**4. Results**

This chapter discusses the findings from the literature search and the first step of the literature analysis. Afterwards, we present the results with a focus on the second and third step of the analysis.

#### *4.1. Findings from Literature Search*

The first step in our literature search was the search in the research field of shadow IT integration. Hereby, as expected due to the topic being not very well covered, only one study emerged. Second, we searched the literature on IT integration. Relying on an earlier literature review [11], we collected 14 studies.

From the additional search in the scientific databases, we only collected one more study, because the others did not mention specific cost dimensions. Vague statements regarding costs are rather common in this field of study [19]. Last, we consulted the literature on shadow IT. Based on an earlier review [28], we collected seven relevant studies that mentioned specific shadow IT benefits. The additional search in the databases resulted in two more studies. In total, after removing duplicates, we collected 25 studies (Table 1).


#### *4.2. Shadow IT Integration Drawbacks*

The second step in our research process was the analysis of the literature using the principles of the PBM. As a result, we found three different dimensions of shadow IT integration drawbacks: the switching cost dimension consisting of the procedural, financial, and emotional drawbacks [36], as well as the loss of shadow IT benefits from the shadow IT literature. The second dimension is the organizational-technical dimension and the third dimension the global-local dimension.

#### 4.2.1. Analysis from a Switching Cost Dimension

We assigned the found drawbacks to the four pre-set codes: financial, procedural, and emotional switching costs, and loss of shadow IT benefits. Table 2 shows the results of our coding and the corresponding sources. This sub-section focuses on the switching cost dimension but also gives reasons why we included them in the context of shadow IT integration.


**Table 2.** Result of Coding of Integration Drawbacks to the Switching Costs Dimension and Sources.

The first factor in the switching cost dimension is financial expenditures that are "financially quantifiable resources" [36]. Those expenditures are monetary drawbacks, such as expenditures for integration hardware, software [12,18] mentioned by a few studies, or the initial adaption of the integration technology to the needs of the organization referred to by more than half of the studies [17,32,45]. Additionally, some include the costs that the maintenance of an IT integration causes [13,18]. Because shadow IT occurs in various forms [5], the monetary drawbacks for implementation, development, and maintenance also differ. Additionally, we include the external support. Although only one source mentioned this factor explicitly [12], depending on the integration technology, organizations might need external support to find suitable integration technology as well as to adapt it. Although the organization also must spend time and effort on the search and the coordination, the resources that the support needs are financially quantifiable.

Some drawbacks consist mainly of the procedural switching costs, which are "expenditures of time and effort" [36] that they cause. Factors included in this category are the employee training as well as the project coordination. Project coordination includes the time of the project team in general [12], but also the effort of coordination between departments [17] or the planning of the process and the needed

resources [46]. Depending on which technology the organization has chosen to integrate the shadow IT with the enterprise system, those expenditures may vary as well. Some studies stress that the organization has to understand [47,49] and then select the right integration technology [12,48]. Besides, research mentions that the organization also has to understand the process behind IT integration, which requires time and effort [15,18]. Additionally, top managemen<sup>t</sup> support is very important [12,13], and the organization also, in some cases, has to restructure their processes [18,30] and communicate the change [46]. Those factors, again, also concern shadow IT integration. Most studies mention *culture change* as important. This includes changing the culture [13,18] due to the resistance to change that it causes in members of the organization [15,46]. Given the high involvement of business units during the lifecycle of shadow IT [4], we assume that this factor is especially relevant for shadow IT integration.

Relational switching cost, the "psychological or emotional discomfort" [36], is the third category. Factors that cause relational switching costs are sharing data [18] and using the new technology [12,30]. Losing power is an organizational factor that causes discomfort in certain members of an organization [18,49]. This factor might be especially relevant, because shadow IT research focuses on it as well [4,57].

The fourth category of switching cost is the loss of shadow IT benefits. Thereby, we found three important factors. Shadow IT is often innovative, in terms of processes or even technology [6,52,54]. An IT integration might eliminate this innovation if done with few considerations [10]. Another factor is the productivity that increases in a lot of cases [5,22,56]. Depending on how the organization handles the IT integration, this productivity increase might ge<sup>t</sup> lost. As a last factor we mention the loss of flexibility. Business units are able to adapt their solution to changing needs very easily [27,55]. An IT integration might eliminate the ability to adapt the solution in an easy and flexible way.

4.2.2. Analysis from an Organizational-Technical Dimension

Besides allocating the integration drawbacks to the switching cost dimension, we also analyzed them on the organizational-technical dimension (Table 3). Thereby, the organizational aspect of shadow IT integration comprises mainly the business unit and its interaction with the system and the avoided enterprise system [4]. Factors that we included here because they are mainly organizational are external support; the top managemen<sup>t</sup> support; the organizational restructuring; the cultural change; as well as the loss of power, flexibility, and productivity.


**Table 3.** Result of Coding of Integration Drawbacks to the Organizational-Technical Dimension.

The technical aspect focuses on the shadow IT technology and its support structures [4]. Here, we associate the monetary factors of the hardware and software.

In other factors, the organizational and the technical aspects are intertwined and a differentiation between the two aspects is hardly possible. Hereby, we assign the development and maintenance of the integration technology, which depends both on the technology but also on the developers [18]. We also included project coordination, because one source mentions the influence of tooling during this task [46]. The employee training and technology understanding both depend on the actors as well as on the technology. Sharing data and using the technology are discomforts that stem both from the user and the integration technology [12,30]. Losing innovation can have an organizational aspect if it comprises process innovation, but also a technical aspect if it involves the introduction of new technology [54].

#### 4.2.3. Analysis from Global-Local Dimension

The third analysis of the shadow IT integration drawbacks focuses on the global and the local level of influence (Table 4). In our analysis, the global level is comprised of the organizational IT managemen<sup>t</sup> that is responsible for IT integration, often represented by the IT department. Here, we assign the tasks of the project team, which are project coordination [46], technology understanding, organizational restructuring, and process understanding [12]. Top managemen<sup>t</sup> support is required from the managemen<sup>t</sup> of the organization [45] and is therefore also a global factor. The loss of innovation is also a factor on the global level, because shadow IT technology is innovative for the whole organization [54].


**Table 4.** Result of Coding of Integration Drawbacks to the Global-Local Dimension.

In our analysis, the local level represents factors that primarily affect the business unit. This is, because it is a crucial actor in the context of shadow IT [4]. One important factor is the culture change that the business unit must undergo due to a new technology [13]. Additionally, they experience discomfort through sharing data [18] and using the technology [12]. Also, business units might lose flexibility in their work routine due to IT integration [55].

We assign some factors to the interaction between the two levels. Expenditures for hardware, software, development, and maintenance of the solution might occur on the global but also on the local level. Who pays for the IT integration depends on the cost structure of the organization but also on the

negotiation between the IT and the business department. Employee training is a factor that occurs on both levels. The IT department needs training to understand the integration technology [18], and the business unit needs training to use the technology [12]. Additionally, either the IT department might lose power [13,57] or the business unit [4]. Losing productivity also occurs on both levels, because while the business department loses productivity in their daily work, this productivity loss also affects the productivity of the whole organization [27].

#### *4.3. Conceptual Framework for Shadow IT Integration and Discussion*

Given the results of our literature analysis, we can reply to our proposed research questions: Which drawbacks do organizations face when deciding about the integration of shadow IT and an enterprise system? Figure 2 summarizes these findings, including the assignment of each factor to the three different dimensions: the switching cost dimension, the organizational-technical dimension, as well as the global-local dimension.

**Figure 2.** Shadow IT Integration Drawbacks—Conceptual Framework.

Our framework presents an overview of the 19 different factors that organizations must consider when integrating shadow IT with redundant enterprise systems. It integrates prior studies on integration costs and integration barriers. Additionally, it enhances and specifies then to the particularities of shadow IT. Thereby, it adds a theory base to IT integration research [19]. With the framework, organizations can first assess for each of the shown factors whether it is existent in their IT integration case or not. Second, organizations can use the framework to derive the impact that the relevant factors have. The visual representation allows organizations to consider multiple dimensions

at the same time. Thereby, it enables organizations to keep an overview about the different aspects that influence the factor and to target it accordingly.

From the switching cost dimension, they can identify whether the factor consists of monetary or non-monetary aspects [36]. A financial factor implies that organizations must assess the costs and include them in their budgeting processes. However, our framework shows that most of the drawbacks are not purely monetarily driven. Therefore, organizations should pay the same attention to each of the other three categories. A procedural factor indicates that organizations must identify the time of involving relevant stakeholders during the integration project. Afterwards, they must observe it in their resource planning. Change managemen<sup>t</sup> is important, as it is a complex task that organizations must plan and execute carefully in a timely manner. An emotional factor highlights the existence of specific fears, which organizations need to address during their communication and change management. The introduction of the fourth category of drawbacks targets the particularities of shadow IT [11]. A loss of benefits has an impact on the choice of the integration technology, which should allow the flexibility for the business unit or keep the innovation. As a result, organizations align with the requirements of digitalization [1].

This overview of the organizational-technical dimension helps organizations to manage the social processes during the IT integration process [40]. Our framework indicates that the organizational aspect is as important or even more important as the technical one, given the number of factors that we assigned to the organizational side. This notion confirms prior research on the emergence of shadow IT [22] and stresses that organizations have to observe and manage the social and organizational processes.

The global-local dimension is especially important in shadow IT integration. Prior studies indicate the importance of the business unit that implements and maintains the shadow IT [4]. Our framework helps to keep in mind the actors on the different levels and enables organizations to target them with the right IT integration measures. Our results indicate that the global tasks of the IT department mainly consist of managing the IT integration process. However, most of the overall tasks occur in the interaction between the two levels. This points at the importance of the coordination and cooperation between the IT department and the business unit, which has been already pointed out by past research [5,22]. Thereby, it stresses the role of the business units and encourages organizations to actively seek and monitor their opinion toward possible IT integration.

#### **5. Conclusions and Future Research**

The goal of this paper was to present a conceptual framework of shadow IT integration drawbacks. To reach this goal, we conducted a literature review on IT integration costs and barriers but also on shadow IT benefits to be able to capture the particularities of shadow IT. As a result, we developed a framework of shadow IT integration drawbacks based on the concept of switching costs in the context of path dependency. The framework is multi-dimensional and includes organizational-technical factors as well as factors of global and local influences.

Practitioners can use our framework during the process of deciding whether to integrate shadow IT and an enterprise system. They can collect evidence following the presented criteria. The switching cost dimension helps to identify what type of drawback a specific factor causes, which organizations can then target using financial resources, change management, planning of resources, and choosing the right integration technology. The global-local perspective helps to identify and target the appropriate stakeholders.

We theoretically contribute to IT integration research by integrating the research streams of costs and barriers. Additionally, we enhance the research on shadow IT that can use this framework when coming to an IT integration decision. Besides, research can benefit from our framework to tackle the problems of digitalization that have to weigh letting business units innovate and integrating IT systems for data integrity in the enterprise architecture. Furthermore, the framework is based on the concept of path dependency and introduces this theory in the context of an IT integration decision.

Certain limitations are also present in our research upon which future research can be based. First, the framework presents factors but no measurement for these factors. Therefore, further studies may provide a measurement. Second, although based mainly on literature on case study research, the framework is only conceptual. It has not been evaluated in practice, which should be a focus of future research. Third, to come to an IT integration decision, the drawbacks must be integrated with the benefits of shadow IT integration. Future studies might develop a framework that includes the benefits as well as the drawbacks to gain a holistic view of shadow IT integration.

**Author Contributions:** M.H., C.R., S.Z. and C.F. are all part of the research project. M.H. performed the literature search and the analysis. S.Z., C.R. and C.F. reviewed and refined the analysis. M.H., S.Z., C.R. and C.F. mutually discussed the results as well as the scientific contributions and determined the structure of the research paper. M.H. wrote the paper and consulted S.Z., C.R. and C.F. regularly to receive their feedback on the structure and the content of the current status.

**Funding:** This research received no external funding.

**Conflicts of Interest:** The authors declare no conflict of interest.
