Implementing IT Financial Analysis

(1)

Implementing IT Financial Analysis

Susana Semedo Velez

Dissertação para obtenção do Grau de Mestre em

Engenharia Informática e de Computadores

Júri

Presidente: Prof. José Manuel Nunes Salvador Tribolet

Orientador: Prof. Miguel Leitão Bignolas Mira da Silva

Vogal:

Prof. José Manuel Costa Dias de Figueiredo

(2)

(3)

Agradecimentos

Ao meu orientador, Prof. Miguel Mira da Silva, pelo apoio, disponibilidade e preciosas sugestões que ajudaram a levar este trabalho a bom porto.

À minha família, fonte de força, inspiração e boa-disposição. Bem sei que uma linha aqui não é muito, mas é um começo.

Ao Nuno, pela presença nas muitas horas de desabafos e pelos risos e alegrias partilhados nas restantes.

Aos meus amigos, que me lembram que a vida é um recreio, que há tempo para tudo e que todos os dias trazem uma nova experiência.

Ao Carlos, pela inﬁnita paciência a responder-me.

Ao grupo do SAS, pela disponibilidade e apoio.

Ao Marco e respectivo grupo, que seria de mim sem o vosso tempo e atenção?

(4)

(5)

Abstract

The highly competitive scenario in the enterprise world associated with the dif-ficult world economic situation drives organizations’ top management to reduce costs. This specially affects IT, since its budget is very often hard to justify. To reduce costs successfully and with long term consequences it is necessary to know where the real costs are. To that effect there are several cost models available.

The problem addressed in this thesis is that although there are cost models that can identify IT costs, namely Activity-Based Costing, it does not handle direct costs, which also contain a fair amount of information, and models that operate with direct costs do not cope with indirect costs.

The proposal to solve this problem is based on a hybrid cost model that combines the advantages of both and therefore provides better cost information. In order to test the hybrid cost model a prototype was developed and data from a real organization was used.

The results obtained with the prototype led to the conclusion that this proposal makes it possible to determine the cost of activities using both direct and indirect costs.

Keywords

(6)

(7)

Resumo

O cenário de grande competitividade empresarial, associado à complicada situ-ação económica mundial leva os gestores de topo a querer reduzir custos. Os departamentos de Sistemas de Informação são especialmente afectados por esta situação, uma vez que o seu orçamento é, muitas vezes, difícil de justiﬁcar. Para a redução de custos ser feita com sucesso e consequências positivas a longo prazo é necessário saber onde se concentram os verdadeiros gastos e para esse ﬁm existem modelos de distribuição de custos.

O problema que esta tese pretende endereçar é que apesar de existirem modelos de custeio que conseguem identiﬁcar os custos dos departamentos de Sistemas de Informção, nomeadamente oActivity-Based Costing, este não consegue lidar

com os custos directos, que também contêm grande quantidade de informação e os modelos que operam com custos directos não conseguem lidar com os custos indirectos.

A proposta para resolver este problema consiste num modelo de custeio híbrido que combina as vantagens dos dois tipos de modelos e portanto dá mais e melhor informação sobre custos. De modo a testar o modelo de custeio híbrido foi desenvolvido um protótipo e foram usados dados de uma organização real.

Os resultados obtidos com o protótipo levam-nos a concluir que esta proposta faz com que seja possível determinar o custo de actividades usando tanto custos directos como indirectos.

Palavras-Chave

Activity-Based Costing, Gestão Financeira do ITIL, Modelo de Custeio, Redução

(8)

(9)

List of Figures

1 Main priorities for organizations in 2010 (IDC, 2010a). . . 2

2 Charge the activity or the client? . . . 4

3 The Action Research cycle (Baskerville, 1999). . . 5

4 ITIL v2 overview (OGC, 2005). . . 10

5 Service Delivery overview (OGC, 2005). . . 11

6 The Activity-Based Costing method (Hicks, 2002). . . 20

7 Beneﬁts of using ABC in various areas (Turney, 1990). . . 21

8 Cost attribution using the Direct Method. . . 24

9 Cost attribution using ABC. . . 25

10 Costs by month from the Financial Management System (Mendes, 2009). . . 27

11 Direct and indirect costs in the Financial Management System (Mendes, 2009). . . 29

12 Service costs in the Financial Management System (Mendes, 2009). 29 13 Direct and indirect costs in Customers in the Financial Manage-ment System (Mendes, 2009). . . 30

14 The conceptual model of SAS ABM (SAS, 2009a). . . 34

15 A conceptual cube of SAS ABM (SAS, 2009a). . . 34

16 The DSI model in SAS ABM. . . 36

17 Diﬀerent types of costs in SAS ABM (model view). . . 37

(14)

19 Cube intersecting the Supplier and Cost Type dimensions (graph

view). . . 38

20 Direct costs in a cube that intersects customer and service di-mensions (grid view). . . 39

21 Theoretical model for the proposal. . . 41

22 Activity costs after the ﬁrst step of the distribution. . . 43

23 Product cost after distributing the direct cost. . . 43

24 Cost allocation using the hybrid model. . . 44

25 Hybrid model with 4 levels. . . 45

26 Data model for the prototype. . . 48

27 The cost attribution screen. . . 50

28 Cost ﬂow for IC (table view). . . 50

29 Cost ﬂow for IC (graph view). . . 51

30 Cost ﬂow for DC (graph view). . . 51

31 Product ﬂow for P1 (table view). . . 52

32 Product ﬂow for P1 (graph view). . . 52

33 Direct and indirect costs from Company X. . . 57

34 Sample of activity costs for Company X. . . 57

35 Project costs for Company X. . . 57

(15)

List of Tables

1 Department costs for the Direct Method example. . . 15

2 The Direct Method (Stinson, 2002). . . 16

3 Organization costs for the ABC example. . . 19

4 Cost drivers for the ABC example. . . 19

5 Activity drivers for the ABC example. . . 19

6 Costs for the example for Section 3. . . 23

7 Cost drivers for the example for Section 3. . . 24

8 Activity drivers for the example for Section 3. . . 25

(16)

(17)

List of Acronyms

ABC Activity-Based Costing

ABC/M Activity-Based Costing/Management

ABM Activity-Based Management

CIO Chief Information Oﬃcer

DC Direct Cost

DSI Direcção de Sistemas de Informação

DW Data Warehouse

ERP Enterprise Resource Planning

IC Indirect Cost

IS Information Systems

IT Information Technology

(18)

(19)

1 Introduction

“ Beware of little expenses: a small leak will sink a great ship.”

- Benjamin Franklin

Financial matters assume a highlight position in organizations’ aﬀairs. Organi-zations are faced with the challenge to adapt to rapid changes in the environment and to use their resources as eﬃciently as possible, in order to gain competitive advantage.

According to an IDC1 _{survey, ﬁnancial departments recognize that IS play an} important part in the organizations welfare (IDC, 2010b) by providing means to obtain management information that can improve the quality of the deci-sions. This situation challenges CIOs to contribute to the business’ good per-formance, as they must deliver high-quality services, improve IT contribution to the business and innovate, but always under the constraint of having to re-duce costs (IDC, 2010a), especially basic operating expenses, which represent approximately 70% of IT budgets (Cullen & Murphy, 2008).

Reducing costs is a priority for most organizations, as shown in Figure 1. This priority is reﬂected through all the organizational units and IT is no exception. Funding to IT has been decreasing (IDC, 2008) and according to a CIO Exec-utive Council2 _{survey, 40% of CIOs reported that they were under "very high"} or "signiﬁcant" pressure to reduce costs (Rosenbaum, 2006).

1

http://www.idc.com/

2

(20)

0% 10% 20% 30% 40% 50% 60% 70% Product/ Service Inovation Process Inovation/Automation Client Service Operational EBiciency Reduce Costs Increase ProBit 2008 2009 2010

Figure 1: Main priorities for organizations in 2010 (IDC, 2010a).

It is still important to note that organizations are no longer trying to reduce costs to stay aﬂoat, but as an attempt to make business grow, by still providing quality services with lower funding, which means doing more with less (IDC, 2010a).

The first step to reduce costs in any sector, but in IT in particular, is to un-derstand these costs. It is not possible to address a problem that one is not familiar with. As such, it is not possible to reduce costs effectively and with long-term results if one does not know where to save (Gruman, 2006). Without a deeper knowledge of costs, the plans to reduce costs will most likely lead to cuts in areas that will affect the organization’s performance (Michaels, 2008).

(21)

ITIL is a set of concepts and practices for IT management, development and operations. Included in the various topics that ITIL addresses is IT Financial Management (OGC, 2005). ITIL defends that IT ﬁnancial management should be done in four main steps, among which can be found IT accounting. This process enables an organization to calculate the costs of providing IT services with the use of a cost model.

From the various existing cost models there is one, Activity-Based Costing (Hicks, 2002), that best fits this purpose. By associating activities with the use of specific IT resources (Cameron, 2002), Activity-Based Costing techniques are capable of distributing IT costs in an effective and exact manner (Mendes, 2009).

However, there are several questions to take into consideration. Activity-Based Costing is a cost model that deals quite well with indirect cost, but an orga-nization does not exclusively incur in indirect costs. Direct costs are a source of information that an organization cannot ignore, just as indirect costs cannot be overlooked. In a situation where an organization has both direct and indi-rect costs, which is the most common situation, how should it deal with them? Management wants to know how much it costs to perform an activity, which implies that Activity-Based Costing is needed, but it also wants to know where to count the direct costs.

1.1 Problem Summary

(22)

Figure 2: Charge the activity or the client?

If the expense counts only for the client, it will not be reﬂected in the activity cost. If the expense counts solely for the activity, then the information that the cost should be associated to a speciﬁc client is lost. The expense cannot count in both, otherwise it will distort the total cost, by counting the cost twice.

1.2 Research Methodology

(23)

Specifying Learning

Diagnosing

Action Taking Evaluating

Action Planning

Figure 3: The Action Research cycle (Baskerville, 1999).

1. Diagnosing: This phase corresponds to the identiﬁcation of the problems underlying the organization’s desire to change and also the interpretation of the organizational problem. This analysis leads to the development of theories about the nature and scope of the problem.

2. Action Planning: Involves the speciﬁcation of the measures to be taken to improve the problem. This phase includes the decision of the organiza-tional actions that will help to decrease the problem, that is, the planning of the actions to be taken.

3. Action Taking: In this phase, the actions planned in the previous phase

are implemented and monitored. The researcher intervenes in the organi-zation’s context, leading to changes.

4. Evaluating: The researcher determines if the implemented changes have

(24)

necessary to perform adjustments to both the plan and the theory about the problem.

5. Specifying Learning: Finally, this phase corresponds to the

documen-tation of the knowledge acquired during the investigation. In the cases where the taken actions were not eﬀective with the problem, the obtained knowledge should be used in the next Action Research iteration.

The fact that it is a cyclical process, allows the acquired knowledge in every cycle to be applied in the next cycle of the research.

1.3 Proposal Summary

To ﬁnd a proposal to improve this situation three research cycles were held.

The ﬁrst research cycle was done using a tool that was developed in the context of a Master’s Thesis. The purpose of this iteration was to evaluate a simple tool’s eﬀectiveness in dealing with both direct and indirect costs.

The second research cycle was done using a professional tool and the purpose of this cycle was to explore the data exploration potentialities of that tool in separating and evidencing the inﬂuence of both types of costs.

Both experiences were unsuccessful so a hybrid cost model was proposed that combines both the advantages of using Activity-Based Costing, which is the identiﬁcation of overhead costs, with the simplicity of a direct cost allocation method. This model was used as a foundation for the development of a proto-type that represents the theoretical model. The protoproto-type was then used with data from a real-world company, which provided material for the conclusions presented.

(25)

1.4 Document Structure

The next section of this document, Section 2, Related Work, focuses on the various topics necessary to understand the area of investigation.

The problem at hands in then clariﬁed in Section 3, Problem. It is important to note that the problem presented here is the basis for the diagnosing phases for all the research cycles.

The following two sections represent two cycles of the research methodology. Section 4 is the First Research Cycle and Section 5 represents the Second Re-search Cycle.

The following sections represent the third, and last, research cycle. The diag-nosing phase is represented by the already mentioned Section 3. The action planning phase is represented by Section 6, Proposal, and here is where the proposal for improving the problem is presented. The next section, Section 7, Prototype, represents the action taking phase and is dedicated to the pro-totype that implements the solution presented in the previous section. The section that follows, Section 8, Third Research Cycle, reviews all the phases of the research methodology for this cycle and adds the evaluation and specifying learning phases. This section includes the description of the experience held in a real-world organization to validate the solution presented, as well as the corresponding results.

(26)

(27)

2 Related Work

In order to fully understand the problem at hands and to formulate a relevant proposal, there are two topics of great interest to this investigation. The ﬁrst is IT Financial Management and the second is Cost Models.

2.1 IT Financial Management

Financial management is about raising capital for business operations and the carefully considered use of that capital. IT financial management can be inter-preted as the financial management responsibilities and activities relating to IT (Sottini, 2009). IT financial management is a deeply explored subject in the context of ITIL.

The IT Infrastructure Library (ITIL) is a vast documentation of best prac-tices for IT service management (OGC, 2005). It is a best practice framework that describes how IT resources should be organized to deliver business value, documenting the processes, functions and roles of IT Service Management. The motivation for the development of ITIL lies in the recognition that organisations are becoming increasingly dependent on IT in order to satisfy their corporate aims and meet their business needs. This leads to an increased requirement for high quality IT services.

ITIL consists of a series of books giving guidance on the provision of quality IT services, and on the accommodation and environmental facilities needed to support IT. Over 30 books where published since it was originally produced, in the late 1980’s. However, in order to make ITIL more accessible and aﬀordable, the publications were consolidated into 2 books, Service Support and Service Delivery, with ﬁve principal elements, as shown in Figure 4. The elements are:

• the business perspective;

• managing applications;

• delivery of IT services;

(28)

• manage the infrastructure.

Deliver IT Services

The Business Perspective

Support IT Services

Manage the Infrastructure Managing Applications

Figure 4: ITIL v2 overview (OGC, 2005).

The element of interest in this thesis is the delivery of IT services.

The delivery of IT services is addressed in the Service Delivery book. This book looks at the services required by the business in order to provide adequate support to the business customers. The following topics are covered, as depicted in Figure 5:

• Capacity Management;

• Financial Management of IT Services;

• Availability Management;

(29)

Service Delivery

Capacity Management

Financial Management

for IT Services _ManagementAvailability

Service Level Management

IT Service Continuity Management

Figure 5: Service Delivery overview (OGC, 2005).

The focus of this investigation lies in the topic of Financial Management of IT Services.

As mentioned, organizations are becoming more dependant of IT services and they are recognized as critical for the organization’s good performance. The increase of user demand has caused IT service costs to grow. By doing financial management as suggested by ITIL, organizations are able to understand the costs of providing a service and manage those costs in a professional manner (Sottini, 2009). It also increases confidence in setting budgets, provides accurate cost information to support IT investment decisions, leads to a more efficient use of IT resources throughout the organization and increases professionalism of staff within IT.

This process comprises three activities. They are budgeting, accounting and charging (OGC, 2005).

2.1.1 Budgeting

(30)

overspent (OGC, 2005). A prediction for the period’s expense is done, typically a year, and the eﬀective costs are regularly monitored. The aim of this process is that the actual costs match the budgeted ones.

This process has enormous influence in both tactical and strategic plans. As such, it is necessary that budgeting is done with the business goals in mind and that its communication to the organization’s management is done effectively and without conflicts. Budgeting includes periodic negotiation cycles between IT and other business departments, in which expense plans are defined and investment programs are discussed. The organization may impose restrictions to IT budget, namely:

• limits on capital expenditure, as the purchase of hardware or software;

• limits on the operational expenditure, as staﬀ costs, maintenance of com-puter hardware and software;

• limits on variance at any point in time, between actual and predicted spend;

• guidelines on how the budget must be used;

• limits on expenditures outside the organization;

• agreements on how to cope with exceptions.

Budgeting is an important process because it allows an organization to predict the necessary funding to manage IT during a determined period of time, guar-antees that it is possible to compare actual with predicted spend at any time and reduces the risk of overspending.

2.1.2 Accounting

(31)

pondered otherwise the process of accounting may become too elaborate and with costs that rapidly overcome the beneﬁts. A way to calculate the costs of IT services, in a somewhat simple fashion, is using a cost model. The cost model

should provide cost information by customer in order to facilitate the charging process. Accounting allows for an organization to:

• account for the money spent in providing IT services;

• calculate the cost of providing IT services to internal customers, such as other organization’s departments, or to external customers, as in outsourc-ing;

• perform cost-beneﬁt or Return-on-Investment analysis.

This process is extremely important, since it provides management information that justiﬁes the cost of IT services. This information is vital to guarantee that the quality of IT services is in accordance to the expenditures made to provide them.

2.1.3 Charging

As referred, accounting allows to calculate the total costs by business unit and that information may, or may not, be used to charge the services to the respec-tive customers (Silva & Martins, 2008). Charging includes a set of procedures that allows IT to recover the costs of the services provided. The charging mech-anism must be:

• Simple: it should not add any more bureaucracy.

• Fair: each client should pay the same money for the same service.

• Realistic: the charging mechanism must be designed to achieve optimal behaviour.

(32)

2.2 Cost Models

In business, retail and accounting, a cost is the value of money that has been used up to produce something, and hence is not available for use anymore (O’Sullivan & Sheffrin, 2003). The cost definition alone expresses the importance of a cost and the need for controlling and managing costs. But in order to do that, one must first know costs.

There are several types of costs. They can be classiﬁed according to their variability as:

• Fixed: expenses that do not change as a function of the activity of a business (O’Sullivan & Sheﬀrin, 2003), such as a hardware maintenance contract or a software licence.

• Variable: expenses that change in proportion to the activity of a business (Garrisonet al., 2008), such as the production of additional requirements in a project.

Another classiﬁcation for costs that is particularly interesting is in respect to how they are attributed:

• Direct: costs that can easily be associated with a particular cost object (Garrisonet al., 2008). A cost object may be a product or a customer.

• Indirect: costs that are not directly accountable to a cost object. Indirect

costs are those that are incurred in activities or services that benefit more than one customer or product (OGC, 2005). Their precise benefits to a specific customer or product are often difficult or impossible to trace. Indirect costs are also known as overhead.

(33)

It is a very useful tool since it allows an organization to be able to calculate costs, which in turn enables their control and management.

There are two cost models of particular interest to this problem and they are the Direct Method and Activity-Based Costing.

2.2.1 Direct Method

The direct method is a very simple cost model. This model considers that orga-nizational units can generally be categorized as production departments, which are directly involved in producing and distributing outputs, and service depart-ments, whose primary activity is providing services to other organizational units. The direct method allocates all costs to the production departments (Stinson, 2002). It ignores service departments even if they play a part in providing ser-vices for production departments. All costs are directly allocated to outputs, bypassing any contributions, hence the term direct. The direct method uses allocation bases to distribute the costs of service departments (Caplan, 2009). There is no set criteria to choose the allocation base. The allocation base should be what best relates the cost to its purpose, in this case, the production depart-ment.

As an example of the direct method allocation consider an organization with two production departments and two service departments. Assume that the de-partment costs come from the number of employees working in each dede-partment and from the rent of the space occupied by each department, as shown in Table 1:

(34)

Department A will have its costs allocated on the basis of number of employees per production department, as seen below.

10 + 5 = 15

24.000∗5

15 = 16.000

24.000∗10

15 = 8.000

In the same manner, Department B will have its costs distributed on the basis of space occupied by square metres (m2), as seen in the calculations below.

75 + 75 = 150

15.000∗75

150 = 7.500

The direct method of allocating the organization’s service department costs to the production departments is shown in Table 2.

(35)

Several aspects of this example should be carefully considered. Even though all departments have aﬀected employees and occupy a determined area, these numbers are ignored when allocating service department costs using the direct method. Only the amount of the allocation base of the production departments is used in the allocation, in this case the number of employees for Department A and the area occupied for Department B. Another aspect to note is that even if one of the service departments provided services to the other service department it would be ignored and its cost would still be allocated to the production department. The ﬁnal aspect to note is that after all allocations have been completed, all departmental costs are limited to the two production departments.

For the purpose of this investigation a parallel can be established between service departments and activities and production departments and products or clients, meaning that the direct method overlooks all activities existing between the costs of the organization and its products.

Although this method is very simple and quite aﬀordable for an organization to implement, it is less accurate than other methods, precisely because it ignores all that is between the cost and the output. This leads to distorted product and service costs.

2.2.2 Activity-Based Costing

Activity-Based Costing(ABC)is a concept around which it can be constructed an economic model of the business that provides the accurate and relevant cost information necessary to support business decisions of all types (Hicks, 2002).

(36)

use of ABC quickly alerted to the distortions inherent to traditional cost man-agement systems, which led to changes in strategies, processes, operations and consequently, to improve the competitive advantage of organizations that used it against organizations that did not (Turney, 1990; Aﬄeck-Graves, 2001).

Using ABC is especially interesting in IT where there are many overhead costs and a broad variety of products and services, which is a fact that ABC copes with quite well (Rau, 2005).

This model considers that costs are divided between direct and indirect. Direct costs are directly attributed to the respective products. As for indirect costs, they are attributed to activities and the activity costs are then attributed to the outputs of those activities, namely the products. The characteristic feature of ABC is how the various costs are attributed, from indirect costs to activities and from activities to products. ABC looks at activities and tries to identify the trigger that makes the workload increase or decrease. That trigger is called driver (Cokins, 2001). The driver is then used to express the activity’s contri-bution to the production of the output. An analogous process is used for cost drivers. The deﬁnition of drivers is a crucial task in ABC and it must be a reﬂexion of the organization’s reality.

The methodology to apply ABC is as following (Hicks, 2002):

• Step 1: Identify and classify activities related to products/services

pro-vided by the organization.

• Step 2: Determine costs for each activity identiﬁed in the previous step.

• Step 3: Identify the cost driver.

• Step 4: Divide total cost of each activity by products/services according to cost driver usage.

(37)

salaries and the facility’s rent, which are both indirect costs and with the values indicated in Table 3.

Table 3: Organization costs for the ABC example.

Once the organization’s activities are defined it is then necessary to identify the drivers. To assign the costs to the activities we will use the number of employees per activity to distribute the salaries’ costs and the area occupied by activity to distribute the facility’s rent. Analyzing the activities, the factor that influences the workload of sales calls is the number of calls made, so that will be the chosen driver. As for shipping, what influences this activity is the number of packages shipped so that will be the driver. All drivers for this example are shown in Tables 4 and 5.

Table 4: Cost drivers for the ABC example.

(38)

The ABC method of allocating costs is shown in Figure 6.

Figure 6: The Activity-Based Costing method (Hicks, 2002).

The advantages of using ABC include (Proctoret al. , 2006; Mendes, 2009):

• identifying unproﬁtable activities;

• identifying opportunities for cost reduction;

• allowing to perform a cost allocation which is closer to the business’ reality;

• helping in making better business decisions.

(39)

Figure 7: Beneﬁts of using ABC in various areas (Turney, 1990).

Still, there are disadvantages associated to the use of ABC (Proctoret al., 2006; Mendes, 2009):

• expensive and diﬃcult to implement and maintain;

• its complexity creates resistance to its use.

2.3 Critical Analysis

(40)

Although ITIL speciﬁes how to do IT Financial Management, few are the organi-zations that have it implemented. In a survey by Evergreen, 78% of respondents admitted they did not approve ITIL implementations due to organizational re-sistance to change and 50% indicated the fact that ITIL’s value is not proven (Evergreen, 2006). In small organizations this situation is even more evident, since the high level of complexity makes the ITIL approach quite expensive. Still, there are indications on what is the route to follow to cost reduction, namely the indication of the use of a cost model to account for the costs in an organization.

The Direct Method is the most common in organizations. It is simple and aﬀordable. However it is too simple to represent reality and that is where ABC takes a step forward. ABC is a robust cost model that can cope with most IT costs and there is vast literature on how to do it. But again, organizations are faced with various obstacles trying to implement it, evidencing the fact that there is a gap between theory and reality.

(41)

3 Problem

The need to reduce costs is present everywhere and IT departments are con-stantly faced with this challenge. To reduce costs it is necessary to know and manage them. But IT departments are complex and it is diﬃcult to distribute costs correctly and calculate an exact value of the services provided. ITIL sug-gests the use of a cost model, but which should be used? The majority of IT cost may be indirect, but there are also direct costs to consider.

The problem addressed in this thesis is that there isn’t a cost model that can deal with both direct and overhead costs. IT departments cannot be limited to use solely ABC to deﬁne the value of its activities, since it is incomplete. It does not include direct costs, which are a valuable source of information. However, using a system of direct cost allocation is even worse, since it cannot deal with the overhead costs, which represent a crucial part in the total costs. In short, we want to ﬁnd a way to deal with direct and overhead costs without having two separate systems.

To better understand the limitations of both cost models, an example is in order. Consider a universe where there are two activities: Activity A and Activity B. Consider also that there are two products: Product P1 and Product P2. Consider now a direct cost (DC) associated to P1 and an indirect cost (IC). The respective values are indicated in Table 6.

Table 6: Costs for the example for Section 3.

(42)

Figure 8: Cost attribution using the Direct Method.

Although simple enough, is far from accurate. The direct cost is easily allocated to its correspondent product, but there is no deﬁned way to treat the indirect cost, so in this case it was simply divided equally by all activities. One obvious disadvantage is that there is no idea of what activities are performed or how much they cost. Therefore it is safe to say that this system does not provide enough information and does not fulﬁl the purpose of this investigation.

Now consider the same universe, where IC has a 50% driver for each activity. Activity A has a 70% driver for P1 and 30% for P2 and Activity B has a 50% driver for each product. This situation is depicted in Tables 7 and 8. The direct cost was incurred while performing Activity A.

(43)

Table 8: Activity drivers for the example for Section 3.

Using ABC, this example results in the situation depicted in Figure 9.

Figure 9: Cost attribution using ABC.

The products’ costs show the difference between both models, and although ABC gives a more accurate figure, it is still not correct. By analysing closely Figure 9 it is possible to see that Activity A receives costs from DC and IC, which add up to 450. This value is then distributed accordingly by both P1 and P2. This means that the contribution from DC is being allocated to P2 instead of being totally attributed to P1. This situation proves that ABC also does not fulfil the purpose of this investigation although it is excellent in treating indirect costs.

(44)

(45)

4 First Research Cycle

The ﬁrst research cycle is based on a Master’s Thesis developed in the academic year of 2008/2009 with the purpose of controlling IT costs using ABC and sending to each business unit, on a monthly basis, the real costs of IT services used, separated by user and by service (Mendes, 2009). As a result, a tool that implements ABC was developed and tested in the Direcção de Sistemas de Informação (DSI) of Turismo de Portugal. This tool was called Financial Management System and allows the user to manage budgets, costs, services and customers among other functions.

In order to implement the solution proposed in the referred Master’s Thesis it was necessary the cooperation of the DSI by giving access to all the invoices and the deﬁnition of services provided by the organization. The Financial Man-agement System registers the costs, that the user must identify as direct or indirect, and according to the information in the invoices, namely supplier and cost type, distributes the costs accordingly. The result of the distribution can be consulted through graphs, as seen in Figure 10.

Figure 10: Costs by month from the Financial Management System (Mendes, 2009).

(46)

4.1 Diagnosing

As mentioned in Section 3, the problem addressed in this thesis is that there isn’t a cost model that can deal with direct and indirect costs.

In this iteration, the DSI is the universe studied. Using data form the DSI, the Financial Management System provides the costs for the services deﬁned as well as for the customers by performing an ABC distribution. The problem in this context is not knowing how the direct costs are processed and how they inﬂuence the calculation of the costs of services and customers.

4.2 Action Planning

This tool was chosen as a starting point to ﬁnd a solution for this problem because it is a simple implementation of ABC, which also considers direct costs. There was also the fact that it was possible to contact the person that developed the work, including the Financial Management System, and therefore have more support.

A few meetings were held in order to understand the concepts and learn how to work with the Financial Management System. It is important to note that this implementation considers IT services as activities and customers instead of products.

In the course of the meetings, we were able to gather the information needed to represent the context of the DSI in the Financial Management System. This information included the list of customers, services and gathering the invoices necessary. There was also the deﬁnition of which drivers to apply in each situ-ation.

4.3 Action Taking

(47)

The various costs are listed and it is possible to distinguish the direct costs from the indirect costs, as seen in Figure 11.

Figure 11: Direct and indirect costs in the Financial Management System (Mendes, 2009).

At the service level, the direct costs have no inﬂuence in the service cost, as shown in Figure 12. The total cost of services is the sum of indirect costs.

Figure 12: Service costs in the Financial Management System (Mendes, 2009).

(48)

Figure 13: Direct and indirect costs in Customers in the Financial Management System (Mendes, 2009).

The experience with the Financial Management System showed that the direct costs can be seen at both cost and customer level.

4.4 Evaluating

In this research cycle the main interest was to evaluate the tool’s eﬀectiveness in dealing with both direct and indirect costs.

The Financial Management System allows direct and indirect costs. It handles indirect costs with the use of ABC. The direct costs are represented and count for the customer costs, and it is possible to separate direct cost contribution from indirect.

(49)

Furthermore, this system is somewhat limited when it comes to data exploration. It answers only the most direct questions. But by being simple, the Financial Management System was quickly learned and accepted by its users.

However its simplicity may be the reason why it cannot accommodate the direct costs, so for the next iteration, a more complex tool will be considered.

4.5 Learning

There are important notes to take from this experience. First, the fact that this is a simple and eﬀective implementation of ABC shows that it is possible to do it in a non-expensive manner. But most importantly this fact is valid not only for ABC, but for any cost model implementation.

Furthermore, by being simple, a cost model implementation becomes more at-tractive to users, which is an important factor for success.

(50)

(51)

5 Second Research Cycle

There are several companies selling tools based on ABC and SAS3 _{is one of} them. SAS is a company that intends to turn existing data into knowledge allowing businesses to respond to changes. SAS oﬀers several products in the Financial Intelligence area and the point of interest resides here.

The fact that SAS has such a tool and favours an ABM approach seemed to make it a favourable place to have a good experience. Therefore, SAS was contacted with the purpose of studying its tool, the SAS Activity-Based Management.

SAS ABM

The SAS Activity-Based Management (SAS ABM) had its origin in the tool OROS, that was used to build cost models based on ABC. This software be-longed to a company named ABC Technologies, which was bought by SAS in April 2002. OROS is no longer sold by SAS and became obsolete with the development of SAS ABM (Blocher, 2004).

SAS ABM is an analytic application that models business processes to deter-mine cost, proﬁtability and drivers (SAS, 2009b). According to SAS, with this solution it is possible to obtain better cost information that can be understood by employees and not just ﬁnance experts, improve processes by identifying ac-tivities that don’t add value and evolve the way an organization does business.

SAS ABM has four conceptual modules, but for the purpose of this investigation we will concentrate only on three. They are the Resource Module, the Activity Module and the Cost Object Module. The Resource Module contains informa-tion about resource costs, the Activity Module contains informainforma-tion about work that is performed within an organization and the Cost Object Module contains information about products and services. The three modules together form a model, a representation of the organization. It is possible to associate these modules to ABC’s concept of cost, activity and product, respectively.

3

(52)

Each module contains accounts. Costs ﬂow through the model at the account level, through assignments, as seen in Figure 14. This tool allows the user to deﬁne drivers in order to make a more accurate model.

Figure 14: The conceptual model of SAS ABM (SAS, 2009a).

But what makes SAS ABM interesting is not the model. It’s the exploring possibilities.

In this tool there are two fundamental concepts. The ﬁrst is the concept of dimension, which represents a description of components of an organization, and where dimensions intersect, unique opportunities for analysis exist. The second is the concept of cube that is a conceptual container of dimensional information (SAS, 2009a). The intersection of two or more dimensions creates a cube, as seen in Figure 15, although a cube can have any number of dimensions.

(53)

(SAS, 2009a):

• Single-Stage: allows the analysis of contributions from one module back

and step-by-step allocations within the module.

• Resource: allows the analysis of contributions from the source, that is,

the Resource Module.

• Multi-Stage: allows the analysis of the assignment structure through

the model by deﬁned stages, which includes all the dimensions.

It is important to note that the cubes are not the only way of analysing the data. There is also a graph view and a grid view.

By deﬁning dimensions that mirror an organization and then intersecting them, this tool provides a powerful way to analyse the results that come from ABC.

5.1 Diagnosing

In the previous research cycle, the tool chosen to address the problem was deemed too simple. Therefore, for this iteration the problem remains the same.

For this iteration it was decided to try a professional tool. The SAS ABM with all of its features and with its data exploration capacities may be able to cope with the indirect but also with the direct costs.

After analysing the possible beneﬁts of using SAS ABM in this investigation, SAS was contacted. This is not a free software so we tried to come to an understanding with the company in order to proceed with this experiment.

5.2 Action Planning

In the ﬁrst meeting it was agreed that it would be possible to use SAS ABM, free of any charge. The team that would be working on this experience would be formed by the researcher and the expert on SAS ABM. Several meetings would be arranged in order to help the researcher learn how to work with SAS ABM.

(54)

5.3 Action Taking

In the course of this experiment 3 more meetings where held.

During these meetings, several aspects of the tool’s functioning were clarified with the expert. Moreover, these meetings were used to build a representation of the DSI in the SAS ABM. The model was built based on the documentation used in the previous iteration and was validated by the expert on SAS ABM. The model used reflected the same universe of the Financial Management System. The Resource Module has two dimensions: Cost Type and Supplier, since this is the information that was used to aggregate cost in the previous tool. Separating the resources in two different dimensions would allow for more information to be extracted from the cubes. The rest of the modules have one dimension, which consists of Services and Customers, respectively.

Having deﬁned the dimensions, the rest of the elements to complete the model where the same as used in Section 4, namely the cost elements and the drivers. The resulting model is shown in Figure 16.

Figure 16: The DSI model in SAS ABM.

(55)

ious modules matched the ﬁgures found for the diﬀerent levels in the Financial Management System.

This model accepts direct and indirect costs. It is possible to assign a cost to a product, but being a simpliﬁed representation, it is not clear the distinction between a direct and an indirect cost, as seen in Figure 17.

(a) A direct cost in SAS ABM.

(b) An indirect cost in SAS ABM.

Figure 17: Diﬀerent types of costs in SAS ABM (model view).

(56)

/Users/fluffs/Tese/myThesis/myDissertation/figures/SAS_ABM_Cost_Flow.pdf

Figure 18: A cube representing a cost ﬂow, from Cost Type to Customers (cube explorer view).

Using the cubes it was possible to have more detailed information, especially about suppliers. The Resource Cube in Figure 19, shows the costs associated to suppliers and cost types.

Figure 19: Cube intersecting the Supplier and Cost Type dimensions (graph view).

Even with an incomplete model, there are cubes that evidence the direct costs.

(57)

Figure 20: Direct costs in a cube that intersects customer and service dimensions (grid view).

The experience with the Financial Management System showed that the direct costs can be seen at both cost and customer level.

Although the model of the DSI used in this experience limited the information that could be extracted from the cubes, it was still possible to detect direct costs in the various views.

5.4 Evaluating

The purpose of exploring SAS ABM was to understand if its data exploration power could make it possible to separate and evidence the inﬂuence of both types of costs in the service costs.

This tool allows both direct and indirect costs and SAS ABM gives the user a wide range of possibilities. The fact that the user can use the cubes to explore the data and is not limited to one view is quite an evolution from the usual cost information. Also the possibility of having n dimensions within each module and the ability to intersect them is without a doubt a striking feature.

SAS ABM is a professional tool and the help of an expert makes a signiﬁcant diﬀerence for its users to be acquainted with and to use it in its full potential.

(58)

and must be defined from the start. However, an organization is not static and if a change happens, it might be significant enough to require the model to change, namely change the dimensions previously defined. Although not very straightforward, it is possible to add dimensions so that the model can still reflect the organization it is representing.

The time reserved for this experience was not enough to design a model that would highlight all the beneﬁts of using a professional tool, therefore the model presented here cannot cope with direct costs.

For the next iteration it was decided to try a new approach, one that involved thinking of the problem from the beginning and propose the solution from there and not from an existing system.

5.5 Learning

Two research cycles were done using ABC implementations. The ﬁrst tool was simple and easy to use and the second is a professional tool, more complex and with great data exploration power.

The data exploration power is a good ally in ABC implementations, since the more powerful the tool, the better cost information is provided.

(59)

6 Proposal

At the light of the learnings from the previous experiences regarding the matter of this thesis, it was decided to try a new approach. This new approach con-sisted of designing a hybrid cost model that would ﬁt best the purposes of this investigation.

The theoretical model that supports this proposal is based on the ABC model. It consists of a sequence of levels. Drivers, as in ABC, relate each level to the next. Although ABC only considers three levels, this model can be augmented with as many levels as needed, as long as there are drivers connecting them. The costs’ total is used as a control ﬁgure, to guarantee the distribution correctness.

Figure 21: Theoretical model for the proposal.

To achieve the correct distribution, we must first allocate direct costs. Direct costs have priority over indirect costs so that they can’t influence the ABC distribution. In the first level it won’t be noticeable, but in the following levels it can lead to distortions.

A direct cost has the indication of the instance in a level where it should count. It also has the information of all the instances in the previous levels where it should count. This information is crucial to get a correct distribution. If we don’t allocate direct costs ﬁrst they will be distributed in ABC and won’t be entirely charged in the correct instance.

(60)

we must allocate the direct costs, independently of where it was previously allocated and how the previous instance is related to the instances in the next level. Then we continue with the ABC distribution, but before doing it, we must note that part of the instances’ costs have already been distributed, so we must subtract all the direct costs from the value of the instances and distribute the result of the subtraction. For the next level, we apply the same procedure.

There is one situation to highlight. When considering a direct cost, we know the output to which it should be associated but it may happen that the level of the output may not be last level in the model. This means that after attributing the direct cost in the level of its output, since the direct cost is not associated to any instance of the next level, it will be used in the ABC distribution along with the indirect costs.

As an example, consider a model with three levels, the ﬁrst being the Cost Level, then the Activity Level and ﬁnally the Product Level. All drivers are represented as a percentage, although the model accepts other types of drivers, just like ABC.

Using the example from Section 3 we will perform a distribution using the hybrid cost model. The Cost Level will have two instances: DC costing 200 and IC costing 500, which amounts to a total of 700. The drivers’ quantities for IC are 50% for each activity.

We will begin the distribution with the direct costs. That means DC is the ﬁrst to be assigned. Since it is a direct cost, it only has one activity associated and that is Activity A. This activity now has a cost of 200.

(61)

Figure 22: Activity costs after the ﬁrst step of the distribution.

For the next step of the distribution, it is known that Activity A has driver quantities of 70% and 30% for Product P1 and Product P2, respectively, and Activity B has driver quantities of 50% for each product. This is the point in the distribution where the hybrid cost model differs from ABC. The direct costs are first attributed to their respective products, independently of the activity where they were attributed first. In this case, it means that P1 has a cost of 200 that comes directly from DC, as seen in Figure 23.

Figure 23: Product cost after distributing the direct cost.

(62)

it has to be taken into account that part of that value was from DC. That means that the direct cost value has to be subtracted from the activity’s value that is to be distributed. This means that Activity A will only distribute 250 of its original value in the ABC method. This results in P1 receiving a cost of 175 and P2 receiving a cost of 75 from Activity A. Activity B has no direct costs in its value, so the total activity cost will be distributed. That means that both products will receive a 125 cost from Activity B. The totals are now 500 for P1 and 200 for P2, which add up to a total of 700. The complete distribution is shown in Figure 24.

Figure 24: Cost allocation using the hybrid model.

(63)

Figure 25: Hybrid model with 4 levels.

(64)

(65)

7 Prototype

In order to test the theoretical model a prototype was developed. The main purpose of this prototype is to test if the theoretical model is practical and if it does in fact help in ﬁnding an activity’s true value. There were other factors taken into consideration, namely the usability of the prototype so that real users could later test it.

7.1 Technology

The prototype that follows was implemented in the OutSystems Agile Platform4_. Several factors inﬂuenced the choice of this technology. The most important were the development time, which is shorter than with other technologies, the time to learn, which is also shorter, and the fact that the resultant prototype is a web-based application. That means that it only needs a browser to run, which makes it highly portable.

7.2 Data Model

The prototype is the implementation of the hybrid cost model with three levels. Although more levels are possible, it is simpler to begin with an implementation closer to the classic ABC. The three levels are the same used in the previous examples, namely Cost, Activity and Product Level.

The data model for the prototype is shown in Figure 26 and comprises six entities:

4

(66)

Activity Id Name Description Value Category (FK) Product (FK) Id Name Id Cost (FK) Activity (FK) Driver Id Name Indirect Total Id Name Indirect Total Id Activity (FK) Product (FK) Driver Category

Cost CostActivityRelation Activity

ActivityProductRelation Product

Figure 26: Data model for the prototype.

• Cost: Represents a cost.

• Category: Entity that indicates whether a cost is direct or indirect.

• Activity: Represents the activities in the hybrid cost model.

• Product: Represents the products in the hybrid cost model.

• CostActivityRelation: Entity that represents the relation between a cost and an activity. This relation is characterized by the cost driver quantity.

• ActivityProductRelation: Entity that represents the relation between an activity and a product. This relation is characterized by the activity driver quantity.

7.3 Main Functions

(67)

◦ Add, update, remove and list costs;

◦ Attribute costs;

◦ Show a cost ﬂow;

◦ Generate graphs.

• Managing activities:

◦ Add, update, remove and list activities;

◦ Attribute an activity’s cost;

◦ Calculate activity cost.

• Managing products:

◦ Add, update, remove and list products;

◦ Calculate product cost;

◦ Show a product ﬂow;

◦ Generate graphs.

This prototype allows calculating accurately an activity’s cost and a product’s cost and also provides two representations for cost ﬂow and that is its most important feature.

7.4 Interface

As mentioned, it is possible to calculate activity’s and product’s costs and have diﬀerent representations for cost ﬂow.

In order to do these calculations it is necessary that the model is deﬁned. The user must provide the activities and the existing products. The costs must also be provided and categorized as direct or indirect.

(68)

Figure 27: The cost attribution screen.

When all data is inserted, the user only has to calculate the model.

Once the model is calculated, the user can consult the total costs in every level, but most importantly the user can choose to see the flow of costs. It is possible to see the flow starting in a cost and observing its distribution, called a cost flow, or choosing a product and observe which activities and costs contribute to that product, a product flow. In both flows, the user can choose to see this in the form of a table or a graph.

Resuming the example used in Section 6, if the user chooses to see the cost ﬂow for IC, the result will be the one in Figure 28.

Figure 28: Cost ﬂow for IC (table view).

(69)

inﬂuence in any product. Alternatively, the user can choose to see this in the form of a graph, and the result is shown in Figure 29.

Figure 29: Cost ﬂow for IC (graph view).

The representation for DC is shown in Figure 30.

Figure 30: Cost ﬂow for DC (graph view).

(70)

Again, using the example from Section 6, if the user chooses the product ﬂow for P1 the result will be the table seen in Figure 31.

Figure 31: Product ﬂow for P1 (table view).

Analysing the table it is possible to see the costs from activities that contributed to the product’s cost, as well as the direct costs associated. Again, there is also the graph view available, as seen in Figure 32.

(71)

(72)

(73)

8 Third Research Cycle

This section describes the experience held in a real-world organization. This organization wishes to remain anonymous, therefore it is referred to throughout this document as Company X.

8.1 Diagnosing

Company X is a company in the IT consulting and software development market that wanted to implement a cost model that would give better cost information and was especially interested in ABC. They used the Direct Method to deﬁne the cost of their products. This means that there was a lot of information about their products that could not be lost with the implementation of a new cost model. This company was a perfect example of the problem described in Section 3.

8.2 Action Planning

As soon as it was established that both parts had an interest in an experience in Company X, it was decided that the researcher would spend a few weeks in the company. This would ensure contact with the problem. There would also be daily meetings with management allowing a better understanding of the expectations for this project.

At the light of the unsuccess of the previous experiences regarding the matter of this thesis, it was decided to try a new approach. This new approach consisted of designing a hybrid cost model that would best ﬁt the purposes of this investi-gation and the company’s wishes. the hybrid cost model proposed is described in Section 6.

8.3 Action Taking

(74)

the crucial part was to define a list of activities. Since this company used the Direct Method, the costs and the projects were easily accessed, but defining the activities would be a more complicated task. As a starting point, the company’s ERP was used. Company X has a project management system incorporated in the ERP where employees indicate, for a certain period of time, the task they were working on. This list was analysed with management and employees, and irrelevant tasks were eliminated. The remaining tasks were grouped into activ-ities, which were presented to management for validation. Another important issue was the definition of drivers. Again, the ERP was of significant relevance. Using the information in the ERP it was possible to associate time to tasks, and consequently activities, therefore achieving a simple driver for the various activities. As for costs, the drivers were found through interviews that took place during meetings.

During the meetings with management the context where the model would function was also an important issue. The levels that were going to be used had to be decided in order to design a hybrid cost model, as the one proposed in Section 6, that would mirror Company X. The model was subject to the company’s management evaluation almost on a daily basis. This allowed the management’s feedback to be quickly incorporated in the proposed solution.

After a few more meetings, a suitable model was achieved and tested in Excel sheets. Once it was approved, it was decided to implement the model in the form of a more satisfactory prototype. The prototype is described in Section 7.

A sample of costs, which included direct and overhead costs, were used as input for the prototype.

(75)

Figure 33: Direct and indirect costs from Company X.

The prototype also showed the costs for the activities deﬁned, as seen in Figure 34. The costs for the activities were obtained from the distribution of indirect costs using ABC and the contribution from direct costs.

Figure 34: Sample of activity costs for Company X.

It was also possible to obtain the cost of each project. The results can be seen in Figure 35.

Figure 35: Project costs for Company X.

(76)

(a) Direct cost flow (table view).

(b) Direct cost flow (graph view).

Figure 36: A Company X direct cost ﬂow.

(77)

(a) Indirect cost flow (table view).

(b) Indirect cost flow (graph view).

Figure 37: Company X indirect cost ﬂow.

(78)

(79)

The prototype showed the inﬂuence of direct and indirect costs throughout the whole model.

8.4 Evaluating

The main purpose of this prototype was to test if the theoretical model is prac-tical and if it does in fact help in ﬁnding an activity’s true value.

The use of this prototype and the data from Company X allowed to see the diﬀerent contributions. Indirect costs were used and distributed with the use of ABC and direct costs are seen contributing directly for a product, but also correctly contributing to the value of an activity. This type of information allows the company to deﬁne the correct value of its activities and products.

It is important to highlight that this model presents a higher cost for activities than the previous tools presented, as seen in Table 9.

Table 9: Activity costs from the three tools.

The diﬀerence is caused by the contribution of direct costs in the cost of ac-tivities. The cost of products remained the same, since the previous systems already considered the contribution from direct costs in the products.

The fact that management was profoundly involved in the design of this model was a factor of success. Management, that also acts as a user, contributed with feedback during the development, which was determinant in creating a real and practical solution for the problem.

(80)

cost model, since it fulﬁls the main requirement, that is to deal with both types of costs. The model copes with overhead costs performing an ABC distribu-tion and with direct costs associating them to the correct product and to an activity. Also there is no duplication or loss of costs while calculating a correct distribution.

According the client’s point of view, here Company X, the proposed hybrid cost model is a contribution to solving the problem. It is a cost model based on ABC and with the advantages of ABC, namely the information that is extracted from the overhead costs, but that allows to maintain and use eﬀectively the informa-tion they already had from the Direct Method. It is a simple implementainforma-tion that provides two diﬀerent ways of visualizing the cost information. It also provides a cost for the activities performed. With this information it is now possible to act on it.

8.5 Learning

The result of this research cycle is a hybrid cost model that copes with both direct and indirect costs. The overhead costs are distributed by performing an ABC distribution and the direct costs are associated to the correct product and to an activity. Therefore, the proposed cost model is a contribution to solving the problem addressed in this thesis.

Another important learning from his experience is that understanding the prob-lem is the most important step in finding the solution. In the beginning of this investigation, the problem was defined in theory, but there was no contact with it. Not being totally aware of the problem was an important factor in the out-come of the first two iterations. Not knowing for sure what to look for impairs the capacity to act on it.

(81)

9 Conclusion

One of the biggest challenges for enterprises today, and IT speciﬁcally, is to reduce costs, but in order to reduce costs it is necessary to identify and control those costs. In IT it is not as straightforward as usual, since most IT costs are indirect costs, that is, the cost cannot be attributed to a single client, product or service.

ITIL provides a set of best practices for IT ﬁnancial management, that includes the use of a cost model to calculate all expenses. The cost models presented in this study were the Direct Method and Activity-Based Costing.

The Direct Method is a simple and very aﬀordable cost model, and is very common in organizations. However, this cost model is not accurate and leads to serious distortions.

Activity-Based Costing is more complex than the Direct Method but when implemented correctly can give very accurate cost information. The problem with Activity-Based Costing is that it can become quite complex and expensive and overlooks the information from direct costs.

Both methods have shown limitations when it comes to deal with the type of costs that organizations have to deal with on a daily basis: direct and indirect. The problem addressed in this thesis is that there isn’t a cost model that can deal with both direct and indirect costs and therefore there isn’t a model that can give true accurate cost information.

A ﬁrst experience with an existing tool called Financial Management System was conducted in order to understand the best way to incorporate direct cost in an ABC implementation. This experience was important to highlight that it is feasible to implement ABC without much expense and with a simple tool.

Implementing IT Financial Analysis