4 Methodology
4.3 The Sample Design Process
During the sample design process the target population is identified, as well as the sampling frame, sampling technique, and the size of the sample.
4.3.1 Define the population
Initially, the target population has to be identified. The target population is the specific group relevant to the research project, the group that posses the information relevant to the researcher. (Malhotra, 1996)
The target population for this study is customer companies and retailers to Jeeves Information Systems AB. The customer companies should operate within the production or service industry. These two industries was chosen since the activities in production and service companies can be quite different and it’s interesting to investigate if this has any influence on the CRM requirements. The limitation, to only look at two industries, was made due to the time constraints for this study.
In order to select suitable study objects, three criterions that the companies had to fulfill were developed. First, the companies should currently apply CRM and use the Jeeves Marketing Module, or the company should have bought the Marketing Module in Jeeves Enterprise and be in the starting phase of implementing CRM. This will ensure that the company already is familiar with the CRM approach, which will enable them to provide relevant and interesting information when being studied. Secondly, the companies should benefit from applying CRM; otherwise they won’t be able to provide interesting information. In order to ensure that the companies would benefit from using CRM, the company should apply more of a relationship marketing
to profit from CRM, as well as their marketing approach needed to be studied. The final criteria, is that the companies have a CRM system with quite high CRM complexity, in order to enable the companies to provide information on CRM requirements within many different business functions. If the company is in the starting phase of applying CRM it is important that they are aiming at a solution with high complexity.
One criterion on the retailers is that they have great experience of CRM systems. This is required to ensure that the retailers will provide high-quality information when being investigated.
4.3.2 Determine the sampling frame
After the target population has been defined it’s time to identify the sampling frame, which is a representation of the elements of the target population. It consists of a list or set of directories for identifying the target population, such as a telephone book, a directory listing the firms in an industry, or a mailing list. (Malhotra, 1996)
The sampling frame for this study is the customer record of Jeeves Information System where all the customers are listed. The sampling frame also includes all retailers of Jeeves Enterprise.
4.3.3 Select a Sampling Technique
When taking a sample there are two major techniques to choose between, these are probability versus nonprobability sampling. In probability sampling every element in the population has a known nonzero probability of selection. Nonprobability sampling is a technique in which units of the sample are selected on the basis of personal judgment and convenience. In judgment sampling the sample is selected based on some appropriate characteristic of the sample members, and in convenience sampling the units or people most conveniently available is selected. (Zikmund, 2000)
In this study, the nonprobability sampling technique have been used. The Marketing Director at Jeeves Information Systems and some retailers were asked to recommend suitable companies. This since they have a lot of knowledge about the customers. Thus, the selection of study objects was based on the judgment of experienced individuals, which is in accordance with the sampling technique called judgment sampling. To receive information from the retailers concerning suitable companies showed to be time consuming, since it was difficult to get hold of the retailers and since they didn’t respond straight away. The selection of retailers was also based on the judgment of the Market Director and the CRM Product Manager at the assigner company.
Convenience sampling is another nonprobability technique that was used in the study. This sample technique was used since the time frame and economical aspects implied restrictions on the geographical area where the samples were situated. However, many of the samples are situated close to Stockholm, thus the restrictions of the graphical area did not affect the selection noticeably.
When selecting which persons to interview the judgment sampling was used, this since it was very important to find the most suitable respondents with good knowledge about their company, as well as about the CRM approach.
4.3.4 Determine the sample size
In this step the sample size is selected, implying that the number of units included in the study is selected. According to Zikmund (2000) it’s difficult to determine the size of the sample, and in order to make the right decision different factors must be considered. The nature of the research is such a factor. For exploratory research, using a qualitative approach, the sample design is usually small. Limitations of money and time also influence the selection of sample size.
For this study a rather small sample size was selected. Two customer companies were chosen, as well as two retailers. The reason to why the sample size was limited to involve four study objects is due to that it’s suitable when applying a qualitative approach. Another reason is the time constraints for this study.
The customer companies selected are anonymous in this study, they are named Company A and Company B. The companies preferred to be anonymous since the study includes information that is sensitive to the companies.
The persons suitable as respondents were selected based on recommendations from the Marketing Manager and the CRM Product Manager at the assigner company. As well as on recommendations from the companies. The position of the respondents are presented below.
Respondents at Company A: Respondent A: Marketing Manager
Respondent B: IT Manager Respondent C: Logistics Manager Respondent D: Export Manager Respondent E: Service Manager
Respondent at Company B: Financial Manager
The retailers selected to represent the retailer case are Jan Stenberg and Pär Heed, who are more thoroughly introduced in the empirical chapter.