evaluate the use of appraisal methods used by organisations when assessing IT/IS investments. It proposes that there are arguments both against and for using standard investment appraisal techniques, there amongst the issue that they are not capturing all the inherent benefits and are too financially oriented, but on the other hand this could be argued as logical from a shareholders’ maximisation perspective. What Ballentine and Stray (1998) found was that a smaller portion of the respondents used more sophisticated appraisal techniques such as NPV and IRR, which may be due to the fact they are not seen as easily adapted to the nature of IT/IS investments. It was also found that the quantification of benefits was more difficult for practitioners, rather than the identification of them.
Irani (2002) states that there is a general difficulty discussed in the normative literature regarding IT/IS evaluations and that there is a difficulty assessing the full impact of such investments. Looking at an IS investment in a manufacturing company, Irani (2002) focus on the limitations of traditional appraisal techniques, and how the financial and conceptual justification was carried through. The manufacturing company lacked experience since precedent investments could be evaluated by standard investment appraisal techniques. Even though benefits were identified, no estimated financial values were attached. Irani (2002) concludes that many social and technical factors make the search for a generic evaluation method impossible. Irani and Love (2002) set out to present a framework for the ex-ante evaluation of IT/IS investments by looking at the appraisal stage of the capital budgeting process, and sorting the different appraisal methods into subgroups. Irani and Love (2002) further discuss the issues of using traditional appraisal techniques when assessing IT/IS investments and argue that more strategic, analytical and integrated appraisal methods should be used. Though, these are not well established amongst practitioners, probably due to the complexity and subjectivity and the fact that focus on expected financial returns in many capital budgeting processes still plays a key role.
Just as Irani and Love (2002) argue that there is a need for new methods to be adopted, Walter and Spitta (2004) look at both the academic contribution as well as practitioners and thereafter presents a framework for classifying appraisal techniques. It goes on to discuss their limitations, and finally reviews findings regarding how well the techniques are established in practice. The appraisal techniques can roughly be divided into either effect-assessing or effect-locating, where effect-assessing are more financially oriented, while effect-locating are more of interpretative nature. Walter and Spitta (2004) presents that there are certain financially oriented methods which try to apply quantitative measures to qualitative benefits. It is also argued that no single method is universally applicable (which is in line with Farbey, Land and Targett (1992)). Similarly to earlier studies Walter and Spitta (2004) found that financially oriented methods are more widely established, which most often leave the intangible values disregarded. The main problems experienced were related to identification, quantification and estimation of benefits when assessing IT/IS investments.
Looking at BI more specifically, Gibson, Arnott and Jagielska (2005) state that there is a limited body of knowledge regarding the intangible benefits of BI, and that these benefits just as other IT/IS investments are difficult to quantify. However, given the lack of research in the area, it is not yet known whether or not BI investments are appraised in the same manner as other IT/IS investments, and if there is a need amongst practitioners to use new methods. Gibson, Arnott and Jagielska (2005) state that this area is of importance to further explore, given that there is a need to justify BI investments, and also a need amongst BI vendors to understand how techniques are used to evaluate their products.
Myrtidis and Weerakkody (2008) go on to discuss the difficulties emphasized in earlier literature (e.g., Powell, 1992; Farbey, Land and Targett, 1992; Hirschheim and Smithson, 1998; and Irani and Love, 2002) and highlights the problem even further, arguing that digitalization has led to escalating costs within the IT/IS investment area. Myrtidis and Weerakkody (2008) build further on Symons (1991)’s arguments concerning IT/IS investments and the political problems within organizations. It concludes that different goals and agendas behind the investment will have an impact on the evaluation. Myrtidis and Weerakkody (2008) highlight the importance and limitations regarding IT/IS appraisal methods used in practice and argues that the appraisals and justifications focus on short-term benefits. Finally, it calls for further research within the area, to create a deeper understanding of the organizational processes.
Song and Letch (2012) reflects on the last twenty-five years body of literature related to IT/IS investments, and highlights where research has been focused, what are the main issues and finally where future research is needed. Song and Letchs (2012) reflection discover that most of the examined articles ignore the purpose of evaluation, described by Farbey, Land and Targett (1992), and most of the articles focus on the ex-post evaluation even though ex-ante evaluation is found to be more prevalent in practice. Finally, Song and Letch (2012) concluded that the methods presented by academia are many which make the field crowded, however few of these methods are used in practice, which is in line with earlier findings from Powell (1992) and Farbey, Land and Targett (1992).
Royer (2013) presents an overview of relevant ex-ante evaluation methods related to IT/IS investments. These methods are further evaluated and examined from the perspective of practitioners. Royer (2013) finds that financial methods e.g. NPV, payback, return on investment (“ROI”) and IRR seems to be most relevant in practice. However, the Balanced IT Decision Card (“BITDC”) scored highest on the evaluation since this method takes most of the effects and benefits into account. Finally, Royer (2013) highlights the limits with BITDC in practice, which is mainly related to the limitation of data and quality.
Looking at decision making in the context of BI investments Frisk et al. (2014) argue for the need of a more interactive, creative and adaptive model for explaining decision making (cf. rational decision making under known circumstances). According to Frisk, Lindgren and Mathiassen (2014), this could enable a better understanding of both structured and unstructured data. Further, it argues that this approach could make investments in BI more easily adopted and analyzed, mainly because it allows managers to measure the tangible benefits while sensing the intangibles.
Ababneh, Zeglat and Shrafat (2017) and Trieu (2017) are both recent literature reviews, which summarize literature dating back to 1988. Despite having been systematically addressed throughout the years, researchers yet have contradictory views regarding what models ought to be used when appraising IT/IS investments. At the same time, many different methods have been suggested, but given the many different situations and contexts in which IT/IS is being evaluated, there is likely no best methodology readily available to suit the different situations (Ababneh, Zeglat and Shrafat, 2017) There is a widespread concern regarding the use of evaluation and justification techniques in organisations, and this issue becomes yet more palpable given the intensive use of IT/IS/BI and the large investments that have been made (Ababneh, Zeglat and Shrafat, 2017; Trieu, 2017), and there has not been much progress in the field (Ababneh, Zeglat and Shrafat, 2017).
Summary of literature review
As presented in several of the articles above, there is clearly a recurring elucidated issue where the investment appraisal techniques in use are not appropriate for the task. There are inherent issues with the use of the methods, and it results in intangible benefits not being sufficiently accounted for. Further, the literature review shows that there are plenty of by academics presented methods for evaluation, though these are not yet used by practitioners. Some researchers argue that there is a need for new methods in order to better fit the needs of practitioners, while others argue that there need to be a better adoption and application of the already available methods. Finally, the issue which has been discussed with regards to IT/IS investments are yet more palpable in the situation of BI investments, given the nature of the systems’ benefits. This area is according to researchers in need to be further explored, and there is a need to understand how practitioners are handling these issues. As a summary, there is inherent criticism amongst the researchers whether certain methods are deemed appropriate or not. Other disagreements between researchers in the area include the discrepancy of whether or not there should be a focus on developing new methods or further development of a large number of already existing methods of which very few are used in practice. For a summary of precedent research, see table 2.0.
Criticism to precedent literature
Consensus exists among the presented literature regarding the inadequacies of methods used by practitioners. However, there is no evidence of a call or need from practitioners concerning improvements of these methods, which Myrtidis and Weerakkody (2008) also mentions when discussing the relevance of further research within the area. Further, the literature presents several different methods to evaluate these investments (e.g. Irani and Love (2002); Farbey, Land and Targett (1992); Ballentine and Stray (1998); Royer (2013)), but details regarding how to carry through an evaluation with these methods and capture the intangible benefits are limited, with exception for the suggestions presented by Klein and Beck (1987). Furthermore, except for Myrtidis and Weerakkody (2008) a majority of the case studies and the literature that focus on describing what methods that are mostly used among the practitioners (e.g. Ballentine and Stray (1998)) are limited to the Anglo-Saxon markets. This could, of course, limit the possibilities to draw conclusions regarding what extent the evaluated cases in this studies are aligned or not with other companies in their own geographical jurisdiction. Finally, the literature presented consists of several literature reviews, repeating each other to a large extent, and there is also few of the articles discussing BI investments specifically and rather investigate IT/IS investments more generally.
Table of contents :
1. 1 BACKGROUND
1.2 PROBLEM DISCUSSION
1.3 RESEARCH QUESTION
2.1. CAPITAL BUDGETING
2.2 BUSINESS INTELLIGENCE SYSTEMS
2.3 THEORETICAL FRAMEWORK
2.3.1 LITERATURE REVIEW APPROACH
2.3.2 LITERATURE REVIEW
22.214.171.124 SUMMARY OF LITERATURE REVIEW
126.96.36.199 CRITICISM TO PRECEDENT LITERATURE
2.3.3 SHAREHOLDER THEORY AND DECISION MAKING
2.4 ANALYTICAL FRAMEWORK
3.1 RESEARCH METHODOLOGY
3.3 DATA SAMPLE AND RESPONDENTS
3.3 DATA COLLECTION
3.3.1 INTERVIEW METHOD
3.3.3 DOCUMENTATION RECEIVED
3.4 DATA ANALYSIS
3.5 METHOD DISCUSSION
3.5.1 CRITICISM OF RESEARCH DESIGN AND METHOD
188.8.131.52 ETHICAL CONSIDERATIONS
184.108.40.206 RELIABILITY, REPLICABLE & VALIDITY
4.0 EMPIRICAL FINDINGS
4.1.1 CASE 1 (“TRADING COMPANY”)
4.1.2 CASE 2 (“CONSTRUCTION COMPANY”)
4.1.3 CASE 3 (“MANUFACTURING COMPANY”)
4.2.4 COMPARATIVE EVALUATION CONDUCTED
5. DATA ANALYSIS & DISCUSSION
5.1 DATA ANALYSIS
5.1.1 BACKGROUND AND INTENTIONS
5.1.2 GENERAL CAPITAL BUDGETING PROCESS
5.1.3 BI INVESTMENTS AND EVALUATION
5.1.4 GAP BETWEEN RESEARCH AND PRACTITIONERS