Development of performance indicators
through cost driver identification
-‐ an IT department case study
SAMI ALMEHDI ÖSTERMAN
CARL LUNDBERG
Master of Science Thesis
Utveckling av nyckeltal genom
identifiering av kostnadsdrivare
-‐ en fallstudie på en IT-‐avdelning
SAMI ALMEHDI ÖSTERMAN
CARL LUNDBERG
Examensarbete Stockholm, Sverige 2012
Utveckling av nyckeltal genom identifiering av
kostnadsdrivare
-‐ en fallstudie på en IT-‐avdelning
av
Sami Almehdi Österman
Carl Lundberg
Examensarbete INDEK 2012:84
KTH Industriell teknik och management Industriell ekonomi och organisation
Development of performance indicators through
cost driver identification
-‐an IT department case study
by
Sami Almehdi Österman
Carl Lundberg
Master of Science Thesis INDEK 2012:84 KTH Industrial Engineering and Management
Industrial Management SE-‐100 44 STOCKHOLM
Examensarbete INDEK 2012:84
Utveckling av nyckeltal genom identifiering av kostnadsdrivare
-‐ en fallstudie på en IT-‐avdelning
Sami Almehdi Österman
Carl Lundberg Godkänt
2012-‐06-‐21
Examinator
Prof. Håkan Kullvén
Handledare
Prof. Håkan Kullvén Prof. Thomas C. Westin
Konfidentiellt Konfidentiellt Sammanfattning
Kostnaderna för informationsteknologi (IT) i stora multinationella företag utgör ofta en betydande del av företagets totala årliga omsättning. Samtidigt som IT-‐organisationer förväntas leverera värde till företaget, utsätts de även för kostnadsnedskärningar. Därför är det inte ovanligt att IT-‐ledningen väljer att fokusera på kostnadsbesparingar, snarare än kostnadskontroll, för att nå bolagets finansiella mål. Det är inte alltid enkelt att förstå vad som driver kostnader i en IT-‐organisation, vilket försvårar besluten om vilka nyckeltal som IT-‐ ledningen bör styra efter. Denna studie tar upp frågan om kostnadskontroll och prestationsmätning i IT-‐ organisationer i form av en fallstudie, som genomförts på IT-‐avdelningen på ett stort svenskt multinationellt bolag i telekommunikationsbranschen. Organisationen kan delas in i Enterprise och Engineering, där Enterprise tillhandahåller IT i form av skrivare, persondatorer, applikationer, IT support och kommunikationstjänster genom nätinfrastruktur och taltjänster. Engineering tillhandahåller infrastruktur för mjukvaruutveckling och en miljö för testning av produkter för forsknings-‐ och utvecklingsenheter. Testmiljön består av både mjukvarutestning i form av simuleringar samt tester av hårdvara i fysiska laboratorier. Med utgångspunkt i intervjuer med chefer och andra nyckelpersoner ansvariga för stora budgetposter, identifierades 130 kostnadsdrivare inom IT-‐avdelningen. Efter reduktion av återkommande kostnadsdrivare och sammanslagning av liknande drivare, filtrerades kostnadsdrivarna med aveseende på kvantifierbarhet. Kostnadsdrivare som var kvantifierbara översattes sedan till nyckeltal med hjälp av SMART-‐modellen. I efterföljande steg poängsattes nyckeltalen med avseende på kostnadseffekt och möjligheten för
implementering. Detta resulterade i en rekommendation av 49 nyckeltal för effektiv kostnadskontroll i hela
organisationen. Ett urval av rekommenderade nyckeltal är totala lönekostnaden / antal anställda, antal virtualiserade servrar / totala antalet servrar samt mängden videorelaterad datatrafik / totala bandbreddskapaciteten. Ambitionen med denna studie är att på ett ett holistiskt sätt kontrollera kostnadsdrivare, genom prioritering av nyckeltal. Även om denna fallstudie specifikt riktar sig mot IT-‐ avdelningar, anser författarna att tillvägagångssättet även kan tillämpas på andra avdelningar och branscher.
Master of Science Thesis INDEK 2012:84
Development of performance indicators through cost driver identification
-‐
an IT department case study
Sami Almehdi Österman
Carl Lundberg Approved
2012-‐06-‐21
Examiner
Prof. Håkan Kullvén
Supervisor
Prof. Håkan Kullvén Prof. Thomas C. Westin
Confidential Confidential Abstract
The costs of information technology (IT) in large multinational companies (MNCs) often constitute a significant portion of the company’s total yearly turnover. IT departments are on one hand expected to return value to business, but are on the other hand often prone to cost reductions. Thus, it is not unusual that IT management chooses to focus on cost cutting rather than cost control in order to meet the company’s financial targets. Understanding cost drivers in IT is not always evident, making it difficult for managers to know what performance indicators that should be tracked. This study addresses the issue of cost control and performance measurement in IT departments, in the form of a case study carried out at the IT department of a large Swedish MNC in the telecommunications industry. The case company is divided into two parts: Enterprise and Engineering, where Enterprise provides the organization with IT in form of printers, personal computers, applications, IT support and communication services through network infrastructure and voice. Engineering provides research and development units with software development infrastructure and testing environments for products. The testing environments comprise of both software testing through simulations and hardware testing in physical labs. By carrying out interviews with managers and key people responsible for large budget items, 130 cost drivers were identified. After reducing recurring cost drivers and merging similar ones, the cost drivers were filtered according to quantifyability. In a second step, performance indicators were developed using the SMART model and then scored with respect to cost impact and ease of
implementation. This resulted in a recommendation of 49 performance indicators to be tracked across the
entire IT department. A sample of recommended performance indicators is total cost of wages / number of employees, number of virtualized servers / total number of servers and amount of video related traffic / total capacity of bandwidth. The ambition of this study is to provide a holistic way of controlling cost drivers through prioritized performance indicators. Even though this case is specific to an IT department, the approach in this research may well be applied in other departments and industries.
Table of Contents
1. INTRODUCTION ... 1
1.1 BACKGROUND ... 1
1.2 PROBLEM DESCRIPTION AND RESEARCH QUESTION ... 2
1.3 THE AIM OF THE STUDY ... 2
1.4 DELIMITATIONS ... 3
1.5 OUTLINE ... 3
2. THEORETICAL FRAMEWORK ... 4
2.1 BUDGETS IN MANAGEMENT ACCOUNTING ... 4
2.2 ACTIVITY BASED MANAGEMENT ... 4
2.2.1 COST DRIVERS AND THE ACTIVITY-‐BASED COSTING MODEL ... 4
2.2.2 TYPES OF COST DRIVERS ... 5
2.2.3 CATEGORIZING COST DRIVERS ... 6
2.3 BUDGET ELEMENTS AND COST DRIVERS IN IT ORGANIZATIONS ... 6
2.4 PERFORMANCE INDICATORS ... 9
2.5 PERFORMANCE MEASUREMENT SYSTEMS ... 11
3. METHODOLOGY ... 12
3.1 JUSTIFICATION OF METHODOLOGY AND RESEARCH PARADIGM ... 12
3.2 RELIABILITY AND VALIDITY OF RESEARCH ... 12
3.3 DATA COLLECTION ... 13
3.3.1 DOCUMENTARY ANALYSIS AND MEETINGS ... 13
3.3.2 INTERVIEWS ... 14
3.3.3 REVISED METHODOLOGY AFTER FIRST INTERVIEW ... 15
3.4 DATA ANALYSIS ... 15
3.4.1 ANALYSIS OF INTERNAL DOCUMENTS ... 16
3.4.2 ANALYSIS OF INTERVIEWS ... 16
3.5 LIMITATIONS OF THE STUDY ... 16
3.6 OUTLINE OF RESEARCH METHODOLOGY ... 17
4. CASE STUDY ... 18
4.1 PRESENTATION OF THE IT DEPARTMENT ... 18
4.1.1 ENTERPRISE AND ENGINEERING ... 19
4.1.2 THE ENTERPRISE ORGANIZATION ... 19
4.1.3 THE ENGINEERING ORGANIZATION ... 19
4.1.4 COSTS IN THE IT DEPARTMENT ... 20
4.2 INTERVIEW OBJECTS ... 22
5. FINDINGS ... 26
5.1 IDENTIFIED COST DRIVERS ... 26
6. ANALYSIS OF FINDINGS ... 31
6.1 ANALYSIS OF COST DRIVERS ... 31
6.2 ANALYSIS OF PERFORMANCE INDICATORS ... 34
6.3 ASSESSMENT OF DEVELOPED PERFORMANCE INDICATORS ... 37
6.4 ANALYSIS AND IMPLICATION OF EXCLUDED DATA ... 39
7.1 MAPPING OF PERFORMANCE INDICATORS ... 40
7.2 INFORMATION ABOUT RECOMMENDED PERFORMANCE INDICATORS ... 44
7.3 CONCLUSIONS AND GENERAL DISCUSSION ... 53
7.4 THEORETICAL CONTRIBUTION ... 54
7.5 PRACTICAL CONTRIBUTION ... 54
7.6 LIMITATION OF THE STUDY ... 55
7.6.1 LIMITATIONS OF THE EMPIRICAL DATA ... 55
7.6.2 LIMITATIONS OF THE METHOD ... 55
7.6.3 TRANSFERABILITY ... 56
7.7 SUGGESTIONS FOR FURTHER RESEARCH ... 56
8. REFERENCES ... 58
APPENDIX ... 61
APPENDIX 1 TABLES OF COST DRIVERS IDENTIFIED IN ENTERPRISE ... 61
APPENDIX 2 – TABLES OF COST DRIVERS IDENTIFIED IN ENGINEERING ... 70
APPENDIX 3 – INTERVIEW QUESTIONS ... 77
APPENDIX 4 COST DRIVERS, PROPOSED PERFORMANCE INDICATORS AND AREAS AFFECTED FOR ENTERPRISE. ... 79
APPENDIX 5 COST DRIVERS, PROPOSED PERFORMANCE INDICATORS AND AREAS AFFECTED FOR ENGINEERING. ... 83
APPENDIX 6 ASSESSMENT OF DEVELOPED PERFORMANCE INDICATORS IN ENTERPRISE ... 85
APPENDIX 7 ASSESSMENT OF DEVELOPED PERFORMANCE INDICATORS IN ENGINEERING ... 86
List of Figures
Figure 1 IT expenses / IT budget as a percentage of sales in 2002. Source: Buchta (2010) ... 1
Figure 2. Aligning IT Performance management to corporate strategy. Source: Buchta (2010) ... 11
Figure 3. Process for selection of interviewees in Enterprise. ... 14
Figure 4. Process for selection of interviewees in Engineering. ... 14
Figure 5. Illustration of the research design and methodology. ... 17
Figure 6. Simplified organizational structure of the IT department, as of March 2011. Source: Company Internal A (2012) ... 18
Figure 7. Illustration of the main process within the IT-‐department. Source: Company Internal B (2012) ... 19
Figure 8. Representation of the division of the IT department into its two main parts: Enterprise and Engineering and R&D IT. Source: Company Internal C (2012) ... 19
Figure 9. The Engineering testing process, with a graph illustrating increased costs further down the chain. Source: Interview with Vice President Service Delivery (2012) ... 20
Figure 10. Enterprise and Engineering respective percentages in entire budget. Source: Company Internal D (2012) ... 20
Figure 11. Pie chart representing the Enterprise budget for 2012. Source: Company Internal E (2012) ... 21
Figure 12. Pie chart representing the Engineering budget 2012. Source: Company Internal F (2012) ... 22
Figure 13 Cost drivers in their categories according to service, with the additional category ”General”. ... 32
Figure 14 Categories of cost drivers in Engineering. ... 33
Figure 15. Decision flow chart for developing performance indicators. The YES/SMART loop was used to provide iterations for more and more suitable performance indicators. . 34
Figure 16. Two criterions performance indicators were assessed on. ... 37
Figure 17. Mapped performance indicators for IT-‐management, Service Desk & Support, and Development and Implementation budget, accounting for 25 % of the total Enterprise budget. ... 40
Figure 18. Mapped performance indicators for Clients, Client Infrastructure Services, Voice and Network which accounts for 34 % of the total Enterprise budget. ... 41
Figure 19. Mapped performance indicators for Platform operations, Application operations, Software Licenses and Storage which accounts for 28 % of the total Enterprise budget. .... 42
Figure 20 Engineering budget divided into Testing, R&D IT and Global Tools. ... 42
Figure 21. The testing branch of engineering representing 75 % o the total Engineering budget. ... 43
Figure 22. Performance indicators that effects all budget posts in testing. ... 43
Figure 23. R&D IT and Global Tools representing the rest of the Engineering budget, 25%. ... 44
List of Tables
Table 1 Table exemplifying structural and executional cost drivers. Source: Fong (2011) 6 Table 2 Table exemplifying operational drivers, commonly used in the ABC model. Source:
Fong (2011) 6
Table 3 Typical budget categories related to IT organizations. Source: Cassidy & Cassidy
(2010) 7
Table 4 The table presents key IT cost drivers, comments and areas affected. Source:
Baschab & Piot (2007) 9
Table 5 SMART model – specific, measurable, achievable, relevant and timed. Source:
Modified model from Platt (2002) 10
Table 6 Seven suggested criterias for performance indicators. Source: Binnendijk (1996) 11
Table 7. Groupings of interviews covered in Enterprise. 22
Table 8 Groupings of interviews covered in Engineering. 24
Table 9 Cost drivers identified in Enterprise 28
Table 10 Cost drivers identified for Engineering 30
Table 11 Cost drivers that were combined in Enterprise 31
Table 12 Cost drivers that were combined in Engineering 33 Table 13. SMART criteria’s applied on Enterprise performance indicators 36 Table 14. SMART criteria’s applied on Engineering performance indicators 37 Table 15. Assessment of developed performance indicators in Enterprise. Recommended
indicators are included in the shaded area. 38
Table 16. Assessment of developed performance indicators in Engineering. Recommended
indicators are included in the shaded area. 38
Table 17. PIs recommended for IT Management. 45
Table 18. PIs recommended for Service Desk & Support. 46 Table 19. PIs recommended for Development and Implementation. 46
Table 20. PIs recommended for Clients. 47
Table 21. PIs recommended for Client Infrastructure. 48
Table 22. PIs recommended for Voice. 48
Table 23. PIs recommended for Network. 49
Table 24. PIs recommended for Platform Operations. 49
Table 25. PIs recommended for Application operations. 49
Table 26. PIs recommended for Software Licenses. 50
Table 27. PIs recommended for Storage. 50
Table 28. PIs recommended for Depreciations. 51
Table 29. PIs recommended for Premises. 51
Table 30. PIs recommended for IS/IT. 51
Table 31. PIs recommended for Consultants. 52
Table 32. PIs recommended for Installation. 52
Table 33. PIs that affects all budget posts in Testing. 52
Table 34. PIs recommended for R&D IT. 53
Preface
This thesis concludes our Master’s degrees in Industrial Engineering and Management at the Royal Institute of Technology in Stockholm, 2012. Foremost, we would like to thank our supervisor at the Royal Institute of Technology, Professor Håkan Kullvén for his support and continuous active feedback throughout the study. We would also like to thank Professor Thomas Westin and the thesis group for their feedback during the project. We are also very grateful for the support from our mentor at the case company, and all other personnel supporting the thesis and participating in interviews. Without them, this study would not have been possible.
Carl Lundberg Sami Almehdi Österman Stockholm, June 18th 2012
Acronyms & definitions
Acronym
AO Application Operations ABM Activity Based Management ABC Activity Based Costing CAPEX Capital Expenditure CIO Chief Information Officer CIS Client Infrastructure Services GB Giga byte
DI Development and Implementation
HSD Hostage Shared Desktop (virtual desktop solution) IT Information Technology
MA Management Accounting MCS Management Control System MNC Multinational Corporation KPI Key Performance Indicator PI Performance Indicator
PMS Performance Management System OS Operating System
PO Platform Operations PC Personal Computer
PDU Product Development Unit R&D Research & Development
SLO Service Level Objective (e.g. service level of 95 or 99 percent) SW Software
TPC Third Party Connection TTM Time To Market
Definitions
Build Programming code generated for simulation in R&D units Client A personal computer
Cost driver An activity that results in the consumption of a firm's resources Level 2 A company and supplier specific support level
1 of 86
1. Introduction
In this chapter, an introduction to the subject is presented, followed by a brief description of the situation and the issue at the department under study. This is followed by stating the aims and objectives of the study as well as its delimitations. In the end of the chapter an outline of the thesis is presented.
1.1 Background
In recent years, companies have begun recognizing the potential of IT as an enabler of business value. According to a study made by A.T. Kearney, IT costs make up about 1-‐7 % of a company’s total sales depending on industry (Buchta, 2010). The telecommunications industry lies in the upper half of this range and though this may still be regarded as a relatively small number, for multinational companies (MNCs) with revenues in the billions of dollars, the cost of IT is significant (Buchta, 2010), see Figure 1.
Figure 1 IT expenses / IT budget as a percentage of sales in 2002. Source: Buchta (2010)
IT value is measurable and thus controllable (Buchta, 2010), thereby necessarily relating to organizational performance. Like any other activity, IT needs to be managed. D’Auria (2009) points out the need to deliver IT capability with less money, in other words being more effective, and having better control of the IT budget. D’Auria, 2009 points out that CIOs at top-‐performing companies are dealing with cost reduction by doing deep cuts across the board rather than performing better IT cost-‐control initiatives. Further, Chief Information Officers (CIOs) are often asked to deliver state of the art systems but are at the same time under constant pressure to contribute to organization-‐wide cost cutting (Varghese and Kurien, 2004). In order to control either an entire organization or a certain division, such as the IT function, management control systems (MCS) may be used. Any activity that managers do to help ensure that strategies and plans are executed is described as a MCS (Merchant & Van der Stede, 2007). Such systems are established to strive for organizational goals by implementing overall corporate strategies. To make certain that strategies are in fact implemented, activities and performance need to be measured and controlled. A famous citation made by performance consultant H. James Harrington explains why:
“Measurement is the first step that leads to control and eventually to improvement. If you can't measure something, you can't understand it. If you can't understand it, you can't control
it. If you can't control it, you can't improve it." (Levy, 1999)
2 of 86 Anthony & Govindarajan (2007) explain that MCSs cover both financial and non-‐financial performance measures, thereby comprising components such as net income and return on equity, as well as factors such as quality, market share and customer satisfaction.
A commonly used MCS is key performance indicators (KPIs), which enable top management to steer the organization according to corporate strategies (Marginson, 2002). Many theories exist of how KPIs should be used but little information is available of how companies operate with them, according to a Master’s Thesis authored by Falck & Karlsson (2011). KPIs are set by senior management in order to monitor the performance of the organization and serve similar to instruments in a vehicle, monitoring and controlling the business (Anthony & Govindarajan, 2007). Though widely used by organizations, KPIs do not explain why certain targets are reached or not (Rodriguez et al, 2009). Consequently, if senior managers lack understanding of the reasons of deviating performance further down in the organization, they will not be able to adequately control the desired performance and thus steering will not be optimal.
1.2 Problem description and research question
In the company of this study, the business control support function is responsible for securing the IT department’s financial governance and its financial model, also being accountable for the transparency and quality of financial data. A central area of work for the business controller function is to track financial performance and to perform analyses on cost trends and cost drivers (Company internal A, 2012).
The IT department under study has a headcount of approximately 1800 globally, excluding consultants. The six main sub divisions with corresponding sub units are therefore not only different in the activities engaged, but also geographically apart. In addition, top management lacks information of financial performance from the operating level. Mainly they lack information of what drivers of costs that exist. Even though scorecards and dashboards are prevalent in the organization today i.e. the steering system is in place, the content in these documents is not fully developed. This leads to an information gap of what performance indicators that should be used. In order to address this issue, this study focuses on answering the following research question:
How can cost drivers be used to develop performance indicators, in order for management to improve the cost control of the IT department?
1.3 The aim of the study
The aim of this study is to identify cost drivers to enable an improved control of the financial performance of an IT department.
3 of 86
1.4 Delimitations
In order for this Master’s Thesis to have a precise focus, two delimitations were made. The first delimitation was that the thesis only focuses on the company’s IT department and the second delimitation involved the global scope of the study since the organization has a global reach: information was, due to practical reasons, mainly gathered in Sweden but with a few complimentary international inputs. The results are still to be considered applicable on a global scale.
1.5 Outline
There are seven chapters in this thesis, consisting of the introduction, theoretical framework, methodology, case study, findings, analysis of findings, and recommendation and conclusions.
Introduction – In chapter 1, the background and the research problem of the thesis are
presented.
Theoretical framework -‐ Chapter 2 contains theory on among other things cost drivers, the
IT budget and performance indicators. This is presented to provide the reader with a holistic understanding of the relevant context that is covered and to fully understand the study. A reader more interested in the findings may go on reading section 5.
Methodology – In chapter 3, a thorough description is given of the structure of the research
process used in order to reach the thesis aim and objectives.
Case study –In chapter 4 the company is presented in its present state, the organizational
structure and relevant parts that are included in this research are elaborated on.
Findings – The findings in chapter 5 presents data gathered from interviews.
Analysis of findings – In chapter 6, the findings are analyzed to make data comprehensible.
Recommendation and conclusions – Chapter 7 presents the results from the analyzed
findings and the final proposed performance indicators to be tracked. A discussion about the research process, its limitations and results are made. The research question is answered and both suggestions for further research and recommendations for future work for the case company are made.
4 of 86
2. Theoretical framework
To meet the objective of this study and to ensure that it is in line with the proposed scientific approach, a theoretical framework is presented. It acts as a foundation for analysis of the derived results in this study and is presented in this chapter. The intention of this chapter is to create a condensed holistic view of the theoretical knowledge in the areas of activity based management, IT budgets, cost drivers, and performance indicators. For further reading, the reader is referred to the list of references in the end.
2.1 Budgets in management accounting
Management accounting (MA) is according to Chenhall, cited in by Malmi & Brown (2008) a “collection of practices such as budgeting or product costing”, while management accounting systems are the systematic use of MA to achieve some goal. These systems include budgets, which are central in data gathering of this study and for cost item selection. Though budgets may be considered to be forecasts of anticipated revenues and expenses, they may also be viewed as a “basis for subsequent evaluation of performance” typically by comparing budgeted with actual results (Madegowda, 2007). However, not all are fond of budgets and some researchers advocate their abolishment. Hope & Fraser (2003) claim budgets prevent the “long-‐running efforts to transform organizations from centralized hierarchies into devolved networks that allow for nimble adjustments to market conditions.” While this may be true, budgetary control systems are still viewed as an effective management tool for minimizing cost and maximizing revenues and profits (Madegowda, 2007).
2.2 Activity based management
According to Trussel & Bitner (1998), one way for companies to take into account both cost management and performance evaluation is by applying activity-‐based management (ABM). Hixon (1995), cited by Armstrong (2002), defines ABM as “the management and control of enterprise performance using activity-‐based information as the primary means of decision support”. Trussel & Bitner (1998) argue that ABM remedies the issue of cost management systems often being ignored when implementing strategic management initiatives. ABM provides both a cost view and a process view and the former involves activity based costing (ABC). The cost view introduces costs of core activities, products and other cost objects, while the process view involves developing financial and non-‐financial key performance indicators for performance evaluation (Trussel & Bitner, 1998). Vazakidis & Karagiannis (2011) choose to summarize ABM as consisting of cost driver analysis, activity analysis and performance measurement. In the cost view, cost drivers are used to assign costs and to clarify what is causing resources to be consumed by an activity and cost object. In the second view, managers need to decide how the performance for each process and activity should be measured (Trussel & Bitner, 1998). This paper does touch on ABM and in fact analyses cost drivers to subsequently develop performance indicators. However, this paper does not entirely apply ABM since Trussel & Bitner’s (1998) model departs in the ABC model see section 2.2.1.
2.2.1 Cost drivers and the activity-‐based costing model
From a manager’s perspective, literature on cost drivers is not necessarily easy to understand, as previous researchers have approached cost driver analysis using linear regression, cost driver optimization models and cost driver correlation (Yang et al. 2010). Even though the identification of cost drivers is central in this study, the study of cost
5 of 86 allocation for specific products is not. This is because this study aims to find out what causes lie behind selected cost items, rather than focusing on how costs may be related to specific products. Therefore, even though the ABC model is not applied here, literature on cost drivers relating to the model is still relevant and useful.
Available theory on cost drivers is related primarily to ABC (Homburg, 2001; Babad & Balachandran, 1993; Johnston & Banker, 1993; Foster & Gupta, 1990). Essentially, the ABC model is a cost management system that traces costs to activities and then to products (Partovi, 1991). The model identifies activities that use overhead resources and collects costs of activities into cost pools. Then, cost drivers are determined to establish how much resources each activity consumes. In the last step, overhead costs are allocated to cost objects, in proportion to their respective cost driver demand (Homburg, 2001). In other words, a cost driver may be defined as “an activity which results in the consumption of a firm's resources” (Babad & Balachandran, 1993; Partovi, 1991). The problem with this approach is that the proportions of the activity actually consumed by a specific product, does not universally correspond with a single cost driver (Marx, 2009).
2.2.2 Types of cost drivers
Mainly two kinds of cost drivers exist: resource drivers and activity drivers. The first refers to the “contribution of the quantity of resources used to the cost of an activity”, and the second refers to the “costs incurred by the activities required to complete a specific task or project” (QFinance, 2012). A simple example of a resource driver would be one kilogram of flour for bread production and an activity driver could be number of inspections or inspection hours (Fong, 2011). Cost drivers may be further divided into organizational activities: structural and executional cost drivers Fong (2011). Furthermore, according to the same author, structural cost drivers relate to business strategic choices about the economic structure, scale of operation or complexity of products, while executional cost drivers relate to execution of business activities such as utilization of employees, product design and manufacturing. Fong (2009) provides an example of organizational cost drivers and operational drivers, the latter classically used in activity-‐based costing, see Table 1 and
Table 2. In addition, Fong (2011) argues that even though organizational activities determine the operational activities, “the analysis of operational activities and cost drivers can be used to suggest strategic choices of organizational activities and cost drivers”. He clarifies his reasoning with an example: “the number of material moves as a measure of the materials moving activity by individual products suggests that resource spending can be reduced if the plant layout is redesigned, to reduce the number of moves required (Fong, 2011).
6 of 86
Table 1 Table exemplifying structural and executional cost drivers. Source: Fong (2011)
Table 2 Table exemplifying operational drivers, commonly used in the ABC model. Source: Fong (2011)
2.2.3 Categorizing cost drivers
According to Cooper, cited by Homburg (2001), using the ABC model primarily entails deciding on the number of cost drivers and which ones to use. The number of identified cost drivers may be large, posing certain problems. On one hand, the number of cost drivers needs to be high enough to accurately reflect the effect on the cost item, and on the other hand small enough to be manageable to work with. Further, Hiromoto, Merchant and Shields, cited by Homburg (2001), claim that a low number of cost drivers is less costly to handle and more useful and clear to managers. Ferrin & Plank (2002) build on this issue and propose a method to categorize and reduce cost drivers. By content analysis, cost drivers may be reduced by removing those mentioned more than twice and then categorized. The formed categories in Ferrin & Plank’s (2002) study constitute “their best estimation of the relationships of cost drivers”.
2.3 Budget elements and cost drivers in IT organizations
As previously discussed, cost items are induced by certain activities or events. Naturally, this also applies for the IT department. In many organizations, personnel, payroll expenses,
7 of 86 and benefits make up the major portion of IT costs and IT often also has the highest overhead costs (Cassidy & Cassidy, 2010). Tracking expenses allows IT teams to optimize resources, reduce IT spend and at the same time increase services, resulting in an improved bottom line (Vanouver, cited by Dubie, 2009). According to an article in the McKinsey Quarterly (Appel et al, 2005), managers in IT organizations should focus on big budget items and the largest cost drivers that consume a significant amount of resources. This is according to Appel et al. (2005) one of the ways of achieving cost transparency in the IT organization, considered an aspiration for IT managers.
According to Cassidy & Cassidy (2010), costs in IT may be divided into operating costs (capital items) and capital costs (expenditure items). In the case of capital items such as servers, the value is depreciated over the item’s lifetime and thus induces a yearly cost. In the case of expenditure items, such as printing paper, the cost is “absorbed by the business during a single fiscal period” (Cassidy & Cassidy, 2010). The authors present a table of common IT budget components, viewed in Table 3.
IT budget category
Amortization Application software and software maintenance
Business continuity Consulting and contract labor
Depreciation Education, training and seminars, books and subscriptions, associations
Facilities and utilities Freight and postage
Lease expenses Network infrastructure
Office supplies Other
Outside services PC's, workstations, and laptops Personnel benefits, fringe Personnel bonuses, overtime Personnel salaries, labor Printers
Recruiting, fees, and ads Repairs, maintenance, and replacement parts
Security Servers and mainframes
Storage Telecom, telephone, long distance, and video
conference Travel expenses, meals, and meetings
Table 3 Typical budget categories related to IT organizations. Source: Cassidy & Cassidy (2010)
Baschab & Piot (2007) present a list of key IT cost drivers. This is a relevant source for this study as it not only reveals key IT cost drivers, but also the areas affected by these drivers. Table 4 displays factors to be taken into consideration when analyzing corporate IT spending and how these factors drive the IT expenditures of the company (Baschab & Piot, 2007).
8 of 86
Key Drivers of IT Costs
It Cost Driver Comments Areas Affected
Industry • Some industries dictate higher IT spending e.g.,
transportation—airline reservation systems • General spending
Company size (sales, profitability, number of end users, type of end users)
• Company revenue
• Number of knowledge workers • Number of professionals
• General spending • Support
• Capital items
Number of computers per
knowledge worker • IT costs rise with the number of personal computers deployed • • Purchase of PCs Support
Complexity of internal
operations • Outsourcing functions should lower IT costs since no longer have to support. Cost will show up in services
• Computational intensive environments will increase IT costs
• Personnel • Hardware • Maintenance • Integration
Historical capital spending • Historical CapEx spending does not drive increased cost however increased depreciation expense will affect the IT budget, e.g.
purchasing Mainframe will affect depreciation for 3–5 years or useful life of the equipment
• Depreciation • Capital expenditures
Current economic/marketplace
condition • Economic pressures will increase need to cut IT spending • Profitable companies tend to spend more on IT
• Personnel • Overhead
Competitive initiatives • Major business transformation projects such as supply chain reengineering will precipitate major IT expenses to support
• Personnel • Software • Hardware
Demands from customers or
suppliers • Pressure from customers or suppliers for electronic information flows and other types of computer related messaging can drive up IT expenditures in the short term
• Software
Merger and acquisition activity • Acquisitions and mergers acquisitions will drive IT integration costs
• Potential economies of scale in the long term
• Personnel • Integration
Age of infrastructure • As age of infrastructure increases, cost to
support generally increases • Maintenance
9 of 86
(Continued)
Key Drivers of IT Costs
It Cost Driver Comments Areas Affected
Central versus decentralized IT
operations • Decentralized IT operations tend to increase IT spending due to lack of controls and volume discounts
• Personnel • Software • Hardware
Number of platforms • Costs increase in relation to the number of supported platforms
• Standardization of environments lowers IT costs
• Personnel • Maintenance
Application complexity • Application complexity drives higher support
costs • Maintenance
Application age • Application age drives higher support costs • Maintenance
Central versus decentralized
purchasing • Decentralized purchasing tends to increase IT spending due to lack of controls and inability to leverage purchasing volume
• Personnel • Software • Hardware
Standardization • Standardization of environment, technical
platform and tools reduces IT spending • • Hardware Support/Maintenance
Chargeback mechanism
employed • Chargeback mechanism can lower IT spending by driving more rationale behavior with business units, e.g., market pricing
• General spending
Table 4 The table presents key IT cost drivers, comments and areas affected. Source: Baschab & Piot (2007)
2.4 Performance indicators
A performance metric is a type of measurement that can be used to quantify the performance of a component of an organization (Holman, 2009). In existing literature, a key performance indicator is a term that has different meanings. Parmenter (2007) claims that many companies are working with measures that are incorrectly termed “KPIs”, which results in few organizations really being able to monitor their true KPIs. Instead, Parmenter introduces three types of performance indicators: key performance indicators (KPI), key results indicators (KRI) and performance indicators (PI). KPIs are the measures that focus on the most critical performance areas within an organization. They reflect strategic value drivers and dictate the success of the organization, while metrics represent anything that is measurable. A KPI is a metric, but a metric is not always a KPI. It is very common for an organization to have several result indicators and performance indicators, but very few KPIs (Holman, 2009). KRIs are long term measures (monthly, quarterly, yearly) which reveal the present state of the organization, and are the results of the organization's actions and used by executive management. Performance indicators (PIs) are common and are measured more frequently (weekly, daily or hourly), and describe what the organization needs to do. These last indicators are mostly used by middle management and staff simply
10 of 86 because they give valuable information of operations and provide management and staff with an understanding of which actions that need to be taken (Holman, 2009).
In existing literature, many different approaches exist on how to successfully develop, implement and review PIs. Cabrera et al. (2001) suggest that at least three conditions are necessary when developing successful performance indicators. First, the data yielded by a performance indicator are meaningful when defined by the user. Second, performance indicators are best when used as a group. The information that they provide should portray a comprehensive picture of an institutional strategic area if they are to support a strategic decision. Third, data should provide information about the input and processes associated with a particular outcome or function. (Carbera et al., 2001).
According to McNeeney (2005) the SMART test is frequently used to ensure quality of performance metrics, such as performance indicators. Table 5 shows a SMART model with issues to keep in mind when assessing performance metrics, with influence from Platt (2002). The criteria’s of the SMART model are Specific, Measurable, Achievable, Relevant and Timed (Black, 2011). Further, a performance audit done by the Commonwealth of Australia in 2011 used the SMART criteria to assess KPI. Both these independent sources point that when assessing performance indicators, the SMART criteria may be used. When developing the SMART model below, see Table 5, a SMART model provided by Platt (2002) was used.
Criteria Consideration
Specific Key question: Is there a description of a precise or specific outcome that
is linked to rate, number percentage or frequency?
-‐ Could a “reasonable person” understand the meaning of the PI? Measurable Key question: Is there a reliable system in place to measure the
performance indicator? Additional: Will the performance indicator show a trend over time?
Achievable Key question: With a reasonable amount of effort can the performance indicator be measured?
Relevant Key question: Does performance indicator link directly to the cost
driver?
Timed Key question: Does the performance indicator provide information in
time for action to be taken?
Table 5 SMART model – specific, measurable, achievable, relevant and timed. Source: Modified model from Platt (2002)
Binnendijk (1996) provides a straight forward process of how to select appropriate and useful performance indicators. The process involves developing a list of possible indicators, assessing them and selecting the best ones. The author suggests that when assessing each performance indicator, seven criteria may be used, see Table 6. With a simple scoring scale, for example 1-‐5, each performance indicator may be rated against each criterion. These ratings aid in the selection process of the performance indicators. It is however important to apply this approach flexibly and with judgment since each criteria might not be of equal importance. (Binnendijk, 1996).
