Data collection

When considering qualitative research, there are multiple different data collection methods available such as interviews, observation, and queries (Metsämuuronen 2008). Depending on the researcher and the research problem, several methods can be used together, or one method can be enough (Tuomi & Sarajärvi 2018). The data for this research is collected with interviews. Alternative data collection methods were considered as well during the thesis writing process. However, since the goal of this study is to gather in-depth information about customer experience from the customers’ point of view, the interviews were selected as the most suitable primary data collection method.

Furthermore, the interviews of this study are semi-structured. According to Saunders et al.

(2015), semi-structured interviews are a suitable, non-standardized, data collection method for qualitative research because of flexibiltiy. In semi-structured interviews, the interviewer has questions and themes which are written down beforehand, but the interview structure can be modified as the interview proceeds. Additional questions can be also presented during the interviews. (Creswell & Creswell 2018; Saunders et al. 2015.) Asking the additional questions can help the researcher to find out new relevant information regarding the theoretical framework of the research (Tuomi & Sarajärvi 2018). Metsämuuronen (2008) points out that if the topic, themes, and interview questions are rather sensitive or difficult, a semi-structured interview method helps to keep the conversation flowing smoothly. Therefore, semi-structured interview is identified as a suitable interview type for this research.

In this study, the interviews are conducted as one-to-one conversations between the researcher and the interviewee. An advantage of one-to-one interviews is their ability to generate a rich, detailed, and in-depth knowledge about the phenomenon under investigation (Creswell &

Creswell 2018; Saunders et al. 2015). Other common benefits from conducting one-to-one interviews are the fact that more information is shared, it is easier to ask follow-up questions, and interviewees are not affected by other candidates. However, the downside is that first impression biases may take place and one-to-one interviews are often more time consuming compared to the group interviews. (Saunders et al. 2015) Based on these advantages and the nature of the research problem, one-to-one interviews were chosen as an interview method for this research.

The final interviews were conducted as phone interviews and face-to-face online video interviews. Because of the current COVID-19 situation in Finland, face-to-face interviews were not possible to organize and therefore online interviews as well as phone interviews were used to gather the data for this thesis. Depending on the customer preferences, some of the interviews were organized online via Microsoft Teams platform whereas others were conducted via mobile phone. Furthermore, the data was captured by audio-recording all interviews.

3.3.1 Interviewee selection

The interviewee selection was based on the goals and delimitations of this study. Since this study is completed in the Finnish market exclusively and in B2C context, all interviewees are Finnish citizens and they speak Finnish language. Most importantly, all interviewees have used at least one time the online store of the case company for ordering products.

In total, 29 individual customers were invited via email to participate in the research interview.

These customers had used the online store for ordering clothes and accessories. The email addresses of the customers were received confidentially from the case company. If the interviewees did not react to the first email invitation, follow-up emails were also sent. The invitation emails were sent during May and June 2021. Despite the effort, it turned out rather challenging to find customers who were willing to be interviewed. Finally, seven customers agreed for the research interview.

An interview questionnaire as well as informed consent were sent to the interviewees beforehand, so they had a possibility to prepare for the interview. According to Tuomi and Sarajärvi (2018), it is important to give participants time to go through interview questions.

Hence the participants feel more comfortable in the one-to-one interview situation and their answers are more thorough (Tuomi & Sarajärvi 2018).

The interviews were organized in May and June 2021. All interviews were conducted in Finnish language since it was the native language of the participants and the researcher. Welch and Piekkari (2006) suggest that foreign language use in interviews threatens the accuracy of responses, which supported the decision to conduct the interviews in Finnish language.

Before the interviews, all participants had read an informed consent about taking part in a research study. They were informed about the purpose of the study, procedure, risks, and benefits of being in the study, anonymity, and voluntary nature of the research. In addition, the participants were aware that the data would be captured by audio-recording the interviews. The basic information about the conducted interviews is presented in the following Table 4.

Table 4. Basic information about the conducted interviews.

The analysis of the interviews started with transcribing the content of audio-recordings.

Transcribing was done soon after the interviews. Thematic analysis was chosen as a method for analyzing the qualitative interview data. Thematic analysis is a popular data analysis method in qualitative research and the goal is to identify and analyze themes which can be found from the data (Braun & Clarke, 2006). First, the material from the interviews was divided into themes to make the analysis easier. The themes are based on the sub-research questions.

The aim is to find common themes, which many of the interviewees brought up. According to Braun and Clarke (2006) thematic analysis can produce insightful and trustworthy findings,