A case study of:
The integration of Data Analytics into Financial Audits
within a Big Four firm.
Name: Michael Trinh Student number: 11371870
Thesis supervisor: Mr. prof. dr. B.G.D. (Brendan) O'Dwyer Date: August 20, 2018
Word count: 19,244
MSc Accountancy & Control, specialization Accountancy Faculty of Economics and Business, University of Amsterdam
2 Statement of Originality
This document is written by student Michael Trinh who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3 Abstract
Purpose – Of this research it to gain insight on why Data Analytics has not been integrated more frequently in Financial Audits.
Research design and methodology – To conduct this research a case study approach has been used within a Big Four audit firm. The Data set contains ten interviews of actors within the firm that are actively involved or influenced by the integration of Data Analytics into Financial Audits. To gain a better understanding of how the involved actors influence this process, Institutional Work was used as a theoretical framework.
Findings – Suggest that a general lack of knowledge about Data Analytics are the root to today’s struggles with regards to a (further) integration of Data Analytics in Financial Audits. But this is not the only reason for the delayed (further) integration of Data Analytics into Financial Auditing. Other reasons are; (bad) experiences with Data Analytics in the past, thresholds with regards to extracting Data from clients, but also the lack of convinced support from the Senior staff.
Limitations – In this research numerous limitations could be identified, as the research methodology might affect the legitimacy of the collected results. First, the interviewees were introduced to me by other interviewees. Therefore, it is possible that the results are biased. Second, the amount of performed interviews was a relatively small sample of the total population that could have been interviewed. Third, the research does not contain interviews with external parties such as; the Dutch regulating authority. Finally, some of the interviewees were well aware that they were being voice recorded, causing them to be more conservative.
Contributions – The contribution of this research is the ability to gain more insight on why Data Analytics has not been integrated more frequently in Financial Audits. Through gaining actual insight it is possible to look into the interaction between the actors that are involved in this process. Furthermore, this research also contributes to the existing theory regarding Institutional Work, but can also serve as a starting point for (further) integrating Data Analytics into the Audit field.
4 Acknowledgements
First, I would like to thank every single interview respondent for their time and effort. Without them this research would not be possible. I would also like to thank the firm for allowing me to write this case study thesis within their firm.
Second, I would also like to thank my thesis supervisor prof. dr. B.G.D. (Brendan) O'Dwyer of the Economics and Business faculty at University of Amsterdam for his guidance.
I must also express my deep gratitude to my parents, my girlfriend and the day ones for their loyal support and continuous encouragements throughout the process of researching and writing this thesis. This accomplishment would not have been possible without them.
Finally, I would like to have some last words for the Master Family, whom I have shared a lot of ups and downs with during this past year. It has been a great ride, meeting and getting to know every single one of you has been my pleasure. Despite the fact that living and studying in Amsterdam for the last two years has been a truly unreplaceable life experience, it was you guys who made going to school fun.
I am grateful. Michael Trinh
5 Contents 1 Introduction ... 8 1.1 Background ... 9 1.2 Research question ... 10 1.3 Research contribution ... 10 1.4 Paper structure ... 11 2 Literature ... 12
2.1 Data Analytics and Big Data ... 12
2.2 Impact Data Analytics on audit behavior ... 13
3 Theory ... 16
3.1 Theoretical framework: Institutional work ... 16
3.2 Three stages of Institutional work ... 17
3.2.1 Creating institutions ... 17 3.2.2 Maintaining institutions... 20 3.2.3 Disrupting institutions ... 22 4 Methodology ... 24 4.1 Research setting... 24 4.2 Data collection... 24 4.3 Data processing ... 27 4.4 Data analysis ... 27 5 Descriptive Analysis ... 30 5.1 Data Analytics ... 30
5.1.1 Definition of Data Analytics ... 30
5.2 Changes and Impact of Data Analytics ... 31
5.2.1 Audit approach ... 31
6
5.2.3 The Potential of Data Analytics ... 32
5.2.4 Data Collection and Availability of Data ... 33
5.2.5 Tools and Innovation ... 34
5.2.6 Expectations of Data Analytics ... 34
5.2.7 Knowledge ... 35
5.2.8 Education ... 36
5.2.9 Structure within the firm ... 36
5.2.10 Procedures and standards ... 37
5.2.11 Guidance during Application process ... 37
5.3 Application Strategies and Actions taken ... 38
5.3.1 (Non-)Coercive approach ... 38
5.3.2 Open-ended approach ... 38
5.3.3 Top-down approach... 39
5.3.4 Bottom-up approach ... 39
5.3.5 Step-by-step approach ... 40
5.3.6 Communication and Promotion ... 40
5.3.7 Social connections ... 40
5.3.8 Maintaining and spreading Knowledge... 41
5.3.9 Providing support and showcase the Added Value ... 42
5.4 Impact on the Actors ... 42
5.4.1 The Regulator ... 42
5.4.2 Senior staff ... 43
5.4.3 Data and Innovation department ... 44
5.4.4 Financial Audit teams... 44
5.4.5 Data Analytics Champions ... 45
7
5.5 Human Behavior and Motives... 45
5.5.1 Personal Motives ... 45
5.5.2 Negative Experiences ... 46
5.5.3 Human Behavior and Mindset... 47
5.5.4 Knowledge and Skillset ... 47
5.5.5 Workload ... 48 6 Discussion ... 49 6.1 Creating ... 49 6.2 Maintaining ... 50 6.3 Disrupting ... 51 7 Conclusion ... 52 7.1 Conclusions ... 52
7.2 Potential limitations and suggestions for future research... 53
7.2.1 Potential limitations... 53
7.2.2 Suggestions for future research ... 54
Bibliography ... 55
Appendix I: Overview of the interview details ... 58
8 1 Introduction
For the past few years, Data Analytics has become more and more embedded in society. As the use of computed and connected devices is growing by the day, so is the amount of Data. This Data, as is defined as Big Data because these sets of Data could be incredibly large in volume, and too complex to directly recognize patterns without the use of the proper tools. Big Data has a significant amount of influence on the way society is living their life. Like it has on our society, Big Data also has a significant impact on the Audit field (Warren Jr., Moffitt, & Byrnes, 2015).
Numerous industries are already trying to find and develop, more effective and efficient ways to analyze Big Data as every organization is seeking competitive advantage. Within the Accounting field this is no different, as the use of Big Data is promised to have great potential. Supposedly, Big Data ensures; clearer insights, more efficiency and an enhanced coordination for risk and compliance procedures (Cao, Chychyla, & Stewart, 2015). Another alleged implication for Financial Audit is the possibility of continues auditing (Alles, Brennan, Kogan, & Vasarhelyi, 2006).
Despite the fact of all these potential benefits, according to Alles (2015) and Vasarhelyi, Kogan and Tuttle (2015) the Accounting industry remains behind compared to other industries, when it comes to the integration of Data Analytics. Prior research (Vasarhelyi, Kogan, & Tuttle, 2015) suggests that there are numerous factors that could explain the hesitation of the Accounting industry to integrate Data Analytics. According to Zhang et al. (2012) one of main reasons for this hesitation is the absence of standards, set by the audit oversight boards.
However, a different researcher Earley (2015), contradicts that the Accounting industry remains behind if compared to the other industries. His research implies that this argument is caused by the fact that organizations within the Accounting field do not provide any information with regards to their Data Analytics integration process to the outside world. Because they are anxious that they might inform their competitors as well.
So as the current position of Data Analytics in the Accounting field remains quite unknown, I will contribute by conducting a research to gain insights in this process, within a Big Four firm.
9 1.1 Background
It is management’s responsibility to provide financial statements and it is the auditor’s responsibility to give the right audit opinion. With an auditor opinion the auditor argues that he is convinced of the fairness and accuracy of the provided financial statements. When an auditor gives an unqualified opinion, it means that the organization does indeed meet these requirements of them depicting a fair and accurate picture of the (financial) state that the organization is in (Hayes & Wallage, 2014).
An auditor’s opinion always carries the risk of being too subjective, making it more susceptible to bias (Bazerman, Loewenstein, & More, 2012). Previous research conducted by Breeze (2006) suggests that because of the lack of knowledge by auditors in statistical analytics, an auditor’s opinion is less reliable. This is due to the fact that auditors have little understanding of statistical principals and therefore rely too much on subjective risk analyses.
Earley (2015) does mention that in other field’s Data Analytical tools are used to fill up the gaps that the lack of knowledge leaves behind. A research by Dowling and Leech (2014) stated that using Data Analytical tools could indeed fill up these gaps, which would mean that audits would improve in terms of accuracy and assurance. This naturally sparked the interest of accounting firms due to the competitive advantage it could deliver.
Research however shows that most accounting firms only use Data Analytics for consultancy purposes and that almost no Data Analytical tools have been implemented in Financial Audits, which is the core business after all (Earley, 2015).
So even though previous literature show that Data Analytical tools have in been implemented in information systems, there is still a lack of research on the actual use of Data Analytics on the core business of an accounting firm, which is Financial Audits (Dowling & Leech, 2014).
A research by Chan and Vasarhelyi (2011) stated that integrating more Data Analytics into auditing will most likely change the role of an auditor. A side from the reduced labor and time, which improves the efficiency and effectiveness of an overall audit (Elliott, 1998). Automatization also requires change in the sense of new procedures such as standardization of data collection and formalization of internal control policies (Chan & Vasarhelyi, 2011).
10 1.2 Research question
For Data Analytics to be (further) integrated in Financial Audits, it is important to study the role of all actors that are involved in the process that allow for innovations such as; Data Analytical tools to be used in the field. While Data Analytics is already being used within public accounting firms frequently for consultancy purposes. This is not yet the case for Financial Audits, which is the core business.
I will conduct a research on the process of integrating Data Analytics in Financial Audits, within a Big Four firm. This provides me a unique opportunity to actively participate in contributing to the audit field. Since there is a lack of research on the actual use of Data Analytics in Financial Audits (Dowling & Leech, 2014). This case study will attempt to answer the following research question:
‘’To which extent has Data Analytics been integrated in Financial Audits of a Big Four firm?’’ Therefore, I will look into the ways Data Analytics are used in the accounting field now and find out what the views are of important actors within the field. In order to do so, I will need to get a clearer overview of who these actors are, and what their influence might be in the process of integrating innovation into the Audit field.
I will investigate the roll of the auditor in this process and look further into how they have been influencing this process until this point. This way, I will be able to find out the current state of Data Analytics in regards with Financial Audits and why it is in this state. By conducting a research on the actual interaction between important actors in this process I could gain a better view on why Data Analytics has not been (further) integrated into Financial Audits yet.
1.3 Research contribution
The contribution of this research is the ability to gain more insight on why Data Analytics has not been integrated more frequently in Financial Audits. Through gaining actual insight it is possible to look into the interaction between the actors that are involved in this process. This way, I will be able to conclude whether it is due to tendencies of the human behavior, or threshold that come with attempting to integrate innovative technology. In this case integrating innovative technology means an institution hesitating on further integrating Data Analytics in audit field.
11 The research is a good match with the perspectives of Institutional work, which looks into the ‘why’s and how’s’, and less on the ‘what and the when’. This research also contributes to the existing theory regarding Institutional Work, but can also serve as a starting point for (further) integrating Data Analytics into the Audit field.
1.4 Paper structure
The remainder of this thesis is structured as follows. The following chapter contains the existing literature about Data Analytics, this chapter should give a better understanding of how Data Analytics came to be as we know it today, in the accounting field. Thus, reader should be able to have a better understanding of Data Analytics after reading this chapter.
In the third chapter, the theoretical framework of Lawrence and Suddaby (2006) is explained in detail. After reading through all the different stages and forms, the reader should be able to have a better understanding of this theory. Making it easier to understand the approach of this case study.
The fourth chapter contains the methodology used for this research. The first paragraph describes the research setting, and the following three paragraphs give a detailed explanation with regards to the; data collection, data processing and data analysis. After reading this chapter the reader should be able to understand how the research will be conducted.
Chapter fifth chapter contains the performed descriptive analysis, which has been separated into five themes. These are the five primary themes as presented in the last paragraph of the previous chapter. This chapter contains the findings, supplemented with quotes with regards to topic. After reading this chapter the reader should be able have a good picture of the integration process of Data Analytics into Financial Audits.
The sixth chapter contains the discussion, this chapter connects the (main) findings from the previous chapter with the used theoretical framework: Institutional work as discussed in chapter three. After reading this chapter the reader should be able to see the link between the used theoretical framework and the main findings from chapter five.
The seventh and final chapter contains the conclusions drawn from the main findings, followed by the potential limitations of this research and suggestions for future researches. After reading this chapter the reader should be able to have a clear understanding of the ongoing process with regards to the integration of Data Analytics in Financial Audits.
12 2 Literature
This chapter provides an overview of the existing literature on Data Analytics, which explains the definition and processes with regards to Data Analytics. It also contains literature about the impact Data Analytics might has on the audit role and auditor behavior.
2.1 Data Analytics and Big Data
The section will be dedicated to give a better understanding of Data Analytics, which can be distinguished in two components; Big Data and Data Analytics. To actually get a better overview and to gain a better understanding about both definitions, this section will start by explaining how data gets collected to form Big Data. It is worth mentioning that there is a lack of in-depth research on Data Analytics and since technical innovations have been growing so rapidly it is making this subject interesting but complex at the same time.
Data Analytics derives from Big Data, which on his turn derives from the Internet of Things. The Internet of Things can be described as everyday devices that are connected to the internet, which are able to identify themselves to other computer systems and devices, such as smartphones and smart refrigerators, but also public devices, such as check-in portals and vending machines. All the collected identifiable data than forms Big Data, making it possible to be analyzed, turning it into Data Analytics (Syed, Gillela, & Venugopal, 2015).
Most recent example is Facebook, which has been in the crossfire of privacy policy issues (Singer, 2018). The whole affair has been about whether and which Data they are legally allowed to share, and to which extend this is controllable and transparent for users. This kind of data is a prime example of how Big Data is formed and being used for commercial purposes. According to Vasarhelyi, Kogan and Tuttle (2015) Big Data can be defined in several ways, depending on the industry or field. In the case of the accounting field Data Analytics could be used to test full data sets rather than only testing samples.
However, even when specifically focusing on the accounting field Vasharhelyi, Kogan and Tuttle (2015) do mention that even within the field there are some differences in the approach with regards to the definition. A smaller accounting firm naturally has a smaller accounting information system, making Big Data actually much bigger when compared to the size any big four accounting information system.
13 Another approach by McAfee and Brynjolfsson (2012) uses volume, velocity and variety as the three pillars to describe Big Data. Volume, because of the size of each data set and it increasing every second. Velocity, because of the speed that a data set is created, real time or nearly real time. Variety, because of all the different forms it used to gather this information, from social media usage to GPS signals it’s all possible due to technical innovations that took place in the last century.
Runkler (2012) defines Data Analytics as the application of computer systems used to collect and analyze large data sets which helps in the decision-making process. His approach uses a four phased process; preparation, pre-processing, analysis, post-processing. Preparation, which is the first phase mostly consists out of planning and data collection and selection. Pre-processing the second phase, involves filtering, cleaning, correcting, completing and transforming data. The third step, the analysis involves running regression, correlations, forecast and classifying the data. Post-processing, the final phase involves documentation, evaluation and conclusions.
Coa, Chychyla and Steward (2015) describe big as the process of; ‘’inspecting, cleaning transforming and modeling Big Data with the aim to communicate useful information and patterns, suggest conclusions to support decision making’’. In his same article it is also mentioned that Data Analytics used in different fields might indeed be also applicable to Financial Auditing.
The existing literature from different authors tend to describe Data Analytics in comparable manner, however these authors due tend to label these procedures differently and might use more or less phases to do so. Overall, the term Big Data is a definition used to describe the gigantic portfolio of data that is growing rapidly (McAfee & Brynjolfsson, 2012). However, Big Data itself as limited value until it actually gets processed and analyzed. This is where Data Analytics comes in, the process of analyzing data with the objective of drawing conclusions used to strategical decision making (Runkler, 2012).
2.2 Impact Data Analytics on audit behavior
In existing research Vasarhelyi, Kogan and Tuttle (2015) state that both accounting academics and practitioners can benefit from gaining more knowledge about the potential of innovative technologies, such as Data Analytics and the challenges, which come along when trying to integrate (more) Data Analytics into the Audit field.
14 Although Data Analytics is recognized as an innovative technology, which in potential could significantly change accounting as we know it today (Vasarhelyi, Kogan, & Tuttle, 2015). According to Alles (2015) the accounting profession has been showing tendencies of repetitively lagging when it comes to integrating innovative technology into their field. Alles (2015), in his research specifically doubts the (un)willingness of actors within the Accounting institution to integrate technologies such as; Data Analytics.
In previous research by Brown-Liburd, Issa and Lombardi (2015) this is confirmed. However, they do add that the integration of Data Analytics in the Audit field is indeed likely to have some behavioral implications from auditors, which could lead to more tempered benefits. These behavioral implications arise from (a lack off) educational background. Throughout their education auditors are thought traditional audit procedures, which might not be suitable for effectively and efficiently using Data Analytical tools or even analyzing Big Data. Brown-Liburd, Issa and Lombardi (2015) also mention that although Data Analytics has its benefits, it will also stumble upon its own shortcomings much like the traditional audit procedures.
Brown-Liburd, Issa and Lombardi (2015) discusses four issues that might affect the audit behavior, which has direct influence on the audit judgement; ‘’information overload, information relevance, pattern recognition, and ambiguity’’.
Information overload is described as receiving way too much information to process at once. So much information that it might actually lead to a negative effect on the audit judgement as the judgement might have been done ‘sloppy’ (Brown-Liburd, Issa, & Lombardi , 2015).
Information relevance, which also has something to do with the amount of information, is described as an issue in which it is difficult for an auditor to separate the relevant information from the less relevant information. Because too much irrelevant information is likely to negatively affect the auditor judgement quality (Brown-Liburd, Issa, & Lombardi , 2015).
Pattern recognition is described as the capability to search and recognize patterns in an ‘overload’ of information. By recognizing patterns in financial information that may suggests misstatements an auditor can actually assess the risk of a client. The ability to do so using Data Analytical tools can quite easily be educated and trained (Brown-Liburd, Issa, & Lombardi , 2015).
15 The last issue is ambiguity, which tends to come along with Big Data, which is quite unstructured in nature. Big Data often comes in different formats and structures, which makes it more complex to use or even create a tool that is compatible to different formats and structures (Brown-Liburd, Issa, & Lombardi , 2015).
16 3 Theory
This chapter provides an overview of the existing literature on the used theoretical framework, which I will use for the descriptive analysis. It also contains an explanation about all the stages and forms within the theoretical framework: Institutional work.
3.1 Theoretical framework: Institutional work
Cohen et al. (1979) is a research article, which focusses on the changes in organizational structure due to (innovative) technology. They state that organizational structure tends to adapt to changes in technology. Using Data Analytics in Financial Audits can be classified as such a new innovative technology that can disrupt the structure of a public audit firm. The research mentions that the integration of technology may also depend on the organizations management staff, its behavior and actions.
Institutional work is frequently used as framework in the accounting research field to get a better understanding of changes in organizational structure (Currie, Lockett, Finn, Martin, & Waring, 2012).
The way organizational institutionalism is understood varies, while Bjerregaard (2011) reports that in order to get insight on the integration of an innovative technology within an existing institution: It is important to recognize that changes in organizational structure can arise in several ways and start from several levels within an organization; top-down, bottom-up. It is important to mention the interplay between social and technological factors, specifically when researching the process surrounding the integration of technology in an existing institution, and having the behavior of actors in play.
Where previous research focuses more on the process, more recent research by Lawrence and Suddaby (2011) emphasized more on the behavior and actions of actors within an organization when attempting to integrate technology into an existing field like auditing. It is stated that institutions arise from purposeful human actions and reactions with one individual key agent of change (Lawrence & Suddaby, 2006).
This research will give insight in the influence that actors within an institution have when it comes to integrating new technology; Data Analytics into the audit field. Like mentioned in the paragraph above, this research will indeed focus on the social elements by looking into the interactions between actors and agents within the institution to gain knowledge of who’s and whys.
17 Lawrence and Suddaby (2006) describe institutional work as a rough life-cycle. This cycle contains three stages; creating, maintaining, and disrupting. Each of the stages contain several sets of practices, which will be discussed and further explained in the next paragraphs.
3.2 Three stages of Institutional work 3.2.1 Creating institutions
This stage of institutional work can be characterized as the ability for establishing rules and constructing rewards and sanctions that enforce these rules. ‘’Creating institutions highlights the effortful and skillful practices of interested actors in order for them to create institutions’’ (Lawrence & Suddaby, 2006, p. 228). This ability can be gained through political and economic processes in which an actor establishes a superior position in the field. Within this stage Lawrence and Suddaby (2006) define nine forms of institutional work. In the paragraphs below, these nine forms are separated in three distinguishable categories, and explained in detail. Advocacy, defining and vesting
The first three forms of creating institutional work involve; ‘advocacy’, ‘defining’ and ‘vesting’ , which discuss political work that actors have to go through to be able to restructure rules, property rights and boundaries regarding the access to material resources (Lawrence & Suddaby, 2006, p. 221). Combining these three mainly rule based driven forms enables them to strengthen each other, which enables the reconstruction of an institution and its structures and practices. Below is a clear description of the three individual forms.
Advocacy is an important form of creating institutional work. ‘Advocating’ involves the process of gaining regulatory and political support by lobbying for resources, promoting agendas and proposing new legislation, which causes it to be a threat to the existing legislation (Lawrence & Suddaby, 2006, p. 221). When used effectively, advocacy can be a powerful tool that permits actors to influence the way norms are perceived and followed. This form enables actors to change an institutional environment and gain cognitive legitimacy (Lawrence & Suddaby, 2006, p. 222). Lawrence and Suddaby (2006) also mentioned that these different forms of advocacy as mentioned above can allow less powerful actors within the institution to shape the institution environment and acquire cognitive legitimacy.
The second form of creating institutional work is defining. Lawrence and Suddaby (2006) describes defining as; ‘’the construction of rule systems that confer status or identity, define
18 boundaries of membership or create status hierarchies within a field’’. Defining involves the creation of standards and actor certification. From the empirical research examined by Lawrence and Suddaby (2006) it looks like most defining work is more focused on actors and the creation of constructive rules, or rules enabling institutional action. However, compared to the prohibitive nature of regulatory activity, defining emphasizes more on establishing parameters for the future or potential institutional structures and practices (Lawrence & Suddaby, 2006, p. 222).
The third creating institutional work form is vesting. It involves the work, focused on the creation of rule structures regarding property right. Vesting arises when government authority is used to reallocate property rights (Lawrence & Suddaby, 2006, p. 222). This form of institutional work refers to the micro-processes when creating new actors and new field dynamics by changing the game and its rules of the existing market relations. A common component of vesting which is the negotiation of a regulative bargain between the state or another powerful authority and third party interested actor (Lawrence & Suddaby, 2006, p. 223).
Constructing identities, changing normative associations and constructing normative networks The second category of creating institutional work refers to the three forms of interaction; ‘constructing identities’, ‘changing normative associations’, and ‘constructing normative networks’. By slowly persuading actors to think differently, slowly influencing them to alter their ways of thinking. Lawrence and Suddaby (2006) refer to this category as the three forms of interaction that provide the basis of forming new institutions. All three forms of interactions focus on the roles, values and norms that are associated with an institution. The institutionalized rules created often complement or are parallel to the established institution. For normative work to succeed it is important to mention that diverging too much due to a conflict within the collective could reduce its force, which gives other actors chance to take over. Below is a clear description of the three individual forms.
Lawrence and Suddaby (2006) refer to ‘constructing identities’ as the form that lays central to the creation of institutions, because it describes the interplay between an actor and its field. In institutional theory, the construction of identities has been commonly associated with the development of professions. Which underlines the importance of collective action in accomplishing the construction of identities as a form of normative institutional work (Lawrence & Suddaby, 2006, p. 224).
19 The second form of this category is ‘changing normative associations’. It is a different form of interaction but also with the aim on creating new institutions. The work involves reformulating normative associations, which means; re-making the connections between sets of practices and its moral and cultural foundations (Lawrence & Suddaby, 2006, p. 224). Time has learned that these practices have frequently led to new institutions, complementary or similar to existing institutions. This form of interaction however, did not involve a direct threat or challenge to pre-existing institutions, but rather supported them. At the same time this lead to actors to questioning them. Changing normative associations refers to manipulating the relationship between norms and the institution that created them (Lawrence & Suddaby, 2006). The third form of creating institutional work in this category aims to create new institutions and is referred to as; ‘constructing normative networks’ by Lawrence and Suddaby (2006). Normative networks are interorganizational connections through which practices become an institution and institutional work. These normative networks possess the power to provide normative compliance, monitor and evaluate of these practices. This form brings together actors from different fields, to form a coalition that is able to alter the relationship between actors in a field by changing the normative assumptions.
Mimicry, theorizing and educating
The final category focusing on the creation of institutions involves; ‘mimicry’, ‘theorizing’ and ‘educating’. Lawrence and Suddaby (2006) describe the actions within this category as; ‘’actions designed to alter abstract categorizations in which the boundaries of meaning systems are altered’’ (Lawrence & Suddaby, 2006, p. 221). These three forms of institutional work emphasize on the cognitive elements of an institution. Cognitive elements contain; ‘’the beliefs, assumptions and frames that inform action by providing meaningful and understandable interaction patterns to be followed’’ (Lawrence & Suddaby, 2006, p. 228).
Mimicry is described by Lawrence and Suddaby (2006) as work that draws on existing institutional work, in order to articulate and legitimate new practices and structures. This means that existing institutional patterns can be used in the process of creating a new institution by leveraging existing sets of practices within these institutions. Associating new elements with existing elements in some way might ease the integration because it decreases the contrast between new and old. Part of the success of mimicry in creating new institutional structures is that it associates new elements with older once to make these new structures clearer and more accessible, while highlighting the insufficiencies and shortcomings of past structures.
20 Theorizing is ‘’the development and specification of abstract categories, and the elaboration of chains of cause and effect’’ (Lawrence & Suddaby, 2006, p. 226). In other words; the development of concepts and beliefs to support new institutions (Lawrence & Suddaby, 2006, p. 228). A key aspect in theorizing is highlighting new concepts and practices, giving them the chance to become part of the cognitive map of the field. Lawrence and Suddaby (2006, p. 226) refer to this as ‘naming’, which represents the first step that provides the foundation a chance for further theorizing. This however does not mean that naming could be referred to as agreeing to these concepts and practices.
Educating is the last form of institutional work in this category, this form aims at educating actors with the skills and knowledge, which enables them to participate in new practices or interact with new structures in support of the new institution (Lawrence & Suddaby, 2006, p. 227). This is important because new institutions are likely to introduce information which was unknown before, therefore it is needed to educate actors with additional skills and knowledge they need (Lawrence & Suddaby, 2006, p. 228).
3.2.2 Maintaining institutions
This stage of institutional work can be characterized as the actions needed to maintain institutions. As most institutions naturally do not have the ability to automatically reproduce itself over time, the emphasis of this stage will be focusing on the work that is required to continue an institution, therefore it is likely that a lot of maintaining work is needed. Lawrence and Suddaby (2006, p. 230) mention that the work of maintaining institutional work in generally involves; supporting, repairing or recreating social mechanisms that ensure compliance. Within this stage Lawrence and Suddaby (2006) identify six forms of institutional work which focus on maintaining an institution. In the paragraphs below, these six forms are separated in two distinguishable categories, and explained in detail.
Enabling work, policing and deterring
The first three forms of maintaining institutional work involve; ‘enabling work’, ‘defining’ and ‘deterring’, these three forms mainly focus on the maintenance of institutions by ensuring adherence to rule systems. These three forms together form a force that is an important foundation and key pillars for an institution. Without these three forms the foundation of an institution would be likely to crumble, leaving only empty threats and promises rather than
21 self-activating means of institutional control. Below is a clear description of the three individual forms.
‘Enabling work’ is described by Lawrence and Suddaby (2006) as the creation of rules to enable the survival and continuation of institutions and their respected activities. This means providing them with facilities, supplements and support, even if this also involves the creation of authorizing agents, new roles or diverting resources (Lawrence & Suddaby, 2006, p. 230).
Lawrence and Suddaby (2006) describe maintaining an institution through enforcement, auditing and monitoring as policing. Policing of is used to ensure compliance, the activities of policing can involve the use of sanctions and inducements, frequently used together at the same time by the same agent (Lawrence & Suddaby, 2006). However, Lawrence and Suddaby (2006, p. 231) do mention that the use of enforcement to ensure compliance is not always a necessary strategy in professionalized fields, because in these cases usually auditing and monitoring is enough to ensure compliance.
The third and final form of maintaining institutional work with the aim to ensure compliance is called ‘deterrence’. Deterrence focusses on ‘’establishing coercive barriers to institutional change which involve the threat of coercion to inculcate the conscious obedience of institutional actors’’ (Lawrence & Suddaby, 2006, p. 232). Lawrence and Suddaby (2006, p. 232) also mention that effective deterrence is very dependent on legitimate authority of a coercive agent and that it may originate from the threat of economic coercion.
Valourizing and demonizing, mythologizing and embedding and routinizing
The final three ‘valourizing/demonizing’, ‘mythologizing’ and ‘embedding & routinizing’ are more focused on maintaining an institution by reproducing existing norms and belief systems. It is worth mentioning that coercive work, which is rule-based used to maintain institutions are more visible than their cognitive or normative counterparts. Actors involved in such work, as well as the complying actors are aware of the effects and purpose of these actions (Lawrence & Suddaby, 2006, p. 232).
Valourizing and demonizing involve work wherein actors identify and review the morality of participants within the field to enact the institutionalized beliefs, but also as a way of maintaining the power of those beliefs (Lawrence & Suddaby, 2006, p. 232). The work of maintenance provided by the form of valourizing and demonizing contains giving positive and negative examples to demonstrate the normative foundations of an institution. As results of
22 these positive and/or negative examples the state of the institution can get reinforced or deteriorated (Lawrence & Suddaby, 2006, p. 232).
Mythologizing represents a form of maintaining an institution that focusses on the past, rather than the present. An important way in which actors aim to preserve the normative pillars of institutions is by mythologizing their history. This means providing work ‘’to create and sustain a myth using a story and occasion to tell it’’ (Lawrence & Suddaby, 2006, p. 233).
Embedding and routinizing the last form within this category of maintaining institutional, Lawrence and Suddaby (2006, p. 233) characterize this form as reproducing through the stabilizing influence of embedded actions and routine practices. Containing actions and practices such as; training, education, hiring and certificated routines and ceremonies of celebration. This form of embedding and routinizing involves integrating these embedded and routine practices into the participant’s daily routines and organizations practices (Lawrence & Suddaby, 2006, p. 233).
3.2.3 Disrupting institutions
The last stage of institutional work are the practices aimed at disrupting an institution, this stage involves an offensive or undermining attitude towards the mechanisms in place at existing institutions. Until today not much is known about disruptive work, and most of what is known does not come directly from consciously undermining an existing institution. Also, the use of disruptive work usually takes an accomplished, experienced actor with a lot of knowledge about the profession, for it to have the right effect. This stage of institutional work focusses mainly on the relationship between an institution and the social controls that give them the ability to continue their practices. This usually involves ‘’disconnecting rewards and sanctions, disassociating moral foundations and undermining assumptions and beliefs all disrupt institutions by lowering in some way the impact of those social controls on non-compliance’’ (Lawrence & Suddaby, 2006, p. 238). Lawrence and Suddaby (2006) identify three forms of institutional work, which focus on disrupting an institution; ‘Disconnecting sanctions’, ‘disassociating moral foundations’ and ‘undermining assumptions & beliefs’. In the paragraphs below, these three forms will be explained in detail.
23 Disconnectingsanctions,disassociatingmoralfoundationsandunderminingassumptions&beliefs Disconnecting sanctions/rewards is described by Lawrence and Suddaby (2006, p. 235) as a form of work within the institutional work aimed at disrupting an institution by disconnecting sanctions and/or rewards. This form involves undermining attitude towards the mechanisms in the existing (leading) institution. A form of undermining is working state and non-state actors through state-systems to disconnect sanctions and rewards from set certain set of practices, technologies or rules (Lawrence & Suddaby, 2006, p. 235). This is mainly done using jurisdiction that has enough impact to directly invalidate the former powerful institutions (Lawrence & Suddaby, 2006, p. 238).
Disassociating moral foundations is described as the potential to ‘’disassociating the practice, rule or technology from its moral foundation as appropriate within a specific cultural context’’ (Lawrence & Suddaby, 2006, p. 236). So rather than directly attacking the foundations, the institution is getting disrupted by indirect sets of practices and actions which undermine this same foundation. Lawrence and Suddaby (2006, p. 237) also found that actors who use these dissociative practices and actions to disrupt an institution are likely to be the largest firms and professional associations, the so called ‘elites’ within the profession have quite a high status within the field due to their level of prestige. While these elites first adopted new practices on the basis of technical reason, studies have showed that actors who were used to using their status to develop and spread technical changes that justified being different. Which put late adopter under pressure to follow and reform (Lawrence & Suddaby, 2006, p. 237).
The last form of disruptive institutional work is undermining assumptions and beliefs. Like the name implies this form of disrupting involves undermining the supposed effects with regards to the changes that established institutions have brought. But it can also involve an innovation that overrides rules or assumptions made by established institutions, or just by slowly proving a concept by an established institution wrong. Although there is not much documentation about these forms of disruptive work, Lawrence and Suddaby (2006, p. 237) still were able to distinguish two different types within this disruptive work. The first one is being breaking the existing institutional assumptions by innovation. And the second one being gradual undermining of existing institutional assumptions through contrary practice.
24 4 Methodology
This section will discuss the research methodology in detail and why the decision was made to carry out a qualitative research. The objective of qualitative research is to get a better insight on how a particular process actually went. It usually describes the perspective of the participants’ experience throughout the process, typically using interviews, observations and documents. (Maxwell & Reybold, 2015). ‘’Qualitative research is multimethod research that uses an interpretive, naturalistic approach to its subject matter’’ (Rynes & Gephart Jr, 2004). 4.1 Research setting
As mentioned in the introduction, the aim of this thesis was to provide more insight on why Data Analytics has not been integrated more frequently in Financial Audits. Institutional Work (Lawrence & Suddaby, 2006) will be used as theoretical framework because this theory is a good match for; gaining more insights as to ‘why and how’ Data Analytics has not been (further) integrated more frequently in Financial Audits.
Although, previous literature shows that Data Analytical tools have indeed been implemented in information systems, there is still a lack of in-depth research on why there perhaps is, still a delay on the actual use of Data Analytical tools in the core business of an accountancy firm; Financial Auditing (Dowling & Leech, 2014).
Through means of a thesis internship at the firm a possibility was opened to indeed gain more insights on the process of these developments.The reason for conducting a case study at the firm is that they are on top of the accounting market, and thus are likely to be the best representative of the changes in the accounting field. More specifically the firm has one of the biggest and advanced Data Analytics and Innovations departments in the industry.
4.2 Data collection
Before starting this research, it was important to separate the key actors from the not actors. In order to do so, it was needed to limit the population to those that have a significant impact on the process of integrating Data Analytics in Financial Audits, or those who will be influenced by the integration of Data Analytics in Financial Audits. In order to know who to filter out, the first step was to get a better understanding of the internal structure.
25 After gaining this understanding the second step was selecting the relevant actors to be interviewed. In order to gain enough insights and a variety of perspectives, the main criteria was that these actors work on different positions and levels within the organization.
Finding interviewees who fit these criteria was not easy, despite the fact that all interviewees were quite easy to approach. In the end I was able to interview ten people (see appendix1); two Senior Audit staff and three Junior Audit staff members from the Financial Audit Department(s), three IT-specialist/developers, one Data Analytics champion and one Senior staff member from [Department X]. Which is the Innovation and Data Analytics department, henceforth named; the Data and Innovation department.
I separated the Senior Audit staff members (A1, A2) and Junior Audit staff members (B1, B2, B3) because this is part of the comparative study, the Audit staff members have been interviewed because they are heavily influenced by the process of integrating Data Analytics into Financial Audits. The Senior staff members because they are in a more powerful situation, where they have a direct influence on the audit approach, and the Junior staff because they are often young and lack the knowledge and experience.
Within this Big Four firm there are three departments that have an above average knowledge of Data Analytics. Two of them are actual functioning departments, while one acts as an assisting/service department for the Dutch Office(s). For the research, I chose to interview the IT-specialist/developers (C1, C2, D2), the Assistant Manager (D1) and a Data Analytics champion (D2) within this department, because the Data and Innovation department focusses mainly on innovating Financial Audits. The choice to interview people in this department was made because within the firm. The Data and Innovation department are the Data Analytics specialist who are responsible for developing, introducing and promoting new innovative Data Analytics tools. Which makes them key actors that need to be interviewed.
The two Senior staff members from the Financial Audit department(s) are experienced auditors and have a managing position within the Audit engagement teams. Interviewee A1 is a manager, whom has been with the firm over the past 9 years. He is an experienced Registered Accountant and is momentarily finishing up his second post-master program to be a Registered EDP-accountant. Furthermore, he is actively involved in the development of new Data Analytics tools. This means that he has knowledge and experience in both side of the coin; Auditing and IT/Data Analytics. Interviewee A2 is also a Manager, she is an experienced Registered Accountant whom has been with the firm for nearly 12 years and has applied Data
26 Analytics in some of her Audit engagements. Both Senior staff members were trained following the traditional audit approach.
The three Junior staff members from the Financial Audit department(s) are all audit trainees and have less than two years of working experience. Interviewee B1 is a young audit trainee whom has been with the firm for only a year and has experienced working on several departments as part of his traineeship. Interviewee B2 has a quite similar resume, but has a background in economics. Just like interviewee B1, he is also in a traineeship where he gets to gain experience in several departments. Interviewee B3 also has been only with the firm for a year. Different from the previous two Juniors, he is following a traditional traineeship where he focusses mainly on Financial Auditing. All Junior staff members do not have years of experience and do not have an above average knowledge about Data Analytics/IT.
All three IT-specialist/Developers from the Data and Innovation department have a different background. They are experienced IT specialists and Data Analytics tools developers, but lack Financial Auditing experience (in the field). Interviewee C1 is a developer with a background in accounting, he has been with this firm for nearly 2 years and before joining the firm was already developing for different organizations, he has experience as an accountant but no (field) experience in auditing. He is currently finishing up his ACCA in accounting. Interviewee B2 has been with the firm for 2 years and has no accounting background, nor does he have any accounting (field) experience. He’s a Data Analytic specialist due to his Statistical/Mathematical skills and learned to program within the firm. Interviewee C3 has been the firm for almost a year, and does not have a background in accounting either. All developers/IT specialist are fulltime involved in the development of Data Analytics tools.
Interviewee D1 is a Senior staff member within the Data and Innovation department. He has been with the firm for over 12 years and is a Registered Accountant. He is responsible for [Tool name] and is involved in the development of Data Analytics within the firm. Interviewee D2 is a Data Analytics champion, he has only been with the firm for a year and has a Data Science background. He is currently pursuing to be a Registered Accountant and is actively involved in the development of Data Analytics. He is also a traineeship where he got the chance to gain experience in several departments. Different from the main developer is that he is working in on Audit engagement. Both are advocates for the use of Data Analytics in Financial Audits.
27 Most interviewees were introduced to me through mutual connections, after a formal introduction in person or via the firms’ internal chat a more extensive introduction got sent through the mail. After that a formal invitation through the mail was sent with date, time and location. Some interviewees wished to not be voice recorded, so for these interviews I was prepared to take extensive notes.
The interviews were conducted in Dutch as much as possible as this was the language in which interviewees felt comfortable to fully express themselves. If the first language of the interviewee was not Dutch, the interview was held in English. All the interviews were transcribed in the languages it was conducted in. As part of the interviews were used for quotations it was needed to carefully translate the Dutch interviews where needed, these sections were directly translated and put into the Data Analysis chapter.
As each of the interviewees had already agreed (or not) to the interview being voice recorded this did not have to be mentioned again during the start of the interviews. This was done to avoid awkward moments in the start of an interview, as interviewees who are not comfortable might not ‘’tell everything’’.
4.3 Data processing
After transcribing all the interviews, the relevant quotes were carefully translated from Dutch to English without losing the true meaning and underlying tone. After this process codes were given to the quotes and while pre-analyzing the Data is became clear that some codes were overlapping each other. It was decided to take some codes out or to merge some codes to avoid repetitive work.
4.4 Data analysis
The research was separated into five primary themes. The first primary theme is ‘’Definition of Data Analytics’’. From the literature it was proven that it was difficult to give an ultimate definition of Data Analytics due to its complexity and because it is a very broad term. Therefore, the theme was divided into two sub-codes: The definition of Data Analytics in General (D/DG) and the definition of Data Analytics with regards to Accounting (D/DA).
Primary theme: Definition of Data Analytics
D/DG The definition, by the interviewee for the term Data Analytics in General.
D/DA The definition, by the interviewee for the term Data Analytics with regards to Accounting.
28 The second primary theme is ‘’Change and Impact of Data Analytics’’, this theme contains topics that deal with the changes and the impact these changes have. The codes have been based upon the all the changes and impact that are directly relatable to the integration of Data Analytics. Furthermore, it was also taking into consideration that some changes and impact have occurred in the past are still going on now or will be relevant for the future.
Primary theme: Changes and Impact of Data Analytics
C/AA Changes and Impact with regards to the Audit Approach.
C/PT Changes and Impact with regards to the Planning and Timeline of Using Data Analytics.
C/PD Changes and Impact with regards to the Potential of Data Analytics.
C/DC Changes and Impact with regards to the Data Collection and Availability of Data.
C/TI Changes and Impact with regards to the Tool and Innovation.
C/ED Changes and Impact with regards to the Expectation of Data Analytics.
C/K Changes and Impact with regards to the Knowledge.
C/E Changes and Impact with regards to the Education.
C/S Changes and Impact with regards to the Structure within the firm.
C/PS Changes and Impact with regards to the Procedures and Standards.
C/G Changes and Impact with regards to the Guidance during the Application process.
Table II: Overview coding scheme ‘’Changes and Impact of Data Analytics’’.
The third primary theme is ‘’Application Strategies and Actions taken’’. This theme deals with the way the application of Data Analytics in Financial Audits is being performed, what actions are taken, but also what application strategies are used.
Primary theme: Application Strategies and Actions taken
S/C Application strategies, in this topic: (Non-)Coercive approach.
S/OE Application strategies, in this topic: Open-ended approach.
S/TD Application strategies, in this topic: Top-down approach.
S/BU Application strategies, in this topic: Bottom-up approach.
S/SS Application strategies, in this topic: Step-by-step approach.
S/CP Actions taken, in this topic: Communication and Promotion.
S/SC Actions taken, in this topic: Social connections.
S/MS Actions taken, in this topic: Maintaining and spreading Knowledge.
S/PA Actions taken, in this topic: Providing support and showcase the Added Value.
29 The fourth primary them is ‘’Impact on the Actors’’. As this a research is about the impact of integrating Data Analytics in Financial Audits. Therefore, the theoretical framework of Institutional work has been used, naturally this involves the ‘’Impact on the Actors’’. This theme addresses the differences between the roles played by the actors that are involved.
Primary theme: Impact on the Actors
I/R Impact on the Actor, in this topic: The Regulator.
I/S Impact on the Actor, in this topic: Senior staff.
I/DD Impact on the Actor, in this topic: The Data and Innovation department.
I/FA Impact on the Actor, in this topic: Financial Audit teams.
I/DC Impact on the Actor, in this topic: Data Analytics Champions.
I/C Impact on the Actor, in this topic: Clients.
Table IV: Overview coding scheme ‘’Impact on the Actors’’.
The fifth and last theme is ‘’Human behavior and Motives’’. This theme contains the perceived reason and given reason by interviewees with regards to the integration of Data Analytics in Financial Audits. The provided quotes can be associated with why certain actions have been taken.
Primary theme: Human Behavior and Motives
B/PM Human behavior and Motives, in this topic: Personal Motives.
B/NE Human behavior and Motives, in this topic: Negative Experiences.
B/HM Human behavior and Motives, in this topic: Human Behavior and Mindset.
B/KS Human behavior and Motives, in this topic: Knowledge and Skillset.
B/W Human behavior and Motives, in this topic: Workload.
30 5 Descriptive Analysis
This chapter contains the descriptive analysis of the gathered data extracted from the performed interviews. The key topics, which were discussed in the previous section will be the guide throughout the analysis below. The analysis will contain a description of the interview findings and where suitable supplemented with notable quotes from the interview.
5.1 Data Analytics
5.1.1 Definition of Data Analytics
As stated within the literature section, Data Analytics can be defined in a variety way’s. Therefore, it is hard to give a clear definition in context of the accounting profession. Throughout the interviews it indeed became clear that it is hard to define Data Analytics, as it can be put in so many forms. The collected data on this topic confirmed this, as all interviewees gave their own definition of Data Analytics, often different from one another. Respondent A1 gave the following definition of Data Analytics:
‘’Data used to gain insights and patterns, that serve the purpose of strengthening the decision making.’’ (M1, Q1) Respondent B1 and B2 mention that they indeed find it hard to give a clear definition of Data Analytics as they state:
‘’It is hard to find a good consensus between what Data Analytics really is and what people within the audit make it to be. Especially in audit where Data Analytics can mean so many different things, from a simple spreadsheet to a tool’’ (T1, Q1 & T2, Q1) Like mentioned in the first paragraph it is also different because depending on the context Data Analytics is in, it can indeed mean a lot of different things. Respondents who were more involved in the IT-Audit part of Data Analytics gave more general answers as their definition of Data Analytics. Respondent C1, but also C3 who shared the same thought about Data Analytics, gave the following description of Data Analytics. Both actually came really close towards the literature described Data Analytics:
‘’Getting insides from big amounts of data, through different kinds of Data Analytics procedures. Ranging from data extracting, data cleaning, data transformation, data analyses and visualization.’’ (C1, Q1)
31 5.2 Changes and Impact of Data Analytics
5.2.1 Audit approach
In several interview it was stated that Data Analytics has (had) a great impact on the Financial Audit approach. Often mentioned was that in traditional audit procedures, sample-testing was used. This meant that if a certain number of samples were taken and if these results of this sample were correct, that this control could be checked-off. The audit approach were sampling got traded it for Data Analytics changed this because Data Analytics provided a different view on these audit procedures. Respondent D1 describes this as the most commonly known change/impact that Data Analytics has had on the audit approach. D1 stated:
‘’The most commonly known example of Data Analytics is that you can test a whole population, instead of just testing a sample. Which enhances the audit quality.’’ (D1, Q6) However, one of the respondents; D2 who has knowledge and experience in Auditing and IT mentions that there is one problem with this approach, he described it with an example:
‘’I once had to check 1,2 million transaction, which was the whole population. When using the traditional approach, I had to take samples of 40 transactions. When you take these 40 samples it will follow a path which we call ‘’Happy Flow’’. Which means that if you have a process that has 90% following this ‘’Happy Flow’’ pattern, that it is likely to be correct. However, with a sample of 40 this is more likely to go well than if you would use a whole population. Because if you take the whole population it is quite likely that 10% will not follow this pattern.’’ (D2, Q2) D2 also mentions that regulation nowadays does not allow for this 10% to be an outlier, and that it has to be 100% correct, which means that even for this 10% of samples that do not follow the pattern, you will need to be able to give a proper explanation. Thus, making this one of today’s Data Analytical flaws. Because of flaws like this the IAASB has since then established more strict rule, adding more controls to be checked off to serve the goal of better reports which should result in a more defined approach in auditing.
However, these restrictions have also caused people to be skeptical or even unwillingly to use Data Analytics, as they are not able to fit in into their audit approach and are not willing to put in audit planning. Respondent A2, added why he considered the new audit approach to be conflicting with a broader integration of Data Analytics in Financial Audits:
32 ‘’An auditor is no longer doing pure Audit tasks. Part of the job nowadays is also check listing the controls in a Data Analytics tool. And while testing a bigger simple might be better for the quality, it was not necessary since the traditional sampling was already sufficient, so in a way we’re are doing more work than needed.’’ (A2, Q4) So with Data Analytics, there also was a shift in kinds of audit tasks. Young auditors no longer have to check off thousands of hours in year reports, which make their work more dynamic and perhaps less boring. Which also enhances audit quality. But the traditional way was already sufficient. However respondent D2 disagrees as he stated:
’’I think the role of an auditor stays quite the same, only thing that changes is the skill set that is needed to fulfill these tasks.’’ (D2, Q14) So as the years progress, the new audit approach is getting more materialized. However, there is some resistance that can be sensed in the interviews, as the audit professionals and IT-professional do not fully agree.
5.2.2 Planning and Timeline of using Data Analytics
What we know now is that integrating Data Analytics in Financial Audits changes the audit approach, but it also changes the timeline of an audit. One of the pillars in an audit timeline is planning. Some respondents mentioned that in most occasions Data Analytics is only used at the beginning phase of the audits and the control phase of an audit, which is actually the end phase. This was often the case when a control could not be performed using traditional methods. Respondent D2 stated that in most cases Data Analytics will only be used in an audit if it has been scheduled in the audit-plan, he stated:
‘’Often there is already an audit-plan and whenever something comes by where Data Analytics may be of use. They often say that there is no time for that, because it has not been scheduled in the audit plan.’’ (D2, Q5) However, D2 does mention that nowadays, he is involved in the audit-planning as well and that they do listen to his suggestions.
5.2.3 The Potential of Data Analytics
From the interviews it seems that all of interviewees have the feeling that the organization is really putting in effort with regards to Data Analytics. As there is no way around Data Analytics, as the competition will strive past our firm if it is not willing to put money and effort
33 in Data Analytics. With a dedicated department that fully focusses on Data Analytics innovation all of the interviewees have the feeling that they are on the right track. The respondents that are actively involved in this process take pride in to doing so. However, respondent A1 stated that although Data Analytics has great potential, he also mentions the following:
‘’A full integration or even automatization with the use of Data Analytics will take a long time, but it might come faster than we expect.’’ (A1, Q10) 5.2.4 Data Collection and Availability of Data
Data analytics can only be used if the (Big) Data needed at the audit is correct and available to use. Respondent A1 mentions that:
‘’The most important thing, especially in Audit is the Data Quality. That is the most important, but at the same time also the hardest part.’’(A1, Q4) The interviews asked about Data Collection have not mentioned any problems regards of Data availability. In the case of this Big Four firm in which this research was executed the Data Extraction, Data Cleaning is done by the department dedicated to Data Analytics and innovation. When this is not the case the Data is being obtained from the client, who in most cases already has a digital documentation.
This might be harder for the smaller client who have not gone digital yet, but in the client portfolio of this Big Four, that is hardly the case. However, if this is the case there are tools that are able to digitalize these documents by extracting the relevant information and putting them in spreadsheets.
Another critical part of Data begins when it has to be prepared to fit the Data Analytical tools used. All IT-Experts interviewed acknowledged that they lack this skill. Knowing that Financial Auditor in general lack the knowledge that is needed to process the Data in a manner that can be used for Data Analytics. Respondent D2 mentions that:
‘’Auditors do a lot of work in Excel. They are true magicians when it comes to Excel.’’ (D2, Q9) This confirms, that within the firm there are a lot of floating Excel files, which makes it harder to transform all these floating Excel files into useable Data for Data Analytics tools.
