The rise of Big Data in Austrian tax consultancies

(1)

Master Thesis

The rise of Big Data in

Austrian tax consultancies

How stakeholders of Austrian tax consultancies assess the potential influence of Big Data

Author: Marc Buchner Supervisor: Krenare Nuci

Examiner: Associate Professor Päivi Jokela Date: 2020-05-28

(2)

Abstract

The fact is that every individual leaves behind vast amounts of data, companies collect this data and use the knowledge gained from it in a variety of ways. One area that is lucrative for the use of Big data is the financial sector. A prominent example of the use of Big data is real-time stock market insights. However, there are still industries in which Big data is not yet used for various reasons. One of these industries is the tax consulting sector, which will be the focus of this research. With its high entry hurdles, direct dependence on the legislator, and the associated atypical data sets, the tax consulting sector represents a special use case within the financial sector.

Because big data has not been used in the tax consulting sector yet and that the setting is atypical compared to other sectors, a closer analysis of potential influences on services, the working environment, and quality is of particular interest here. This analysis is the core of this study and was carried out using an interpretative qualitative approach in the form of a case study. In this case study, the three most important stakeholders of Austrian tax consultancies- employers, employees, and clients - were interviewed on the one hand through interviews and on the other hand through a survey with open-ended questions. The results were then compared in the discussion with the changes that studies in other fields have identified.

The results of the study showed that the stakeholders predominantly assume that the quality of services will improve significantly through the use of big data, especially in accounting and business management services. Stakeholders also predicted a positive development concerning the range of services offered. It was also predicted that the range of services offered could increase on the one hand and that services of a business management nature could benefit enormously on the other. In the area of the working environment, employees said that increased training activity and process adaptation would be the only significant changes. In the area of risks, all three stakeholder groups agreed and mentioned data protection. Interesting differences between the three stakeholder groups were on the one hand that the employers gave very detailed answers, which allows the assumption that they have already thought carefully about the topic of big data. On the other hand, in contrast to the other two groups, the employees did not primarily think of their area (work environment) in the analysis, but of that of the clients and thus of the provision of the service. This underlines the strong focus on client satisfaction and encourages a more intensive involvement in the design process.

In contrast to other studies, this thesis analyses the influences on the areas not from a retrospective point of view, but a prospective point of view. This approach allows an unbiased look at the opinions of stakeholders and thus provides the best possible information for the design of big data tools for the tax consulting sector. Besides, by comparing this with changes found in other studies, it is possible to estimate how the use of big data in the tax consulting sector differs from other sectors.

Keywords: Big Data, tax, Austrian tax consultancies, consulting, information and communication technology, Big Data tools, Information Systems

(3)

List of tables

Table 1: Participants, collection methods and codes 21

Table 2: Answers to research question 1: Quality of Service 31

Table 3: Answers to research question 2: Service Range 32

Table 4: Answers to research question 3: Work Environment 32

Table 5: Answers to research question 4: Main Concerns 33

Table 6: Answers to research question 6: Main Differences and Similarities 34

List of figures

Figure 1: Thesis Organization ... 3 Figure 2: 4Vs - Volume, Variety, Velocity and Veracity (adapted from Abbasi, et al., 2016) . 5 Figure 3: 4Vs that define big data (adapted from Abbasi, et al., 2016) ... 6 Figure 4: Change in the treatment of data (adapted from Hilbert, 2016) ... 7 Figure 5: Austrian Tax Consulting study dimensions: ... 18

(4)

1. Introduction

Chapter 1 introduces the reader to the topic of the research study and provides a brief overview of big data and the Austrian tax consulting sector. Subsequently, the research questions central to the purpose of the study are named, which are rounded off by the scope and limitations of the thesis.

It has been said that everything that can be digitized, will be digitized. As an accessory of this development, increasing amounts of data are generated, both personal and non-personal data (Egner, 2019).

In fact, the data and their use are increasingly becoming the focus of attention in science and industry. The umbrella-term for this phenomenon is big data, which is understood to be large collections of data and information from various sources. The process of deriving the data from various sources lets us understand that the derived data are of various format, thus they are not only structured data, which has a normalized form but also unstructured data, which are normally not stored in databases. Examples of structured data would be names, addresses, or even bank details that are stored in a structured manner, and that could be manipulated from the standard tools and databases. Whereas, those data that are stored unstructured but also have different formats of representations such as videos, photos, or audio recordings and could not be manipulated by standard databases are encounter as unstructured data. This data is collected daily in large quantities (Schroeck, et al., 2012).

A glance at the media landscape reveals that big data has already massively changed sectors such as logistics, production, or healthcare. These changes essentially based on the information advantage that can be achieved utilizing sophisticated analysis methods. Big data has therefore fundamentally changed the way we look at data and the way it is obtained; when talking in respect of the value of data, some even speak of data as the new gold (McAfee, et al., 2012).

This saying is certainly true in the financial sector, but in the investment sector, for example, a head start in the stock market can be a massive advantage.

However, surprisingly there are sectors in which the new technology does not play a role yet.

One of those branches is the Austrian tax branch, which differs from other branches. The main reason why it differs from other branches is, that the tax branch is protected by high entry requirements, which are six exams where the pass rate is about forty percent (Bär, et al., 2016).

Another reason is that the setting in the industry is extremely complex due to tax regulations.

In the past, only a few service providers were able to provide good software for tax consultants, which is why most tax consultants shared the same technological basis. Subsequently, the pressure to change was historically very low. The missing pressure manifests itself in outdated organization structures and insufficient adaptability. Taking a closer look at a typical tax consultancy in Austria reveals that except for the use of computers, nothing has changed since the early eighties.

But recently, new technologies started to put pressure on the tax consulting branch. On the one hand, this pressure stems from the fact that the services of tax consultants are very time- consuming. The effective use of new technologies can save time and thus the service can be sold more cheaply, which puts pressure on competitors. On the other hand, many services are based on experience and sector-specific know-how, which can be partly compensated by better software. Therefore, in some cases, lower qualified workers can use the software to provide a service of approximately the same quality as the specialists without good software. A study recently published by Mindtake in cooperation with Datev, LexisNexis, and Future-Law, which surveyed tax consultants and employees in tax consulting firms on digitization issues, showed that 83 percent of all firms are planning concrete digitization steps within the next two years (Mindtage, LexisNexis, Datev, Future-Law, 2019). The planned digitization steps include the implementation of new technologies, such as artificial intelligence tools or big data tools. It can, therefore, be said that the growing pressure is setting the sector in motion. The potential

(6)

changes resulting from the digitization steps are still very unclear at this point. This fact is also reflected in the study, where the effects of digitization range from changes in the way employees work to changes in business models (Mindtage, LexisNexis, Datev, Future-Law, 2019).

However, these subjective views vary between the different stakeholders.

A closer look at the tax consulting branch in Austria, therefore, reveals that stakeholders have an idea of potential changes caused by new technologies (Bär, et al., 2016). It can be assumed that these ideas are not only influenced by the previous environment of the industry, but also by the significant changes that new technologies have brought about in other branches.

Since stakeholders have a significant share in the successful implementation of new technologies, it is important to research the opinions of stakeholders and take them into account during implementation (Davenport, 2014).

Concluding, the rise of big data tools in Austrian tax consultancies is still in a very early stage, the employees, employers, and clients seem to have a certain opinion about the use of big data in tax consulting. To analyze the different opinions, this thesis seeks to answer questions, that focus on the opinion of the three stakeholders (employees, employers, and clients) on certain topics. According to Shah, et al. (2017) this is vital for successful implementation and designing chance projects because stakeholders are able to visualize the change by comparing past experiences with future perspectives.

1.1 Purpose Statement

The purpose of this thesis is to originate the understanding of different opinions on potential influences of big data tools on Austrian tax consultancies. This goal is achieved through interviews with the three most important stakeholders, which are employers, employees and clients, and an open-ended survey. The focus of the interviews and the survey is to identify subjective opinions on the potential impact of big data on specific areas. Those areas are the service range, work environment, main concerns, and the specific area of each stakeholder (service for clients, employment for employees, and business model for employers). Since the opinions of the three most important stakeholders of Austrian tax consultancies are taken into consideration, this thesis will be able to draw a holistic picture of the examined topic. Therefore, this thesis will contribute to research on subjective viewpoints on the implementation of big data tools, and in the context of Austrian tax consultancies using a prospective viewpoint.

1.2 Research Questions

In order to fulfill the stated purpose, this thesis aims to answer the following research questions.

R1: How do the stakeholders think big data will change the service quality of Austrian tax consultancies?

R2: How do the stakeholders think big data will change the service range of Austrian tax consultancies?

R3: How do the employees and employers of Austrian tax consultancies think big data will change the existing work environment?

R4: What are the main concerns regarding the use of big data in Austrian tax consultancies?

R5: What are the main differences and similarities between the opinion of the three stakeholders?

(7)

1.3 Scope and Limitations

The field and literature of big data in general, and its technical, social, economic, and ethical facets, are very extensive and complex. Besides the further development of technology and the creation of ethical frameworks, the implementation of big data tools and their impact on companies is one of the most interesting areas of big data research. Since big data tools are already being used in most industries, the adoption of technology and its impact on the company is usually examined more closely. However, a largely unexplored area is how a company’s key stakeholders think big data tools will impact the company at a state where no big data tools are used yet.

The scope of this thesis includes subjective perspectives of employers, employees, and clients of Austrian tax consultancies regarding the potential impact of big data tools on companies, services, and work environment. It should be emphasized that big data tools are currently not or only rarely used in tax consultancies and therefore it is not yet certain in which area they will be used. The subjective opinions of the stakeholders are therefore based, for example, on media reports from other industries, or experiences from other companies, which are combined with the knowledge of the tax consultancies in its current state. To work out the opinions of the stakeholders in an unbiased manner and to capture all facets of the topic, no concrete tool or scenario is analyzed, but rather the general expectations, assumptions, and fears surrounding future implementations of big data. Therefore, instead of analyzing opinions after tools have been implemented, the thesis analyzes opinions before big data tools get implemented. This guarantees that the results of the thesis reflect an unbiased picture and therefore do not suffer from limitations that might arise from the use of tools. This means that none of the participants has been influenced by any use in the past so that the opinions expressed are those which are merely based on the underlying needs in day-to-day business. Furthermore, the viewpoint of Austria as a legislator is not explored in the thesis, because although it sets the legal framework, it does not directly influence implementation within companies.

1.4 Thesis Organization

The thesis is divided into five chapters as shown in figure 1. In the first chapter, the introduction, the research questions, scope, and limitations of the thesis are explained in more detail.

In the second chapter, the Literature Review, the literature on big data, Austrian tax consultancies, and the influence of big data on various business areas is analyzed.

In the third chapter, the methodologies of the thesis are discussed in more detail and justified.

In the fourth chapter, the empirical data is analyzed, and the research questions are answered, in the fifth chapter, the results are discussed and in the sixth chapter, the results get summarized.

Figure 1: Thesis Organization

(8)

2. Literature Review

This chapter establishes the literary framework by reviewing the existing literature on the changes in companies triggered by big data. Besides, this chapter provides an insight into the structure of Austrian tax consultancies and their services. This chapter introduces the chapter by explaining the importance of stakeholders in the design process of information systems.

2.1 The Role of Stakeholders in the Design of Information Systems

The stakeholders of a company play a central role not only for the business model but also in the design of Information Systems (Beynon-Davies, 2009). According to Beynon-Davies (2009), a comprehensive consideration and involvement of stakeholders of an organization is an essential part of the socio-technical design of an information system. Beynon-Davies (2009) argues that by involving stakeholders, a better match between the requirements of Information Systems and the needs of stakeholders can be achieved. This ultimately results in greater stakeholder commitment to the new sociotechnical system. Von Hippel (1976) also underlines the high value of user contribution in the innovation process. The author claims that the users of an organization surprisingly often generate ideas for innovations and therefore have a significant influence on the success of a new information system. Rogers (1996) also analyzed the role of stakeholders in the innovation process and claims that innovation systems must be collaborative and therefore stakeholders should participate in the design of the systems.

The analysis of stakeholders and their requirements follows the infological approach defined by Langefors. The infological aspects of an information system deal with the information that is provided by the system and thus serve the end-users. The counterparts to the infological aspects are the datalogical aspects, which deal with the structure and operations of the information system. Thus, infological problems mostly deal with the interaction between human and system and data logical problems deal with the technical components of the system (Langefors, 1980).

2.2 Austrian Tax Consultancies

The tax consulting sector in Austria differs from other industries, such as financial advisory or risk advisory, in some respects. Firstly, the requirements for setting up a tax consultancy are among the highest in Austria. One of those requirements is that the entrepreneur must have a tax consultant license, which requires passing one of the most difficult exams in the country, and on the other hand that only tax consultants may buy shares of the company, which makes the profession unattractive. Secondly, Austrian tax law is one of the most complex in Europe, so employees of tax consultancies must be highly qualified (Kammer der Steuerberater und Wirtschaftsprüfer, 2020). Besides, tax laws change several times a year, so a lot of training is necessary to ensure that employees are always aware of the current legal situation. Thirdly, the tax consulting sector is highly regulated and directly dependent on the legislator, which is why tax consultants are only allowed to provide tax-related services within the given framework, which makes it difficult to use new technologies (Kisslinger-Popp, 2014).

The services offered by tax consultancies generally include the following:

• Accounting - Tax consultancies prepare the accounts for their clients on a monthly or quarterly basis. Based on this accounting, a VAT return is then submitted, which the entrepreneur is obliged to do by law. For this purpose, tax consultancies often offer services that are indirectly related to accounting. These activities, such as payment management or dunning management are also carried out by accountants (Kammer der Steuerberater und Wirtschaftsprüfer, 2020).

(9)

• Balance sheet accounting - Accountants prepare annual financial statements and, based on these, the tax returns for companies. Besides, they are also responsible during the year for declaring any taxes and preparing reports to authorities such as Statistik Austria (Kisslinger-Popp, 2014).

• Payroll accounting - Payroll accountants prepare the payroll, calculate monthly salary payments, and prepare reports for authorities. Furthermore, they advise companies on labor and social law and personnel management (Kisslinger-Popp, 2014).

• Management consulting - In some tax consultancies there are also management consultants who prepare planning calculations, carry out monthly controlling, assist clients in financing projects, and advise clients on special business management issues (Kisslinger-Popp, 2014).

• Tax consulting - The tax advisers prepare tax concepts for clients, discuss the tax return and the annual financial statement with the clients, and support the client in dealing with civil servants on tax law issues (Kammer der Steuerberater und Wirtschaftsprüfer, 2020).

It should be mentioned that all services can also be used independently, as they are only conditionally interdependent. Furthermore, not all tax consultancies offer all services, as it is quite common for the consultancy to specialize tax issues and therefore not offer accounting services (Kammer der Steuerberater und Wirtschaftsprüfer, 2020).

2.3 Big Data

Big data defines large, diverse sets of data, which are steadily increasing (Hilbert, 2016).

Baesens, et al. (2016) states, that in opposition of the era data management, where most of the data was captured within business boundaries, in the era of big data, data outside of those business boundaries become increasingly available. Moreover, current usage of the phrase ‘big data’ refers not only do the datasets but also to the analytics of those datasets. The concept of big data is defined by four elements (4V) as shown in figure 2, namely Volume, Variety, Velocity and Veracity (McAfee, et al., 2012; Abbasi, et al., 2016). However, the definition of big data varies in the literature. Sometimes a fifth V, the Variability, or a sixth V, the Value, is mentioned (Moura & Serrao, 2015).

Figure 2: 4Vs - Volume, Variety, Velocity and Veracity (adapted from Abbasi, et al., 2016)

(10)

The element of volume refers to the quantities of data in datasets. According to Statista Inc.

(2018), 33 zettabytes of data were created in 2018, that is one trillion gigabytes, which are approximately 8.25 trillion DVDs (Cisco Systems, Inc., 2018). According to McAfee et al.

(2012), the amount of created data doubles every forty months, which will lead to 175 zettabytes of data in 2025 (Statista, Inc, 2018).

The element of variety refers to the different kinds of data. Those are for example pictures, sensor-based data, social media data, text, videos, and much more (Schroeck, et al., 2012).

Therefore, big data is not only the commonly known row and column database, which is known as structured data but also unstructured data (McAfee, et al., 2012).

The element of velocity refers to time, which is needed for the data to be created. When it comes to big data, huge amounts of data are created within seconds. For example, every second, 6.000 tweets are posted on twitter and one single flight with an airplane produces eight terabytes of data (Abbasi & Adjeroh, 2014).

The element of veracity refers to the credibility and reliability of the involved data sources (Abbasi, et al., 2016). Abbasi and Adjeroh (2014) for example, state that social media platforms are highly susceptible to noise due to spam, which constitutes approximately 20 percent of all content on the web. Figure 3 shows those four components of the big data definition with their main characteristics.

Figure 3: 4Vs that define big data (adapted from Abbasi, et al., 2016)

However, Baesens, et al. (2016) and Chiang (2018) argue that the phenomenon of big data is defined insufficiently by those four elements. They argue for a fifth element, which describes the phenomenon out of a business perspective. The authors propose the element named ‘value’.

Chiang (2018) substantiates his proposal with the fact that a lack of generated value constitutes no contribution to an organization. Baesens, et al. (2016) in opposition argue that to describe the fifth element, it is necessary to understand the four-element of big data and think beyond.

Furthermore, Chaudhary, et al. (2015) even claims that big data is mostly seen as business transformation, not IT transformation. The author moreover claims that due to new insights and new levels of prediction-based decision making, big data made business analytics suitable for the mass market.

(11)

However, Hilbert (2016) claims that with the development of big data, the way data is treated changed massively. The author traces this back to increased information flow, information stock, and information computation which results in the emergence of five demarcated characteristics, which are shown in figure 4.

Figure 4: Change in the treatment of data (adapted from Hilbert, 2016)

Firstly, big data merges different sources with different data types. According to Hilbert (2016), this process called ‘data fusion’ can compensate for messy and incomplete data by data redundancy from different sources.

Secondly, big data can replace random sampling on some occasions because big data captures all available data (Hilbert, 2016).

Thirdly, big data is not only the accumulation of data but also the analysis of the data (Fuller, 2015). According to Hilbert (2016), intelligent analytics of large datasets is the key feature of big data, which constitutes a crucial advantage in decision making.

Fourthly, big data is commonly accessible in real-time (Hilbert, 2016). According to Bär, et al.

(2016), real-time analytics promises increasements in terms of business efficiency and profitability.

Lastly, the data that makes up big data is produced whether it is collected and evaluated. This is because the digital footprint is becoming increasingly larger and easier to capture due to social media, the Internet of Things, etc. (Hilbert, 2016).

Concluding, several authors state, that big data is already a well-defined phenomenon (Pospiech

& Felden, 2016; Abbasi, et al., 2016; Loebbecke & Picot, 2015) although many questions are still unanswered (McIntyre, 2017; Lu, 2011). Pospiech & Felden (2016) for example developed a descriptive big data model and proposed a theoretical foundation for big data, which justifies future research in big data. Abbasi, et al. (2014) also claim, that due to big data’s impact on people, there is a need to do further research, but not only out of technical viewpoints, but in other areas such as economics, behavioral science or design science.

2.4 Big Data Tools

As already stated above, Fuller (2015) and Hilbert (2016) claim, that intelligent analytics of large datasets constitute the key feature of big data. Several authors support this claim (Chen, et al., 2012; Loebbecke & Picot, 2015 or Abbasi, et al., 2016) and Baesens, et al. (2016) even claim, that the evolving positive view of big data can be traced back to increasing intelligent analytics. Therefore, analytics of big data is a major part of the whole big data paradigm and demonstrates value for an organization, which can gain a competitive advantage through intelligent analysis (Ghasemaghaei & Calic, 2020). The mentioned intelligent data

Data Fusion

Replacing random sampling

Intelligent Analytics

Real-time Access

Digital Footprint

(12)

analysis is also called big data Analysis in the context of big data and is an important part of the discussion about big data. Abbasi, et al. (2016) sums this up by stating that in organizations, big data is utilized for presenting and monitoring key findings to managers and analysts by using tools that combine traditional data with novel unstructured data such as sensor data, clickstreams or social media data.

Loebbecke & Picot (2015) broadly define the term ‘big data analytics’ as a tool for analyzing and interpreting all kinds of information. Furthermore, the authors claim, that through technical and analytical advancements, big data not only defines the functional range of contemporary tools but also constitutes a crucial factor in the development of artificial intelligence and Business Intelligence. Chen, et al. (2012) also claim, that the combination of big data and intelligent analytics supported a broader development of Business Intelligence, which gets used by organizations to analyze critical business data for a better understanding of services they offer, the market they compete in or customer habits and subsequently for making better decisions within a shorter time.

But Business Intelligence is not the same as big data Analytics (Baesens, et al., 2016). Most of the time, Business Intelligence is confused with big data Analytics, but there are some differences (Baesens, et al., 2016; Chen, et al., 2012).

Business Intelligence (short BI) is an advanced type of reporting, which is based on data management and warehousing (Chen, et al., 2012). BI uses tools to extract, transform, and load (short ETL) to manipulate data, which is then loaded into tools, where it gets analyzed and visualized. Whereas, big data Analytics on the opposite goes beyond analyzing and visualizing datasets for better insights. According to Baesens, et al. (2016), big data Analytics can be described as an analytical method that penetrates the depths of data sets and links certain business transactions with explanatory variables. These links can be used for causal analysis or for predictive analysis, which both should facilitate decision making for organizations.

Predictive analytics especially is extremely interesting for organizations (Chen, et al., 2012;

Wilson, et al., 2017). Egner (2019) traces this back to the fact that from an entrepreneurial point of view, everything is getting faster and faster and the future is much more uncertain than it was twenty years ago.

Wilson, et al. (2017) claim, that big data offers incomprehensible volumes of data, which makes it possible to reveal trends, patterns, and relationships, that could not be revealed with small data sets. Based on these patterns, statistical methods are used to make predictions for the future, which are utilized to support strategic business decisions (Baesens, et al., 2016). Luckily, due to broad interest in big data Analytics, most of the leading companies, such as Microsoft, IBM, or SAP, already implemented those technologies in their tools (Sallam, et al., 2012). Tools that could be used in tax consulting are for example the following.

• Talend (https://www.talend.com/)

• Hadoop (https://hadoop.apache.org/)

• Knime (https://www.knime.com/)

Since the focus of this thesis is not on the technical aspects of big data tools, but on their impact on Austrian tax consultancies, the tools are not discussed in detail.

(13)

2.5 Big Data Change

Big data is already causing significant changes inside and outside of companies, largely due to the high speed, high market complexity, and the need for quick decisions (Kiron, et al., 2014).

Rifkin (2014) for example trace this back to the fact that storage, processing, and analyzing large amounts of data are nowadays at extremely low cost. Therefore, big data can change nearly every process, where big data is involved (Loebbecke & Picot, 2015). Hilbert (2016) breaks this down to a conceptional framework, where the digitization of processes is interpreted as an interplay between policy strategies, technologies, and social change. Therefore, organizations need skilled employees, generic software, and solid hardware. Furthermore, Jones (2018) claims, that big data provides numerous opportunities for companies to learn and enhance their performance because exploring new information constitutes the base for organizational learning. Sooner or later, these opportunities will lead to changes within and outside the company, which will have a significant impact on employees, processes, customers, and technology (Abbasi, et al., 2016). Baesens, et al. (2016) divide those changes into three types, those are altering the scope of optimization challenges, a shift of power in decision making and impact on internal and external processes. Yan, et al. (2015) underline the changes brought by big data on the financial industry, where financial services can benefit enormously from advanced analysis capabilities and outperform traditional banks through economies of scale. Baesens, et al. (2016) therefore emphasize the importance of adopting the new technologies as they can generate enormous added value, which is why the authors claim that big data and big data analytics can be disruptive and speed can be a decisive competitive factor.

Another area that will be significantly changed by big data is human resources within organizations. In their paper Shah, et al.,(2017) analyzed the changes in this area and divided them into five sub-areas.

The first area that big data changed significantly in human recourses is corporate strategy. In his analysis, Marr (2015) has identified several components which, from the employees' point of view, are less customer-focused and more oriented toward the purpose of change. These are day-to-day operations (task distribution, responsibilities, capacities), corporate resources (infrastructure, internal business processes, corporate culture, mission, and vision) and finances (development costs, recruitment costs, and change costs).

The second area in human recourses, which according to Marr (2015)) has already been changed by the influence of big data, is the measurement of metrics and data. This applies not only to external data, such as data from social media, sensor data, surveys, transaction data, etc., which provide data in an unstructured and structured form but also to internal data, which can be used to analyze the performance of individual employees in more detail. This data includes data generated by employees, such as activity logs, interactions with customers on various platforms, or employee surveys, as well as data that is not primarily generated by employees, such as ambient temperature, weather, or the operability of IT systems (Marr, 2015).

The third area in human resources that big data changes is the analysis of the collected data.

According to Rai, et al., (2015), the greatest benefit of big data in the HR context lies in the ability to tap into a variety of sources and obtain valid data from them, and in the ability to analyze the collected data in a meaningful way. According to Angave (2016), one point of discussion is still the extent of the analyses. Without a doubt, there are numerous statistical models and different analysis techniques for all data types, but the authors ask the question of the extent to which these analyses should be performed on employees and thus point to ethical and legal concerns related to the use of big data Analytics in the HR context. The authors, therefore, claim that just because certain analyzes are possible, it does not mean that companies should or may perform these analyses (Shah, et al., 2017).

(14)

The fourth area that big data changes is the reporting of the analysis. Reporting the results of the analysis is an important part of the analysis process since the decisions made based on the analysis depend largely on how the results are reported to the decision-makers (Iran, et al., 2014). The focus lies on the choice of meaningful visualizations that present the findings in the right context and do not distort the results by incorrect scaling. Unstructured data can also play an important role in this process, for example, which data can be meaningfully represented in heat maps, decision trees, or other mappings (Shah, et al., 2017).

The fifth and last area in HR that big data is changing is the transformation of the company.

According to Angrave (2016), the HR department plays a key role in ensuring that data is connected and interpreted in the right context. According to the authors, this means that the HR department is responsible for obtaining the right insights from the analyses and ultimately supporting the management in making decisions. According to Shah, et al.,(2017) both intrinsic and extrinsic causes of poor and good performance should be identified and, if necessary, changed. These changes can ultimately lead to a transformation that simultaneously creates new business opportunities (Shah, et al., 2017).

Taking a closer look at the stakeholders and how they contribute to the success of big data change it stands out, that digital leaders and management play a key role in all processes of change in big data. Liang, et al. (2007) argue that the involvement of managers plays an essential role in the adoption of new technologies. A low participation rate of managers in implementation projects could result in a significantly lower willingness to accept new technologies at the employee level. Galbraith (2014) also discusses this willingness to accept, which in his opinion is measured by whether the new technology destroys or promotes competencies. Therefore, understanding employee attitudes toward new technology is vital for successful change (Shah, et al., 2017).

But big data not only triggers changes but also supports companies in making changes. In the literature, various scholars name different areas in which changes can benefit enormously from big data (Chen, et al. 2012; Galbraith 2014; Marr, 2015). This areas essentially encompass management engagement (identifying issues, data-based decisions, and digital vision), talent management (developing individual resources for transformation), technology design (better design of systems based on usage analysis), decision making (better decision basis) and corporate culture (better knowledge transfer and usage practices) (Shah, et al., 2017).

Big data will also bring about several changes in tax consulting firms. Even though big data currently plays little or no role in tax consultancies, potential changes in the tax industry can already be predicted due to the changes already brought about in other industries. Alnoor &

Willcocks (2014) argue that big data's influence on the collection and analysis of corporate data inevitably changes its role in the provision of financial information. Furthermore, Smith and Payne (2011) point out that the latest technological developments have a potentially transformative impact on the tax consulting sector and that adapting to new technologies is a major challenge.

Furthermore, McIntyre (2017) points out that tax consultancies already manage corporate data, but they should focus on more intelligent analysis of the data in the future. The author argues that this change of focus is since the analysis of financial data combined with business strategies can create massive added value for companies, especially in the area of tax and corporate planning. This massive added value is also emphasized by Alnoor & Willcocks (2014), who claim that accounting information can be used to develop a deeper and broader level of analysis through the use of analytics, which can form the basis for business decisions. Egner (2019), on the other hand, emphasizes that big data itself is only an interim solution, as the goal should be to promote the introduction of Smart Data so that new contexts take center stage.

(15)

2.6 Big Data in Business

Big data and big data analytics get utilized by organizations to gain competitive advantage, create new service, or improve the quality of their services. Davenport, et al. (2010) claim, that big data analytics mainly focuses on improving business processes to gain a competitive advantage in the long term. Highlighting possible competitive advantages, Abbasi, et al. (2016) claim that due to the proliferation of self-service analytics and data-driven decision making, managers and executives increasingly start to use big data for decision making. A study conducted by Accenture and General Electric in 2014 reports that 87% of enterprises believe, that their branch will be redefined by big data in terms of competition and services. Moreover, 89% believe, that companies, which will not adopt a big data strategy, will start to lag, and finally lose market share and momentum (Columbus, 2014).

However, big data can also be a problem for organizations. Davenport (2014) claims, that although organizations can gather data from many sources such as E-Mails, social media, IoT- sensors, or transactional systems, they are challenged to find out how to gain insights or find patterns. Therefore, Abbasi, et al. (2016) claims that organizations are routinely interested in spotting useful data sources, carve out the relevant information, and quantify the value of those sources.

Kiron, et al. (2014) also highlight possible problems questioning the quantification of the value using big data and claiming, that investing in big data demonstrates a huge challenge because although many companies already adopted big data strategies, only a few could increase firm performance. Abbasi, et al., (2016) also think that quantifying the value of big data becomes a major discussion topic from a business value perspective and traditional accounting perspective. Aktar, et al. (2016) for example claims, that there is no reliable link between big data investments and firm performance, whereas Davenport, et al. (2010) claim, that the success of big data tools is based on the design of the tool itself, which should maximize potential competitive advantages. Furthermore, Shi, et al. (2016) also take into consideration, that many experts doubt, that using big data intelligently will improve firm performance, although business press demonstrates the value of being data-driven.

Chen, at al., (2014) on the opposite conducted a study, where the results indicated that the overall IT capability of companies is an important factor when it comes to generating real economic payoffs. Hence, the authors conclude that investments in overall IT capability are important, which is also relevant for big data.

Another study, that encourages those claims was conducted by Shi, et al. (2016) in partnership with McKinsey’s business technology office and students from Wharton Business School and MIT, where 330 executives of public North American companies have been interviewed about their technology and organizational management practices. The data gathered from the interviews were then compared with performance data, which was gathered from annual reports and independent sources. The study shows a positive correlation between companies characterizing them as data-driven and better firm performance. In detail, companies with data- driven decision making were 5% more productive and 6% more profitable than other companies. This was confirmed by Wu, et al. (2015), who examined the effects of big data on overall firm performance. They identified a positive relationship between aligning IT strategy with the business strategy and better overall performance.

Marshall, et al., (2015) looked at the quantification of business value from a different perspective. Their study examined how the use of big data within innovation processes contribute to the success of organizations. As a result, they divided the organizations into three groups based on the effects of innovation processes on firm performance: leaders, strivers, and strugglers (Marshall, et al., 2015). The leaders focus mainly on collaboration and big data Analytics supports innovation processes within a structured approach. The strivers fail to define the most important processes for innovation and therefore are not as successful as the leaders, although they invest a fair amount of money. The last group, defined as strugglers, underly

(16)

other challenges within their company and have no clear innovation process (Marshall, et al., 2015). The authors conclude that companies, where innovation processes are supported by big data and big data analytics, are more likely to outperform their competitors and gain more revenue growth and operating efficiency (Marshall, et al., 2015).

The increased performance of companies using big data tools in innovation processes reflects the necessary changes in business models (Sambamurthy, et al., 2003). Business models, which means how organizations create, deliver, and capture value (Osterwalder & Pigneur, 2010), underly certain pressure, which applied by new technologies such as big data. According to Loebbecke & Picot (2015), big data challenge existing business models in many industries, which forces companies to adapt. Kiron et al. (2014) claim that this can be successfully done when companies manage to establish a culture, which is open to new ideas. On the other hand, Loebbecke & Picot (2015) claim that many established organizations fail to use the opportunities, which are created by new technologies such as big data, and fail to adapt their business models to new economic conditions.

Big data is also an enormous opportunity for tax consulting firms. According to McIntyre (2017), the financial sector, which also includes tax consultancies, offers excellent conditions for the use of big data tools. The author justifies this with the fact that tax consultants receive an enormous amount of valuable data from companies, from which valuable insights can be gained. Bär, et al. (2016) also underlines the far-reaching opportunities for tax consultancies, which result from the development of data sources and the analysis of the data obtained. These are not only limited to measures in the area of process optimization, but also concern new services, which support the company in operational management by analyzing the data of the respective company. However, Alnoor & Willcocks (2014) argue that despite the positive influence big data can have on financial and management decisions, there is no empirical evidence that the industry is taking advantage of the wealth of opportunities.

2.7 Big Data in Services

An important part of companies, which is enormously influenced by big data, is the products and services offered by the companies. Johnson et al. (2017) for example claim, that big data offers the opportunity to effectively and efficiently increase the fit between consumer preferences and product and service characteristics, thereby changing the innovation landscape in all areas in a sustainable manner (Ghasemaghaei & Calic, 2020).

Furthermore, Baesens, et al. (2016) claim, that big data has already begun to break down traditional business boundaries. The authors use the Alibaba company as an example of this phenomenon, which has entered the banking sector through new opportunities and has recently started offering savings funds and insurance services, thus causing massive disruptions in the Chinese banking sector.

Big data and big data analytics are not only a new product in themselves but also enable many new products and services by adding value to data at various levels (Galbraith, 2014; Lehrer, et al., 2015; Loebbecke & Picot, 2015).

Lehrer, et al. (2015) for example claim, that due to the flexibility of big data, it is predestined for evolving around human and material agency and providing spaces for human actors.

Loebbecke & Picot (2015) on the opposite think, that physical and analog products will be either complemented or even replaced by big data, which will serve as a substitution for existing products and services. However, Lehrer, et al. (2015) claim, that companies should take advantage of the synergy effects of new technologies when renewing their products and services, especially concerning the storage, processing, and analysis of collected data.

According to Baesens, et al., (2003), these synergy effects can also make a significant contribution to increasing the company's agility. This is mainly possible because big data enables decisions to be made independently of processes by explicitly modeling decisions and defining logic (Baesens, et al., 2016). However, this clear separation of processes and decisions

(17)

requires sophisticated decision analysis techniques, which benefit to a large extent from standardized decision making (Taylor, et al., 2013).

According to Galbraith (2014), one new aspect of new software is the mode in which it is developed. Nowadays most software applications are developed with agile methods. In this mode, feedback from customers is already collected during the development process and considered for future development stages. This results in a continuous and repetitive development process, which continuously enhances the products and thus adapts them to new requirements and framework conditions. Reflecting on those new development processes, Loebbecke & Picot (2015) think that the new products offer higher benefits to business partners and customers.

By gaining a better insight into the needs and requirements of customers, companies could highly customize their products and services to meet the needs of their customers. This is mainly since big data analyzes customer behavior patterns and, based on this knowledge, tailors’

services to the individual customer. Lehrer, et al. (2015) claims that the high degree of individualization made possible by big data generates enormous added value through improved customer experiences, which has a corresponding impact on success.

In addition to enabling new products and services, big data and big data Analytics can also be used to better analyze customers (Ghasemaghaei & Calic, 2020; Baesens, et al., 2016). Tan, et al. (2015) underline this assertion and state, that companies try to make unstructured data about their customers' access to improve the understanding of their problems and needs (Ghasemaghaei & Calic, 2020).

Erevelles, et al. (2016) also claim, that extracting and analyzing rich customer data provide previously unknown insights to companies. Furthermore, Dong et al., (2018) claim, that the discovery of new customer insights may lead to an improvement in future product developments and therefore a better satisfaction of customer needs.

Ghasemaghaei & Calic (2020) explain this with an example where a company analyzes reviews on public platforms and links them to purchases made on their platform so that they can better analyze preferences and incorporate them into future product developments.

Johnson et al. (2017) furthermore claim, that it is important for organizations to collect both structured and unstructured data from different sources, because gathering different data types from various sources support the implementation of new ideas and products which aim to better satisfy customer needs. Furthermore, the combination of structured and unstructured data helps organizations to view innovation problems from a different perspective and therefore reduces effort and time for innovations.

A look at the tax consulting sector shows that big data is also bringing about changes in the services offered there. On the one hand, big data supports tax consultants in simplifying complex information that represents added value in everyday consulting (McIntyre, 2017). Bär, et al. (2016) mention in this context, for example, well-founded industry evaluations, or more precise forecasts, which identify problems and their effects in advance.

2.8 Big Data and Work Environment

As explained above, big data and big data analytics affect both companies themselves and their service and product offerings. It is therefore not surprising that big data will also impact the working environment within companies. Loebbecke & Picot (2015) address this issue and claim that new technologies will force new forms of division of labor and cooperation within and outside companies. These changes not only concern the flexibilization of working hours, but also the decoupling of work and place of work, as technological progress makes work processes more flexible in terms of space and time. This is due both to the establishment of digital working tools and to the creation of new products and services that do not require spatial presence.

(18)

A major influence of big data on the working environment concerns the speed of the working world in general (Davenport, 2014). With better technology, richer data sets, and better analysis capabilities, more decisions are made in less time. This, of course, also has an impact on the working environment of employees, who need to adapt their work to the changes (Galbraith, 2014).

Another aspect is that big data will also change the company structure itself in the long term.

Galbraith (2014) even claims that a successful implementation of big data even requires a modification of the organizational structure. The authors believe that big data evokes the need for an executive with outstanding knowledge of data and analytics. This leader, known in many organizations as the Chief Digital Officer (CDO), must establish big data as a strategic asset of the organization and ensure its successful impact on the organization's success. Furthermore, Galbraith (2014) proposes, that organizations should hire data scientists and analytics experts to support the use of big data in decision processes. Besides, all departments should have digital experts dedicated to the successful use of data to support the CDO's strategy (Davenport, 2014).

This personnel restructuring within the organization shifts power toward digital skills.

Moreover, the restructuring leads to a shift from judgmental decision-makers toward digital decision-makers, which is particularly important, because decision-makers have a major impact on investment decisions, customer priorities, and new products (Galbraith, 2014). The long term development is to establish a profit center, which will generate its own revenues. Besides, a successful restructuring will create a new organizational structure that will contribute to the successful implementation of the digital business strategy.

However, this restructuring within the organization also requires digitally skilled employees (Davenport, 2014; Galbraith, 2014; Shi, et al., 2016). According to Marshall, et al. (2015) digital leaders view big data as a ubiquitous technology that can add value in all areas. Hence, they invest in intelligent tools intending to make them available to all employees. Furthermore, digital managers invest in the training of their employees so that they have the necessary tools for future challenges. However, Loebbecke & Picot (2015) claim that in the future the workforce will not necessarily have to be employees, but that companies can also fall back on freelancers or cooperation with other companies who can sufficiently meet the company's requirements.

Galbraith (2014) illustrates how a work environment could look like, after real-time analytics and big data have been implemented, with an example. Starting at the bottom, the data and analytics department must handle the incoming data from different sources such as social media, sensor data, financial data, etc. and analyze it. The results are used then for several activities. Firstly, it is used by software developers to create new applications for customers and to adapt existing applications to changing customer needs. Secondly, hardware engineers use the results from the analytics department to check if the production machines work as they should in terms of quality, quantity, and efficiency. Moreover, predictive analytics should tell engineers when they need to maintain which machine. Thirdly, business managers of the organization analyze the firm performance and predict future problems and finally, social media experts use big data to better understand the customers, who use the online community they manage. Melville, et al. (2004) reflect the necessary changes and states, that to successfully implement big data, significant organizational change is often needed. Those changes include not only power shifts, but also policies, rules, culture, and workplace structures (Brown, et al., 1995).

Apart from the necessary changes within the organization and the increasing speed, which could be interpreted positively, there are also negative applications of big data in the working environment (Davenport, 2014). One possible negative effect of using big data within companies is stated by Wilson, et al. (2017), who claim that employers will monitor employee productivity and collect data about computer use, phone calls, and other activities of employees, which could be used against them. But Davenport (2014) claims that this phenomenon of using big data in workforce management is not new, only the volume of data has increased rapidly.

(19)

Furthermore, Loebbecke & Picot (2015) claim, that big data-based systems operate cheaper and with fewer errors than humans. Therefore, more and more humans are replaced by machines with increasing cognitive activities. According to Davenport (2014), big data and big data analytics have only substituted activities with low or medium qualification levels up to now. In the future, however, even highly skilled activities could be substituted by improved technologies.

Another possible negative effect is stated by Shi, et al. (2016), who noticed that many organizations pretend to make data-driven decisions, although the data supporting the decision is insufficient. The authors furthermore claim that many executives use data to support their decision, but they only use data, that support the decision, they already made without the data, which means that decisions are made with bias.

Although the general conditions in the tax consulting sector are different from those in normal companies, certain changes in the working environment are also applicable in this sector (Egner, 2019). The biggest change will probably be in the area of qualifications and training.

At present, employees of tax consulting firms have excellent training in the tax and business management fields, but in the IT field, employees only have a basic knowledge. According to Egner (2019), the existing training courses will in the future be supplemented by more in-depth IT content, which will cover the enormously increasing digital demands on employees. The goal of the training programs should be that employees are specialists in areas such as accounting, controlling, and tax consulting, but also have IT skills that enable them to optimally operate new technology so that the highest level of efficiency and service quality is achieved.

2.9 Big Data – Problems and Concerns

Big data as technology has many advantages but it can also cause some problems. These problems do not only relate to big data itself but also to its influence on organizations and their services, which some authors have reflected in their papers (Baesens, et al., 2016; McIntyre, 2017; Lu, 2011)).

Lu (2011), for example, claims that big data is overly well portrayed by the media, leading many companies to invest excessive amounts of money in tools that can lead to poor performance and reduced agility.

On the surface, the use of big data seems simple, as new technologies such as the Internet of Things make it easier and easier to collect data (Baesens, et al., 2016). The difficult part, however, is to analyze the enormous amount of data and to extract value from it. That is why organizations need to ask themselves whether big data helps them to add value. For example, whether big data helps to improve the current market position, whether big data helps to improve service quality, or whether customer needs can be better identified (Chaudhary, et al., 2015).

Baesens, et al. (2016) also argue that not only a large amount of data from internal and external sources can be a problem, but also the structure of the data. The authors attribute this to the fact that more and more data, especially data from external sources, is unstructured and the effort to analyze this data in a meaningful way is accordingly higher.

Another problem that could arise is the failure to consider the economic value of big data.

However, according to Baesens, et al. (2016), the economic value-added generated by the use of big data tools is a key factor in ensuring that managers gain confidence in analytical models and consequently use the tools for decision making. The authors attribute this to the fact that while some decisions are consistent with results from statistically significant models, the economic value does not meet the expectations of managers, which ultimately leads to distrust.

If we now go from general problems of big data to context-specific problems of big data in the tax consulting sector, it can be seen that the tax consulting branch is sitting on a valuable treasure of data, but has problems monetizing it. This is due in no small part to the fact that many tax consultancies use more than one program to provide the service (McIntyre, 2017).

(20)

One consequence of this is that the data is distributed among various programs with different business logic. Merging this data itself is no longer a major problem, but since tax factors must also be considered when analyzing the data, it is much more difficult to analyze tax data than to analyze customer data, for example (Egner, 2019).

Another major problem regarding the use of big data in tax consultancies is regulatory uncertainty. McIntyre (2017) traces this back to the fact that big data and big data analytics are a relatively new phenomenon, so the use of these technologies in the tax field has not yet been tested to the extent necessary for legislators. It is therefore uncertain at this point how the legislator will shape the future framework conditions, which is why any advances in terms of big data represent a certain risk (Bär, et al., 2016).

One potential point of conflict in the use of big data in the tax consulting sector is the Data Protection Act, which stipulates that data may only be collected to the extent necessary. It is therefore important to find out how to collect data in compliance with the law and then analyze it within the legal framework (Bär, et al., 2016). The minimal approach to the use of data is also supported by Alnoor and Willcocks (2014). The authors explicitly point out that consultants should not be tempted by the amount of data to create a flood of information that would destroy the knowledge gained.

Another point of conflict is mentioned by McIntyre (2017), who points out that the tax authorities are also interested in the data of companies to analyze them for possible tax fraud.

If the data of companies is passed on to supervisory authorities and the underlying context is not disclosed, massive misinterpretations can arise, which can have negative effects on the company. For this reason, all activities relating to big data in the tax field must be carried out with sensitivity.

(21)

3. Methodology

This chapter deals with the research methodologies used to answer the research questions of this paper. This chapter, therefore, provides a brief overview of Information Systems

research, then goes into more detail about the research design, describes the process and methods of data collection and data analysis, outlines the methods used to ensure quality, and concludes with the ethical considerations for this research.

3.1 Information Systems Research

Information Systems (IS) is not only a relatively new scientific discipline, but it also has an interdisciplinary character and thus has a broad field with fluid boundaries to other disciplines (Hirschheim & Klein, 2012). Despite the ongoing discussions concerning the identity and core character of IS (Orlikowski & Iacono, 2001), Information Systems has developed into an influential scientific discipline (Agarwal & Henry, 2005). Baskerville and Mayer (2002) furthermore claim that many disciplines now refer to Information Systems.

This thesis deals with the solution of the infological problem and thus follows the concept as established by Langefors (Langefors, 1980). Due to the importance of the individual views and opinions of the users, the interpretative paradigm was chosen in the thesis. This ensures that the focus is on individual opinions and thus provides deeper insights (Orlikowski & Baroudi, 1991).

Interpretative Research aims to explore phenomena through individual opinions that people give them (Myers, 1997). The core of the interpretative worldview is that social realities are not given, and reality can only be understood through the opinions people give it (Creswell &

Creswell, 2018). Therefore, the interpretative paradigm is ideally suited to explore the expectations, fears, and opinions of stakeholders in Austrian tax consultancies about the potential impact of big data on the branch.

3.2 Research Design

The big data phenomenon has already been the subject of intensive research in recent years, including thousands of papers on ethical, technical, or economic aspects. However, it is currently almost unexplored how stakeholders, whose sector has not yet been directly influenced by big data, think about big data and what expectations, and fears they have. To explore this topic, the case study design was used, which according to Yin (2009) is perfectly suited to explore topics that are contemporary and currently unknown. Yin (2009) defines case studies as empirical inquiries that focus on investigating a contemporary phenomenon within its real-life context. Benbasat et al. (1987) also argue that case studies are viable methods for researching Information Systems.

Furthermore, the case study design was chosen because the research in this thesis concerns a real situation within a complex and unique social system (Hunt, 1991). The opinions of stakeholders must be examined in the current context. The context is that in many industries big data tools are already used and have had a significant impact on these industries. The use of big data tools in the tax consulting sector is also more than likely, therefore stakeholders already have certain prejudices and expectations which they derive from the impressions they have gained from reports, etc. from other industries. According to Shah, et al. (2017) stakeholders can compare those impressions with future perspectives to predict how big data tools could impact Austrian tax consultancies. Using the case study design, it was possible to get to know the phenomenon better and analyze the complexity (Land & Galliers, 1987).

According to Yin (2009), another strength of case studies is the possibility to handle multiple sources of evidence such as interviews, documents, observations, etc. This strength is used in the thesis in that the data comes partly from interviews and partly from a survey with open- ended questions.

(22)

3.3 Empirical Setting

The empirical setting of the study is the Austrian tax consulting sector. As shown in figure 8, the focus lies on those people who have the most contact with tax consulting firms. These are the employers who run the businesses of the tax consultancies, the employees who are employed by the tax consultancies, and the clients who use the services of the tax consultancy.

This constellation is illustrated in figure 5.

Figure 5: Austrian Tax Consulting study dimensions:

These three groups of people are called stakeholders in this thesis. The in-depth examination of the opinions of these three stakeholders concerning the potential impact of using big data in the tax consulting sector should provide a detailed picture of their expectations, fears, and opinions and thus form the basis for the socio-technical design of big data tools for the tax consulting sector. It could be argued that the Austrian state is also an important stakeholder and should, therefore, be considered in this thesis. However, this thesis should analyze the opinions of those three stakeholders who directly benefit from better design of the information system. Since in this case the government merely provides the framework conditions through laws, the functionalities of future systems are irrelevant. Therefore, no representative of the Austrian legislative was considered in this thesis.

3.4 Methods for Data Collection

As described above, two data collection methods are used to collect the data. The use of interviews combined with a survey with open-ended questions is intended to ensure that, on the one hand, the phenomenon can be carefully analyzed and, on the other hand, that the context is sufficiently explored. Furthermore, some researchers suggest, that through the triangulation of data, the quality of the empirical data increases (Burgess, 1984) and it assures the validity and reliability in single case studies (Dubois & Gadde, 2002) In the following section the two data collection methods are described in detail.

The rise of Big Data in Austrian tax consultancies

Master Thesis