In second phase of the study (figure 1), qualitative semi-structured interviews based on the digitalization framework were conducted. Qualitative research ap-proach “is an apap-proach for exploring and understanding the meaning individu-als or groups ascribe to a social or human problem” (Cresswell, 2014, 4). In other words, the aim of qualitative research is holistic understanding of the phenome-non in a real-life context from the study subject point of view (Erikson & Ko-valainen, 2008, 4-6; Hirsjärvi, Remes, & Sajavaara, 161). The aim of the empirical part of this study was to gather information how digitalization is managed in the selected companies and what where the experiences of the study subjects in do-ing this in the context of business management in general. Thus, qualitative re-search approach was found to be suitable for the study. Next, the sampling, study subjects, interview data collection method, and data analysis process are de-scribed.
4.2.1 Sampling
In quantitative studies, sampling is crucial for the validity of the results.
Sampling must represent the population that the hypotheses concern. In qualitative studies, the population concept can also support the validity of the results. Sampling within a defined population can reduce variation caused by external factors and, thus, it supports the within population validity of the results (Eisenhardt, 1989). In this study, the aim was to minimize variation by focusing on small service companies on the Central Finland area.
Partly the selection was done according to the convenience sampling princi-ples. In convenience sampling, study subjects are chosen from those that are most available (Cooper & Schindler, 2014, 152). What this meant in this study that the focus of the study was partly determined by the willingness of the study subjects to participate the study. Furthermore, the companies were selected among those, that had been participating another study that focused supporting digitalization in Central Finland area. Furthermore, the problems in digitalization concern mainly implementing new technologies to support business (Digibarometri, 2017;
TIVIA, 2015). Based on the previous study with these companies, it was known that they had applied digital technologies to support their business and they were not originally born as digital companies but developing as digital compa-nies. Consequently, they all had experience in digitalization and managing it and, thus, suitable as study subjects. However, sampling was also based criteria that narrowed and defined the target population as explained in the introduction.
Five companies were selected to be interviewed for the study and the per-son of the company who was responsible of managing the digital was chosen to be interviewed. The persons were also the managers of the company or co-man-ager in case of one study subject. Thus, single person interview per company captured both the business and technology management views. According to Ei-senhardt (1989) the sufficient number of cases for theory building is usually be-tween 4 and 10. However, optimally in qualitative studies sampling should not be predetermined and sampling should continue until no new insights are pro-duced within the scope of the study, although it is recognized that time and money are limiting factors (Eisenhardt, 1989). This study was not aiming to cap-ture all the variation and insight of the phenomenon to build new theories (Gregor, 2006) and timeframe was limited. In addition, the analysis of the results demonstrated widely homogenous views among the interviewed companies.
Thus, the sample size was considered to be suitable for the study.
4.2.2 Study subjects
Table 2 summarizes the basic information of the study subjects including infor-mation about the company and the interviewee. All the companies were about the same age, established 5 years ago and they all provided some type of service for private customers. Besides C2, whose customers and operations covered the whole Finland, all the companies were operating highly locally. When the inter-viewees were asked about the service they provide from the customer point of view, most of them were able to formulate a value proposition. However, C3 and C5 described only the tangible output of the company instead of demonstrating the ability to see the customers perspective. Further, C1 had not defined any spe-cific customer group for the service. On the other hand, the health services pro-vided by C1 and C4 are not very specific in terms of customer types but con-sumed by all types of people.
Companies C1-C3 were all run by single entrepreneur or a married couple whereas C4 and C5 gathered together several single entrepreneurs under an um-brella company. When asked about partnerships, none of the companies replied to have any. However, some were involved in activities that could be classified as partnerships. C1 was also working as a subcontractor within a larger company proving wider variety of health services. Thus, the companied were in a mutually beneficial relationship. C2 on the other hand was entering a platform that com-bines the sales of many similar service companies.
Besides C5, all the interviewees were in the managing position within their companies. However, in company C5, the interviewee was responsible of the managing the digital component of the company, like all other interviewees within their companies. Thus, considering the aims of this study, they all were correct choices to be interviewed. Furthermore, all the interviewees expressed innate interest in digital technologies and developing them within their compa-nies. All companies had also applied a variety of digital technologies to support their business and they all had continuously been developing this area since the establishment of the company.
To summarize, there were many similarities between the companies but also major differences. One main difference was the company size between C1-C3 and C4-C5. In addition, C2 was operation on a wider area than the other and, finally, in C5 the person responsible for technological issues the primary manager of the company but co-managing it with others.
TABLE 2 Basic information of the companies
C1 C2 C3 C4 C5
Market area local Finland local local mainly local
Main
manager manager manager CEO co-manager
4.2.3 Interviews
Surveys are the dominant data collection technique in information system studies and can be used for qualitative data collection (Palvia et al., 2015), and interviews are another major data collection method in qualitative studies (Myer & Newman, 2007). Surveys use questionnaires, for example, with multiple choice or
open-ended questions and they have the possibility to collect data easily from wide group of subjects (Cooper & Schindler, 2014, 152-165). On the other hand, in interviews the researcher is in direct interaction with them in the data collection.
In the interviews the researcher is in a direct contact with the research subjects which has many benefits goals. The main benefit is the ability to adapt the data collection depending on the subject, which support better understanding of the questions and the answers (Cooper & Schindler, 2014, 152-165; Hirsjärvi, Remes,
& Sajavaara, 2009, 204-207). Since the variation of IT competencies in the companies was expected to high, the interview was a good choice for data collection.
There are also many alternatives of conducting interviews depending on goals. Erikson and Kovalainen (2008, 80) summarize the types of qualitative in-terviews as follows:
1. Structured interviews have predefined questions that are repeated simi-larly in all interviews
2. Semi-structured interview has predefined topics, issues, or themes and variation with the questions and order of them can vary between inter-views.
3. Unstructured interviews can have a defined starting point, but they pro-ceed freely.
Structured interviews give the most comparable results and are more suitable for unexperienced researchers. However, limitations include the lack of flexibility.
Semi-structured interviews, on the other hand, provide more flexibility, which allows to follow unexpected directions in the interview, and gives more tools to support mutual understanding. Furthermore, this type of interview allows wider types of question (what and how) to understand the views of the interviewees.
On the other hand, one limitation of the semi-structured interviews is that it re-quires more experience (Erikson & Kovalainen, 2008, 81-82). However, in this study the interview was based on well-defined concept, which helped to coordi-nate the interview and gave systematicity even without extensive experience.
Thus, semi-structured interview type was selected.
Interviews conducted face-to-face support the interaction with the study subjects and separates them from surveys. However, there some are benefits of using new technological solutions for interviews. Phone interviews support bet-ter time management, and the inbet-terviewee can select most convenient time and place for interviews, which can also lower the social dissonance i.e. raise the com-fortability of the situation in many ways (Myer & Newman, 2007). Furthermore, Sturges and Hanrahan (2004) suggest that phone interview can provide compa-rable results with those done face-to-face. Consequently, phone interview was chosen to be suitable for this study as they it removed the limitation of reaching companies around the Central Finland area.
The interview language was Finnish, and the structure and the questions were planned in that langue. Appendix I presents original interview structure and appendix II a version translated in English. The interview reflects the themes
of the digitalization framework (figure 5) through which the digital business components and management were conceptualized. Thus, the questions were de-signed to reveal how digitalization was managed and what was challenging in it.
Further, the questions followed the teleological view on change which was done to capture the change management capabilities in the companies.
Interviews were conducted during October 2017. The companies were con-tacted earlier about the willingness to participate the study, the aim of the inter-view was explained to them, and time for the interinter-view was agreed based on the interviewee preferences. The lengths of the interviews were 41, 48, 28, 30 and 17 minutes for companies C1-C5 respectively. All the conversations were recorded using Call Recorder software by Appliqato for later extraction and analysis.