Internal Marketing Communication

(1)

Hanna Sloberg Sara Nilsson

Internal Marketing Communication

Alpha, a Machinery Business

Business Administration Master’s Thesis

Science in Business and Economics

30 ECTS

Term: July 2019 Supervisor: Bo Rundh

(2)

Acknowledgements

The idea for this thesis springs from our joint interest in the field of science in business and economics, and more specifically the field of marketing. We thought that it would be interesting to work together with a company, and our paths with ‘Alpha’ was crossed as they wanted help with examining their internal marketing.

Respectfully we would like to thank the Senior Manager of Marketing and Communications at Alpha Machinery Business Line, for including us in this journey of investigating the strategies associated with internal marketing. We would also like to show our appreciation to all managers and supervisors as Alpha that has given us insightful feedback along the way. A special gratitude is expressed to all of the respondents in Finland, China, Sweden, Italy, the U.S.

and Canada for participating in the study.

We express a gratitude towards our own dedicated contribution and our ability to speak freely and be open for each other’s ideas. It has been a rewarding process. Lastly, but not least, we appreciate the help and guidance that we have received from our supervisor professor Bo Rundh at Karlstad University, Sweden.

Thank you!

May, 2019

Hanna Sloberg and Sara Nilsson

(3)

Abstract

The aim of this study is to investigate the effectiveness of the internal communication, as a dimension of internal marketing, in a global multicultural company. The study was based on a conceptual model that is a miscellany of theoretical concepts of how to create effective internal marketing communication for global organizations. Through the investigation of the company referred to as Alpha, a quantitative study across six countries was conducted to further the knowledge on how to address the needs of internal customers and adjust the internal marketing strategy thereafter. A questionnaire was sent to a random sample, with a response rate of 215 employees that together reflected the total population of 2831.

Through the usage of ANOVAs, the findings displayed significant differences of how the employees in all countries perceived the internal communication at Alpha. The study also compared differences within two countries where there was enough data to investigate differences between categories of employees, these results were not significant. In general, it can be said that the results were grouped by the differences of the three European countries against the three non-European countries. Another prominent finding was that China was separated from the other countries, this was also the instance collectively shown for Sweden and Finland as they were often grouped together. The implications are that possible differences in business culture may have affected these results, which further studies need to investigate. The results jointly report that the internal customers are not satisfied with the internal communication. The conclusion is that the needs of the employees at Alpha should be addressed much further, as a part of the company internal marketing communication strategy.

Key words: Internal Marketing Communication, Internal Marketing, Internal Communication, Internal Marketing Strategy, Internal Corporate Communication, Internal Customers, Employee Loyalty, Social Capital, Commitment, Trust, Associability, Finland, China, Sweden, Italy, The U.S., Canada.

(4)

1. INTRODUCTION

The first chapter includes the following sections: Background, Problem Discussion, Aim, Delimitations and Contribution of Knowledge.

Background

The notion of the ‘internal marketing’ concept was developed about four decades ago (Berry 1981 in De Bussy et al. 2003) and includes interventional coordination for implementing marketing strategies effectively (Ahmed & Rafiq 2003). A way to understand internal marketing is to define companies as markets at which the employees are the customers (Varey 1995; Berry 1981 in De Bussy et al. 2003). This is why employees can be seen as internal customers and are thus just as important as the external customers (Akbari et al. 2017). It is central to establish a strategy for how to communicate effectively with the internal customers (Berry 1981 in De Bussy et al. 2003). There are established methods of measuring the needs of the internal customers in relation to an effective internal communication, where the following three components are established concepts: ‘commitment’, ‘trust’, and ‘associability’ (Leana & Van Buren 1999;

De Ridder 2004). Although, practitioners and scholars express the area of internal communication as a crucial and challenging field and some even neglects the importance of a joint vision of the communication and marketing functions (FitzPatrick 2004; Robertson 2004; Welch & Jackson 2007). Welch and Jackson (2007) express a concern that the external and internal communication should no longer be viewed as independent and disconnected research fields without analyzing the intertwining linkages among them.

It is further noted that international corporations have managerial challenges to operate the internal communication effectively, aggravated by a cultural aspect that is inherent in multicultural global organizations today (Johnson 2001;

Prajováa et al. 2016; Sombultawee & Boon-itt 2017). This is tied to the ability to cherish the social capital within an organization in order to grow employee loyalty across functions and departments (De Ridder 2004). Additionally, this is especially important for the internal marketing and communication field, to promote an effective and appreciated internal communication (Akbari et al.

2017). It is apparent that the scholars express that the research field of internal marketing and internal communication is not complete (Welch & Jackson 2007;

Akbari et al. 2017). In the proceeding thesis, it is therefore necessary to further the understanding of internal marketing and internal communication as a joint concept known as internal marketing communication (Welch 2015).

(7)

Problem Discussion

In an increasingly competitive environment, Internal Marketing Communication (IMC) help leverage business performance and provide a strategic guidance to increase profitability (Ahmed & Rafiq 2003). The researchers claim that IMC requires further clarification within a given setting as well as further research. There is a further need for investigating the role of internal marketing (IM) when assessing the strategies of internal communication (IC) (Ahmed & Rafiq 2003). IM is described as an area that is both crucial and challenging for multicultural businesses today (FitzPatrick 2004; Robertson 2004). Most companies misunderstand how the connection of IM strategies enable a more efficient IC (Ahmed & Rafiq 2003; Costa & Jongen 2006;

Jacobsen et al. 2014). Akbari et al. (2017) points out that by focusing on IMC, organizational goals and strategy may be realized through an increased level of employee satisfaction and loyalty. Even if companies manage to integrate a communication strategy and focus on IMC, it is not a straightforward action to increase one's social capital (Ruppel & Harrington 2001; Ahmed & Rafiq 2003).

These relationship-networks are tied to the manager’s responsibility to skillfully imbed corporate core values when trying to create a business culture that promotes good communication and knowledge sharing structures. There is a knowledge gap concerning the comparison of perceptions between managers and non-managers in a multicultural global organization (Jaworski & Kohlin 1993; Palmer and Hartley 2002; Schein 1984 in Welch & Jackson 2007; Jacobsen et al. 2014; Akbari et al. 2017). It includes the perceptual differences of IC at different levels within the organization (Jaworski & Kohlin 1993; Jacobsen et al.

2014; Akbari et al. 2017).

The connection of IM and IC strategies narrows down to how communication is measured. This comprehensive task has not yet seen a sound praxis of measurements within the joint research field. First, Caruana and Calleya (1998) and Grönroos (1994) has connected company IC performance to employee commitment. Second, Akbari et al. (2017) similarly argues that employee commitment and trust will add value to the organization and increase the level of fulfillment of IM goals. Third, the way employees feel associated with the company is tied to the ability of the organization to understand and address their needs of communication (Grunig et al. 1992; Welch & Jackson 2007). It is essential that research should investigate the effectiveness of communication in a more complex cross-organizational structure to fully cover the research gap (Jacobsen et al. 2014). Aiming at an understanding of the underlying challenges

(8)

of IMC will start by examining how the workforce is engaged, through the IC, at Alpha.

Aim

The aim of this study is to examine the effectiveness of internal communication, as a dimension of internal marketing in a global multicultural company.

Delimitations

The study takes a marketing perspective where IM as well as IC are interlinked in the field of IMC. The thesis examines the case of a leading global company within the manufacturing industry, referred to as ‘Alpha’. Alpha has four established business lines, where this thesis will examine the one referred to as the ‘Machinery Business Line’. The specific business line is chosen as the management expressed a concern for the IC, presented further in chapter four.

Contribution of Knowledge

By investigating the needs of the internal customers at a multicultural global company, this thesis will commit to increase the knowledge insight of how organizations, like Alpha, can examine their current internal marketing communication and increase its effectiveness.

(9)

2. THEORETICAL FRAMEWORK

The theoretical framework consists of five sections: Organizational Structure and Culture; Internal Marketing Communication; Internal Corporate Communication; The Conceptual Model; The Hypotheses.

Organizational Structure and Culture

The first section will deal with three subsections that describes the surrounding environment of a multicultural global company. The effectiveness of IC is affected by these circumstances: The hierarchical setting; the business culture and strategic managers.

2.1.1.The Hierarchical Setting

The structure of an organization affects the IC and has to be taken into consideration when studying a multicultural global company (Ruppel &

Harrington 2001; Jacobsen et al. 2014, Men & Stacks 2014). Two aspects stand out when considering the effects that a hierarchical culture has on the IC effectiveness, the level of (1) formalization and (2) centralization (Jacobsen et al. 2014). These structural mechanisms are facilitators of communication.

Formalization is the first element, which emphasizes following procedures and rules, with definite divisions of responsibilities and explicit guidelines (Moenaert et al. 1994; Dasgupta & Gupta 2009). The second element relates to the level of decision-making which is linked to the level of empowerment, positively associated with a better communication flow (Song et al. 1996; Moenaert et al.

2000; Dayan & Basarir 2010).

The marketing department is required to manage the communication flow by following detailed guidelines of how to implement the IC strategy (Costa &

Jongen 2006; Jacobsen et al. 2014). A common tool for managing communication in complex organizations is the use of an intranet to enable better information sharing (Ruppel & Harrington 2001). Ruppel and Harrington (2001) has found a positive relationship between a hierarchical culture and intranet implementation. However, using communicational tools to fit organizational goals may instead serve as the hierarchical organization’s controlling mechanism and hinder an effective IC (Ruppel & Harrington 2001).

Also, it is a common fashion of hierarchical organizations to favor the use of a one-way communication as it is unavoidable when trying to communication with a large sum of people. This is a hinder for effective communication and

(10)

organizations should instead encourage the use of a two-way symmetrical communication as it is more transparent and conceive an organization that is willing to listen and respond to its employees (Grunig 1984; Welch & Jackson 2007). The organizational structure is not necessarily a hinder for an effective communication if the internal customers’ needs are fully addressed (Grönroos 1994; Ahmed & Rafiq 2003; Akbari et al. 2017). This is a strategic matter were the formation of IM strategy will account for the effectiveness of IC (Welch &

Jackson 2007; Men & Stacks 2014; Welch 2015). This further implies that marketing operations should be aligned across organizational borders, which is a difficulty for organizations with a large hierarchical structure (Sombultaweea

& Boon-itt 2017).

2.1.2.Business Culture

Culture and the emergence of subcultures, are factors that affects the strategic decisions and impacts the effectiveness of communication throughout the organization (Handy 1985; Anderson & West 1998; Smidts et al. 2001; Palmer and Hartley 2002; Dackert et al. 2003; Chen & Lin 2004; Schein 1984 in Welch

& Jackson 2007). Depending on the ability to fully integrate corporate policies, values and culture, employees will have different relationships to the organization (Ruppel & Harrington 2001; Mphidi & Snyman 2004; Akbari et al.

2017). The level of integration is tied to the implementation of the business culture which is a matter promoting knowledge sharing and making the IC more effective (Ruppel & Harrington 2001; Ahmed & Rafiq 2003).

2.1.3.Strategic Managers

The strategic managers responsibility expands through the business culture and manages the IC with all internal customers at all levels within an organization (Welch & Jackson 2007; Men & Stacks 2014). It consists of a cross-functional process that resembles a continuum ranging from informal (non-task-related) and formal (task-related) communication (Ahmed & Rafiq 2003; Welch &

Jackson 2007; Men & Stacks 2014). Researchers confirm that by aligning the perception of goals and values, for the internal customers and the organization respectively, management is assured to be effective in reaching the intended IM goals (Narteh & Odoom 2015; Akbari et al. 2017). These strategic managerial actions and systems should be the consequence of the organization’s IM strategy and plan in order to be implement communication more effectively (Ruppel &

Harrington 2001; Ahmed & Rafiq 2003; Akbari et al. 2017). The ability of implementing the IM process effectively has been proved to be positively related

(11)

to a higher level of employee loyalty and social capital (Fulford & Enz 1995;

Ruppel & Harrington 2001; Ahmed & Rafiq 2003; Christen et al. 2006; Tsai &

Huang 2008; Awwad & Agti 2011).

Internal Marketing Communication

Marketing communication is defined as both a tool for external marketing efforts as well as marketing for internal purposes, where the focus is on marketing principles and the process of value creation (Mphidi & Snyman 2004;

Prajováa et al. 2016). IMC reassembles a complex system of identifying internal customer signals, managing effective internal relationships accordingly, and maintaining that external communication as well (Mphidi & Snyman 2004;

Prajováa et al. 2016). When evaluating corporate communication strategy there are indeed many flaws of the communication quality that springs from a disoriented system (Mphidi & Snyman 2004). Welch (2015) stresses the importance of management focus on IC within the marketing framework, to be able to address the needs of the internal customers and enable the creation of an effective communication.

2.2.1.Internal Marketing

IM is a branch from marketing communication, which is oriented towards the internal customers, and therefore it is guiding departments, functions and every single employee, by their continuous needs (Ahmed & Rafiq 2003; Akbari et al.

2017; Sombultawee & Boon-itt 2017). Analyzing a company’s IM will also create an understanding of structures and politics which may hinder company performance internally as well as externally (Grönroos 1994; Ahmed & Rafiq 2003; Cheney & Christensen 2001 in Welch & Jackson 2007). IM intends to clarify the organization’s understanding of its own internal environment, which can lead to better adaptation of the surrounding culture (Welch & Jackson 2007). The mentioned above should be a result of the implemented IM strategy which is a key factor of organizational effectiveness where managers play a key role in reaching IM goals by adopting an efficient communication (Ahmed &

Rafiq 2003; Akbari et al. 2017).

2.2.2.Internal Communication

Within the IM strategies toolbox, IC is a tool for managers when establishing effective communication with the entire workforce (Johnson 2001; Men &

Stacks 2014; Yeoman 2006 in Tench & Yeomans 2017). There is a positive

(12)

2007; Welch 2015) in which implementation of a supportive attitude lead to a more effective communication (Narteh & Odoom 2015; Akbari et al. 2017).

Berger (2008) defines IC as a central process where employees participate in information sharing to make meaning and establish organizational values and cultures. This is a way of creating a supportive attitude towards the organization, which is measured through three components of social capital and employee loyalty, these are: ‘commitment’, ‘trust’ and ‘associability’ (Mowday et al. 1982;

Leana & Van Buren 1999; De Ridder 2004; Jo & Shim 2005). These components are parts of the Internal Corporate Communication (ICC) plan, included in the overall IM strategy, and act as a guidance for the strategic managers, formed to oversee the business culture and influence the internal customer perspectives about the organization (Welch & Jackson 2007).

Internal Corporate Communication

ICC encompass the communication with all employees and in that sense, managers should commit to building employee loyalty and social capital through a continuous and consistent communication (Welch & Jackson 2007). This is achieved through a consistent, clear and continuous communication by top managers and strategic managers (Welch & Jackson 2007). On the contrary, ineffective communication will hinder the development of these goals through a use of incorrect strategies (Etgar 1979). The three interrelated goals of ICC are: (1) commitment, (2) trust, (3) associability (Welch & Jackson 2007). It is possible to influence the employee’s perception of the company though these components (Welch & Jackson 2007). Separating the terms of ‘commitment’

and ‘trust’, De Ridder (2004) differs between information types in terms of quality communication. The type of information is either directly, or not directly related, to the task, the former being an indicator of ‘commitment’ while the latter serves as an indicator of ‘trust’ (De Ridder 2004). Downs and Adrian (1997) also advise that both ‘commitment’ and ‘trust’ should be applied as performance indicators for IC.

2.3.1. Commitment

De Ridder (2004) describes ‘commitment’ in terms of individual involvement with the company and as a positive attitude amidst the employees driven by their motivation for the organization. The research of ‘commitment’ concentrates on managerial communication as a daily basis (Allen & Meyer 1996; De Ridder 2004) and concludes that the accuracy and relevance of communication is key to promote ‘commitment’ (Welsh & LaVan 1981; Trombetta & Rogers 1988).

(13)

Allen and Meyer (1996) has established the prominent technique and a questionnaire for measuring ‘commitment’ which has been validated and is used as a common praxis (Alam 2011). The perception of the quality of task-related information will be high if internal customers experiences the information as supportive for their task performance (De Ridder 2004). It is of great importance for the task-related communication that the instructions and feedback of the daily work is accurate for the communication to be effective (Guest & Conway 2002).

2.3.2. Trust

Communicational activities that relates to company openness, such as information concerning organizational policy, is directly related to the concept of ‘trust’ (Butler 1991; Ruppel & Harrington 2001; De Ridder 2004; Du Preez et al. 2017). When having policies in place, managers also need to consider the quality of the information (Kreijns & Kirschner 2004). This may be achieved by adjusting and separating the information into different types to better suit the organizational goals as well as the needs of its employees. The needs may for example be different when it comes to the amount of information received as different tasks require different content (De Ridder 2004; Kreijns & Kirschner 2004). When it comes to ‘trust, it is important that the employees get the right quantity and quality of the non-task-related information.

2.3.3. Associability

Associability is defined in terms of individual perception of belonging to the organization and an attitude of support (Smidts et al. 2001; Welch & Jackson 2007). It is the ability and willingness of each employee to break down the collective goals and adopt them collectively (Leana & Van Buren 1999). This

‘we’ feeling will enable employees to identify oneself with the organization and these social needs of communal identities are found on subgroup and corporate levels throughout the organization (Cornelissen 2004; Welch & Jackson 2007).

The ICC play a role in developing associability in terms of communicating the organizational objectives and to develop an understanding of the changes within the company (Grunig et al. 1992; Smidts et al. 2001; Welch & Jackson 2007).

(14)

The Conceptual Model

The three above theoretical sections were: organizational structure and culture;

internal marketing communication; and internal corporate communication, which together are the foundation of the communication process in large organizations like Alpha. The first part of the IMC model (Figure I), outlines the different strategic components that work together to create a strategic direction for the organization. It is vital that all of these components play a key role when establishing the IM strategy. First, it is the organizational structure and culture that pave the way for the directives of how the strategy should be created. Secondly, the IM strategy should be a guidance for how the needs of the internal customers should be cared for. Thirdly, the IC is a central process when it comes to accurately share information throughout the organization, articulated in the ICC plan. IM and IC are components of the IMC and align cross-functional processes, which are important as a means of enhancing the company effectiveness. Through the alignment of organizational goals and values, effective communication is created by the means of the strategic managers and their dedication to implement the strategic directions of the organization.

The social capital and employee loyalty are two factors of an effective IC from which the strategic managers should focus their attention on employee

‘commitment’, ‘trust’ and ‘associability’. This is the reason why the arrow of strategic direction leads into a funnel that will shape the outcome of the communication in Figure I below. The outcome of the three components is also affected by a possible difference of how the IC is perceived among the internal customers. It is likely that a large workforce will assimilate different needs as there are differences in tasks, such as the differences between a ‘white’ or ‘blue’

collar employee. ‘’White’ collar employees are officials with more administrative tasks than ‘blue’ collar employees, which instead are referred to as workers. In the model, these employees are illustrated by two cogwheels as their needs and perceptions of communication may differ and need to be considered for an effective IC. The different needs within the workforce may also be a consequence of a difference in nationality, or other prominent subcultures of the organization, indicated in the model by the underlying layer of the cogwheels. The three mentioned components of IC are substantial measurements of effective communication and will be used further to deepen the insights of weather the communication is effective or not (Allen & Meyer 1996; Leana & Van Buren 1999; De Ridder 2004; Welch & Jackson 2007).

(15)

Figure I. Internal Marketing Communication Model

Nationality

Effective Communication

(16)

Hypotheses

The hypotheses are used to support the aim of the study:

The aim of this study is to examine the effectiveness of internal communication, as a dimension of internal marketing in a global multicultural company.

Many researchers have measured the effectiveness of IC before this study, such as De Ridder (2004), Welch and Jackson (2007) and Akbari et al. (2017).

Effective communication is seen as a dimension of IM and motivates the use of hypotheses that are tied to the components of IC: ‘commitment’, ‘trust’ and

‘associability’. The null hypothesis is formulated first, followed by three alternative hypotheses.

Null Hypothesis

The internal communication is effective as there are no clear differences between how employees perceive the communication, shown by a significant level of employee commitment, trust and associability.

The null hypothesis refers to when the IC at a global multicultural company is fully effective and working just as intended to. There is not much to improve if there is proof for that the null hypothesis should be accepted. Instead, the following hypotheses should try to indicate why or how the null hypothesis can be rejected. If there is support for any of the following alternative hypotheses, there is evidence that the null hypothesis is to some extent not valid.

Hypothesis 1: Commitment

There are differences between how employees perceive the quality of task-related communication.

Hypothesis 2: Trust

There are differences between how employees perceive the quality of non-task-related communication.

Hypothesis 3: Associability

There are differences between how employees perceive the organization’s strategic direction.

‘Commitment’ is according to De Ridder (2004) related to the kind of IC that directly concerns the daily work of the employee while the more general non- task-related information is applicable to ‘trust’. These two components may vary in quality as employees have different needs regarding different types of information, described further in the next section. Associability is connected to whether the business culture is fully established or not and is seen as a kind of acceptance of the organization’s strategic direction (Welch & Jackson 2007).

(17)

3. METHOD

The methodological chapter has six sections: Methodological approach; Study Design; Data Collection, Data Analysis, Reliability and Validity, Ethical and Critical Considerations.

Methodological Approach

This thesis is based on the lack of knowledge of the differences between employees concerning their perceived effect of IC in a complex organization structure. Since this is a strategic matter it covers the research field of IMC and is examined by conducting a cross-organizational investigation.

The thesis takes a naturalistic approach, which according to David and Sutton (2016) is a method built on positive epistemology where the focus lay on collecting data to get objective knowledge. The positivistic approach leads this study to a desire to identify whether a theory is right or wrong (David & Sutton 2016). Therefore, the thesis has a deductive approach were the study will test the hypotheses through a quantitative method. In these terms, the empirical study aspires to reject or accept the determined hypotheses. The majority of the theoretical framework is based on articles that concentrate on IC within the fields of marketing, and secondly the field of management. Many of these articles use a quantitative approach, where hypotheses are formulated together with a description of the components used as measures. The study keeps an objective approach through carefully formulating any statements or questions in the survey by following the established praxis and previous research within the field of IMC. Hypotheses are predictions of the direction of the results based on the relationships between variables (Ross 1998), as are they in this thesis.

Study Design

Denscombe (2016) describes the usage of surveys as a way to collect data to test a specific theory. The thesis has used a web-based survey as it aligns with Denscombe’s (2016) arguments for the advantages of a big geographic distribution possibility and the flexibility of the data collection. The selected population is spread throughout six different countries which complicates the handling, as well as the distribution, of the surveys. The usage of web-based surveys reduces the risk of human error through automatizing the data collection (Denscombe 2016). The program for the web-based survey is evaluated through; (1) available item types, (2) ability to export data, (3) ability

(18)

As a result, the system ‘Survey & Report’ was chosen. A random sample of participants are selected from the list of employees at Alpha. A random sample is described by Denscombe (2016) as beneficial when there is a complete access to the entire population of company files of employees.

Data Collection

The entire population of employees at Alpha Machinery is 2831 in total, see the spread in Table 1 below. The population is spread across eleven different countries, where the following six countries are included in the study as they represent the majority of 2831 employees; Finland, China, Sweden, Italy, the US, and Canada. The other five countries had too few employees, in total 14 employees, and were excluded in the Machinery research. The survey was professionally translated into 5 languages to fit the needs of the population at Alpha: Finnish, Chinese, Swedish, Italian, and English. The English version is included in the Attachment section. A closed population is defined by membership in terms of employment (Schonlau 2002) and is applicable in the case of Alpha. As communication differs between levels within the hierarchical structure, the closed population is further categorized into two subgroups depending on the collar color, which represents the level in the company structure. Employees are either a ‘white’-category with more managerial and administrative tasks, or a ‘blue’-category and thus more focused on production.

The study was conducted from January until June 2019, and the survey was released during three weeks in April.

According to Denscombe (2016) a probability sampling is based on the population. In the case of Alpha this population consists of the 2817 employees in the six countries. A stratified sample is characterized by the possibility to divide the data into further variables (David & Sutton 2016) and was thus used in this study. In the case of Alpha, the list of employees was separated by means of belonging to a ‘category’ and a ‘country’, see Table 1. The goal was to reach at least 20 respondents in all groups, as Mendenhall and Sincich (2013) argue that it is the lowest amount needed to be able to compare differences between groups statistically and generate generalizable results for the whole population.

However, to be able to reach that goal, the survey was distributed to 50 randomly sampled respondents in each group, and later the number was increased by another random sample as there were too few responses within some groups. As some groups were smaller to begin with, and other groups did not reach the enough amount of surveys, there are differences in sizes in

(19)

especially the ‘blue’-category groups. The sizes of each group are shown in Table 1, where the sampled size is shown together with the total number of employees in that group (n). The results will be comparable with groups that exceeds 20, if there are more respondents in the results will instead be more precise in those cases. From the table below the total sampling is indicated by country and category. The response rate table is instead included in the findings-chapter.

Data Analysis

A population can be divided into sub-groups in order to enable comparisons between groups (Bryman & Bell 2013). To take these implications into consideration, the first priority is to examine each category independently in each country where there are enough respondents for comparison. If there is not enough data, the second priority is to combine the two categories and instead examine differences across national borders.

The questions are mainly ordinal, and the nominal questions are questions with both interval and scale data. Denscombe (2016) explains that the formation of the survey in terms of ethical dilemmas, such as the perception of sensible questions, may not be answered with full honesty. Another common error made in surveys is limiting the respondent by giving a set of coded answers which can create an underlying frustration (Denscombe 2016). It is avoided to some degree in the study by not coding all answers as well as including a comment section

(20)

where the respondent can write other answers and thoughts. Further, the study avoids questions about organizational level of the individual and by an anonymity disclosure. The questions were then included in three subsequent pilot studies, where respondents and managers from Sweden and Finland gave feedback to improve the survey. A pilot study is a way to ensure that the questions are understood but also as to ensure the adequacy and the structure of the survey (David & Sutton 2016).

When using a Likert scale, such as 1-5, the data is scaled ordinal (Denscombe 2016). This was the case of the Alpha study and later on, these values were re- coded and adjusted to each of the hypotheses. Primarily, the scale represented the level of agreement, where (1) agree and (5) disagree. Secondarily, the scale in other cases represented (1) too little and (5) too much. There were three ways that the variables were re-coded. First, it was interesting to see which respondents that were located in the agreeing-end of the scale, which is value 1 and 2. Second, values 4 and 5 were also separated from the remaining values to locate the disagreeing end of the spectrum. These two sets of values are referred to as outermost values in the findings-section. In these two scenarios, the variables were re-coded so that the outermost values, either 1 and 2 or 4 and 5, were labeled 1 and the remaining variables was re-coded to 0. Third, it was relevant to look at the value 3, to compare the perceived amount by the respondents and locate those who received either a deficit (values 1 and 2) or experienced a surplus (values 4 and 5) of information. For this study, the value

‘3’ concerns questions that regard the amount of information, and ‘3’ is thus interpreted as enough of information perceived by the respondents. This is the foundation of re-coding the variables, further outlined in the findings-chapter.

The null hypothesis is built on the assumption of being correct with alternative hypotheses that questions that assumption (Denscombe 2016). The analytic method of analysis of variance (ANOVA) is used to test the hypotheses and the variation between the categories of ‘white’ and ‘blue’ employees, as well as the differences between countries on a 95% significance-level. To complement the analysis of independent variables, two separate indexes are used for hypothesis I and III. The usage of index facilitates the handling of many variables to help summarize the findings (Mendenhall & Sincich 2013). The use of index is common when there are multiple variables that are used to explain a common cause.

(21)

Reliability and Validity

Reliability specifies how trustworthy the study is in terms of achieving the same results if repeated (David & Sutton 2016). Since this study has access to the entire population, a future study with samples from the same population should give a similar result. Although, it is possible that results will change due to a change of the population as hired employees are not a constant number. If the study was to be conducted many years from now, a different result may be achieved due to changes in the population structure as well as a change in the business culture or as an effect of improved IMC. Another aspect of the reliability is to have a representative sampling (David & Sutton 2016). Following the arguments of the data collection section, the selection of participants is in accordance with the statistical literature methods and thus assures the reliability for this study.

Validity concerns the relevance of how results of the study corresponds with the reality (David & Sutton 2016). In other words, it is a measure of how well the theoretical components assort with the intended conclusions. The internal validity is based on how reliable the data is and if it reflects the reality of the individuals whilst the external expresses the connection between the sample and the whole population, the generalizability (David & Sutton 2016). The internal validity is accounted for in terms of translating the survey into the respective languages. Professional translators were hired to translate the English version into Chinese, Finnish and Italian. Still, there could be issues for the respondents to understand questions, and in China for example, there are several languages and cultural differences within the group. The English and Swedish versions were included in three pilot studies, reviewed by three supervisors, as well as examined by the management of marketing and IC at Alpha.

The external validity is strengthened by formulating questions similar to those in previous research (David & Sutton 2016). The articles of the study were validated in the sense that they were being published in respected journal. There was also a review-article on measurement praxis of the components used in this study, which further strengthens the relevance of the questions in the survey.

The combination of previously used questions from different articles together with questions formulated based on the aim of this study increases the validity.

A weakening of the validity may be an effect of the authors limited experience.

(22)

Ethical and Critical Considerations

To ensure a good ethical level of the survey, the confidentiality of the respondents was guaranteed, with their consent as a criterion to participate in the study. Confidentiality and consent are important for the ethical practice (David & Sutton 2016). There might still be a pressure to participate due to the fact that this an in-house research on behalf of the respondents’ managers. Also, the result might affect the respondents and therefore have an ethical implication that is difficult to overcome. The survey is further in accordance with European law and the GDPR standards for the collection of personal data. The respondent was not asked to indicate which category or country they belong to, instead the authors separated these groups prior to distributing the survey. The anonymity of the data is acknowledged by numerating the participants instead of using individual names. Also, the gathered personal data will not be disclosed for anyone but the authors of the thesis.

The further critical aspects include the translation of the survey from English to the original language of the respondents. Even though the questions will be translated and thus assimilate the same content, there is still a possibility of misunderstandings. For example, managers in China have indicated that there will be differences in perceiving the Chinese language different depending on city and region. The reason for translating is to create comparable answers of the survey, independent on which group one belongs to as there is no language applicable to all (Berry 1981 in De Bussy et al. 2003). McGorry (2000) expresses the concern to consider cultural differences in relation to translation. The importance of cultural differences is to some degree accounted for in terms of a including a cultural aspect in the thesis.

When reflecting upon the use of sources, it should be noted that the majority of the theoretical framework consists of marketing and management sourcing and with a complementary set of articles from other related fields. The majority of the references are peer reviewed articles and published in respected journals within their field. There is also a criticism of sources, as there are four cases of a use of secondary sourcing. It was a necessity to use these secondary sources as the original source could not be found due to an inadequate referencing in the articles. These articles were all peer reviewed and found in trustworthy journals. They are correctly referenced in the theoretical sections by indicating what article the secondary source is located in.

(23)

4. FINDINGS

This chapter has four sections, Alpha is first presented, followed by the response rate. The last two sections present the statistical findings.

Presenting Alpha

Alpha is the leading global producer of manufacturing products and services, in the machinery industry. In 2017, Alpha had a net sale of 3.1 billion Euro and has currently more than 12 000 employees positioned in 30 different countries.

Alpha is a multicultural global company and naturally faces challenges of communication. There are expressed issues of IC throughout the organization.

Further, Alpha is a hierarchical organization where the global and national information takes a top-down direction that assembles in a one-way communication. The thesis will focus on the Machinery business line, primarily located in Sweden and in Finland.¹

The Senior Manager of Marketing and Communication (SMMC) of the machinery business line, express Alpha’s communication challenges as an overflow of information from various internal channels, outlined in the block quote below. According to the SMMC, the employees at Alpha have diverged from the current platform of communication, their intranet, and instead adapted their own channels of communication through many different uncoordinated channels. SMMC further explains that there are numerous initiatives internally at Alpha that compete to get employees attention, a result of a problematic and insufficient coordination of information. The communication is spread through a diverse set of channels, creating an information overflow for the employees.

The intention is to separate what information is being spread by using different channels in an attempt to meet the different needs of their employees.²

Communication Challenges

To reach out through the information flow

Many “big” initiatives competing about attention at the same time To create a sense of urgency and call to action

To define actions that really makes a difference in behavior and mindset To be relevant and inspiring

To respect people’s time and workload

To coordinate and prioritize communication activities from different levels.³

1 Senior Manager of Marketing and Communication, (SMMC) at Alpha Machinery, 2019-02-06.

2 Ibid.

(24)

Alpha has categorized different types of information as either ‘nice’, ‘good’,

‘need’ or ‘unnecessary’ to try to differentiate between the qualities of communication. In the survey questions the employees are asked to classify the amount they get of the different types of information. The types are re-labeled and described to create a better understanding for the employees as they are not used to seeing the information categorized according to quality. All questions are included as attachments to this thesis in English.

Response Rate

The sample population of Alpha has a total population of 2831 of which a sample was summed to 564 respondents. Of these sampled respondents, there were 215 completed surveys which is outlined by the response rate in the table below:

The table above separates the respondents by country as well as by category. To illustrate an example, the Finnish group has a total of 198 respondents of which 49 completed the survey. There were 50 respondents in the ‘white’-category group for Finland, of which 29 completed the survey. Only two countries have enough (x ≥ 20) respondents in each category to perform comparisons between the categories; Finland and Sweden. The other countries have a total population per country that exceeds x ≥ 20 which means that they are comparable between countries but not by categories. An exception is Canada which has 16 respondents out of 30 which assembles a response rate of 53.33% of the entire population in Canada. This large sample from of the entire population presumed to be a sufficient amount for trustworthy sample and is thus included in the study.

The lack of respondent answers occurred due to a lack of email addresses of many of the ‘blue’-category employees. Their main communication channel is verbal with the shop foreman, production manager and in some cases, they have

(25)

hall meetings. The ‘blue’-category therefore have limited access to the communication channels, in comparison with the ‘white’-category group.

Statistical Findings by Category and Country

This section presents the findings of Finland and Sweden as they are the only countries with enough respondents in both categories of employees. The thesis uses a significant level of ∝=0.05. There were a few ‘marginally significant’

results significant, 0.05<∝<0.1, which are included out of interest, not for their significance. The insignificant values with a higher p-value, above ∝=0.1 indicates that there are not differences between the categories other than what chance can account for, at a 95%-degree of certainty. The insignificant variables will not be analyzed further in this section, with a respect to both category and group, instead the subsequent section will analyze the differences by country and excluding the category-variable.

As the data has been re-coded, the Likert Scale from 1-5 has been organized into three distinguishable groups and therefore the mean values in the following section do not represent the mean value of the five Likert scale values. It should be noted that the mean values below represent groups of values, indicated and explained for each of the results below as well as described further in the Data Analysis section 3.4.

4.3.1.Hypothesis I

For the first hypothesis, only question three had significant results which are presented below. The respondents used a scale to indicate the amount of information that they received, where the two outermost values were (1) too little and (5) too much.

Question 3: Estimate the amount of information that you receive that directly concerns your daily work.

The compared differences between the employees concerns the information that was directly related to their daily work, also called task-related information.

From the scale above, the term ‘enough’ (3) was analyzed in comparison with the other terms (1, 2, 4, 5). Table 2 summarizes the results by an ANOVA for the significant variable of both Sweden and Finland. Of the four types of information, the ‘nice’ type was the only significant variable. The complete table for all variables is found in the Appendix (Table I-II).

(26)

Table 2: Question 3. Enough of ‘nice’ task-related information

0=values 1, 2, 4 and 5 3=value 3

The result of the ANOVA display that for one of the four types of information, there is a significant difference between how the two categories of employees perceive the amount information at a 5%-significance level. The information type ‘nice’ is significant for Sweden F(1,55)=4.71 and p=0.034 and for Finland F(1,47)=4.260 and p=0.045. The results suggest that the differences by category is explained by a difference in mean values, visualized in the means plots below and enlarged in the Appendix (Figure II-III).

Figure II and III: Question 3. Task-related, ‘nice’-information

0=values 1, 2, 4 and 5 0=values 1, 2, 4 and 5

3=value 3 (Enough information) 3=value 3 (Enough information)

The y-axis ranges from 0 to 3, where a value closer to 3 indicates that the respondents perceive the amount of information as enough. The value 2.4 is a point of reference indicating that if the mean value is below this value, the average employee does not anticipate the amount of ‘nice’-information as enough. Instead they either get a deficit or surplus of information. The means plots above exhibit a similar pattern for both countries, where the ‘white’- category have higher mean values compared to the ‘blue’-category. The ‘white’- category values are 1.45 in Finland and 1.37 in Sweden and the ‘blue’-category

(27)

values are 0.6 in Finland and 0.55 in Sweden. The higher values of the ‘white’- categories are not really high in the sense that they are close to 3, but rather high in comparison to the ‘blue’ responses. This is the difference by category that is not explained by chance. It is also noticeable that even though there are differences between the mutual categories, they are similar when comparing by country.

4.3.2. Hypothesis II

For the second hypothesis, question four and twelve had some significant results which are presented below. The respondents used a scale to indicate the amount of information that they received, where the two outermost values were (1) too little and (5) too much.

Question 4a - Enough information: Estimate the amount of information that you receive that not directly concerns your daily work.

Out of the four types of information, the ‘need’ and ‘good’ type had significant results where the categories had perceptual differences when it came to the amount of non-task-related information received in Sweden. To compare the differences between categories of employees, the data was coded so that the term ‘enough’ was analyzed in comparison to the other terms. Table 3 summarizes the results by an ANOVA for the ‘need’ and ‘good’ type of information. The complete table for all variables is found in the Appendix (Table III-IV).

Table 3: Question 4a. Enough of ‘need’ and ‘good’ non-task-related information

0=values 1, 2, 4 and 5 3=value 3

The result of the ANOVA above display that there is a significant difference between the perceptions of the two categories of the Swedish employees. The

‘need’-type of information is significant F(1,55)=9.561 and p=0.003 and marginally significant for ‘good’ F(1,55)=3.462 and p=0.068. The significant

(28)

levels of the variables in Sweden can be explained by differences in the category mean values, illustrated below with enlarged figures in the Appendix (Figure IV- V).

Figure IV-V: Question 4a. Non-task-related ‘need’- and ‘good’-information

0=values 1, 2, 4 and 5 0=values 1, 2, 4 and 5

3=value 3 (Enough information) 3=value 3 (Enough information)

The y-axis ranges from 0 to 3, where a value closer to 3 indicates that the respondents perceive the amount of information as enough. The horizontal line represents the value 2.4 and is a point of reference that indicate that if the mean value is below this value, the average employee does not anticipate the amount of ‘need’- and ‘good’-information as enough. There are similarities in the pattern, ‘white’-categories have higher mean values than the ‘blue’-group. The

‘white’-category group represent the respondents that got enough of the information that they needed that does not concern their daily work. The

‘white’-category has a higher mean value in both variables, ‘need’=2.66 and

‘good’=1.97. While the ‘blue’-category variables have ‘need’=1.64 and

‘good’=1.23. The results are significant and therefore the differences in mean values cannot be explained by chance.

Question 4b - Outermost values of ‘need’-information: Estimate the amount of information that you receive that not directly concerns your daily work.

As a development from the question above, the ‘need’-information variable was selected for further analysis and coded to see if the respondents experienced the amount of non-task-related information as too little, or partly too little.

Therefore, the values (1) and (2) was labeled ‘1’ and the values (3), (4), (5) were labeled ‘0’. This resulted in a new variable for the Swedish categories, named Q4_1s2s_nwneed, the results are summarized in the table below. The complete table for all variables is found in the Appendix (Table V-VI).

(29)

Table 4: Question 4b. Non-task-related, ‘need’-information

0=values 3, 4 and 5 1=values 1 and 2

At a 95%-level of significance, the result of the ANOVA above display that there is a significant difference between the perception of employees in the two categories in Sweden, F(1,55)=7.604 and p=0.008. This difference can be explained by the way employees perceive the amount of non-task-related information. The difference in mean values are illustrated in the means plots below, enlarged in the Appendix (Figure VI).

Figure VI: Question 4b. Non-task-related ‘need’-information

0=values 3, 4 and 5

1=values 1 and 2 (too little, partly too little)

The y-axis ranges from 0 to 1, where a value closer to 1 indicates that the respondents perceive a deficit of information. The value 0.6 is a point of reference that indicate that if the mean value is below this value, the average employee does not anticipate the amount of ‘need’-information a deficit. The

‘white’-category mean equals 0.236, while the ‘blue’-category mean equals 0.477.

(30)

Question 12: The following statements represent core values, which four do you align with Alpha?

The respondents were given seven statements of which they were instructed to choose the four that they align with Alpha´s core values:

1. We move our customers' performance forward

2. We know it takes people with different ideas, strengths, interests, and cultural backgrounds to make our company succeed

3. We promote new ideas to create the future

4. We support independent contributions to deliver collaborate results 5. We improve every day to deliver results

6. We work together to make a difference

7. We strive to minimize our negative impact on the environment There were four correct answers (1, 3, 5, 6), three that were incorrect (2, 4, 7).

The correct answers were coded ‘1’, whereas the incorrect answers were coded

‘0’. Table 5 below summarizes the significant results by an ANOVA for the correct answers in Finland. The descriptive table and complete ANOVA is found in the Appendix (Table VII-VIII).

Table 5: Question 12. Correct answers of core values

0=values 2, 4, 7 1=values 1, 3, 5, 6

The result of the ANOVA above display that there is a significant difference between the two categories of employees in Finland, F(1,47)=10.097 and p=0.003. In the Appendix (Table VII) it is noteworthy that the standard deviation for the ‘blue’ group equals 0.00 which could imply that the results of the two categories are not comparable. However, the explanation is found by studying the answers of the ‘blue’-category, as none of the employees managed to identify all of the correct core values of Alpha. The means plots below illustrates these differences further, enlarged in the Appendix (Figure VII)

(31)

Figure VII: Question 12. Correct answers of core values

0=values 2, 4, 7

1=values 1, 3, 5, 6 (Correct answers)

The y-axis ranges from 0 to 1, where a value closer to 1 indicates that the respondents had correct answers. The value 0.43 is a point of reference, as it represents 3/7 of the scale, that indicate that if the mean value is below this value, the average employee cannot identify the core values at Alpha. The variable has a difference in mean values for ‘white’ 0.34 and ‘blue’ 0.00 which indicates the effect of an external factor. The means plots show that both categories in Finland are below the point of reference which means that they have not fully implemented the core values.

4.3.3. Hypothesis III

For the third hypothesis, question seven and ten had significant results which are presented below. The respondents used a scale, varying from (1) agree to (5) disagree, when relating to the statements.

Question 7: I think that we use too many communication channels.

To analyze the respondents’ perception of the variety of communication channels the data was coded so that ‘1’=Agree and partly agree, and that ‘0’=the remaining three values where the respondents were not in agreement. The results were significant in Sweden which is summarized in the table below by an ANOVA. The complete set of tables for all variables is found in the Appendix (Table IX-X).

(32)

Table 6: Question 7. Outermost values for the strategic direction of IC channels

At a 5%-level of significance, the result of the ANOVA above display that there is a significant difference between the two categories of employees in Sweden, F(1,55)=3.979 and p=0.051. The difference between the categories is visualized by the means plots below, enlarged in the Appendix (Figure VIII).

Figure VIII: Question 7. Outermost values for strategic direction of IC channels

0=values 3, 4 and 5

1=values 1 and 2 (agree, partly agree)

The y-axis ranges from 0 to 1, where a value closer to 1 indicates that the respondents perceive the amount of communication channels as a surplus. The value 0.6 is a point of reference that indicate that if the mean value is above this value, the average employee anticipates the amount of communication channels as a surplus. The ‘white’-category mean equals 0.71 and the ‘blue’ mean equals 0.45. This indicates that the ‘white’-categories of employees perceive an overflow of communication channels and that the difference in means can be explained by other factors than chance.

(33)

Question 10: I feel a strong sense of belonging to Alpha.

To differentiate between the respondents who perceived themselves as belonging to Alpha and those who do not, the values (1) and (2) was coded to

‘1’ and the other values into ‘0’. Table 7 summarizes the results by an ANOVA for the respondents who has a sense of belonging to Alpha, and those who do not. The complete set of tables for all variables is found in the Appendix (Table XI-XII).

Table 7: Question 10. Outermost values of belonging to Alpha

At a 95%-degree of certainty, the result of the ANOVA above display that there is significant difference between the two categories of employees in Finland, F(1,47)=3.892 and p=0.054. The differences between the two categories is outlined in the means plots below, enlarged in the Appendix (Figure IX).

Figure IX: Question 10. Outermost values of belonging to Alpha

0=values 3, 4 and 5

1=values 1 and 2 (agree, partly agree)

The y-axis ranges from 0 to 1, where a value closer to 1 indicates that the respondents agree with the statement. The value 0.6 is a point of reference that indicate that employees above the line feel a belonging to Alpha. The ‘white’-

(34)

category means equal 0.72 which indicates that the category perceive a sense of belonging whilst the ‘blue’ do not, with a mean that equals 0.45. This shows that the difference between the groups is based on other factors than chance.

Statistical Findings by Country

This section presents the variables that were not significant when comparing categories of ‘white’ and ‘blue’ employees. Instead, this section focusses on the differences among the respondents by country. In this section, there will be a slight variation in the degrees of freedom (df) which is explained by the fact that some questions have a few answers missing. This does not significantly affect the overall results as there are enough answers already. It should be noted that the mean values below represent re-coded values, indicated and explained for each of the results below as well as described further in the Data Analysis section 3.4.

4.4.1. Hypothesis I

Question three is presented separately and the remaining questions, associated with hypothesis I, are collectively presented as an index.

Question 3: Estimate the amount of information that you receive that directly concerns your daily work.

The question had an interval of 1-5 to indicate the how the respondents perceived the amount of information, varying from (1) too little to (5) too much.

When comparing results by country, instead of categories as above, the results are significant for two types of information. These two types are ‘need’ and

‘unnecessary’ and relates to the perception of receiving ‘enough’ task-related- information, see the complete ANOVA in the Appendix (Table XIII-XIV).

These results are compared further by the following two ANOVAs which will display the two outermost values for each type, see tables below. The complete set of tables and figures for all variables is found in the Appendix (Table XV- XVIII, Figure X-XI).

(35)

Table 8a: Question 3. Task-related, ‘need’-information

Table 8b: Question 3. Task-related, ‘unnecessary’-information

The tables above display that the respondents either recognize that they get too little information that they need or that they get too much of information that they find as unnecessary. The result of the ANOVA indicates that there is a significant difference between how employees in the six countries perceive the amount information’. The information type ‘need’ is significant F(5,200)=7.423 and p=0.000 and the type ‘unnecessary’ is significant at F(5,194)=4.428 and p=0.001. Having a result that is significant by p=0.000 means that the results are highly significance and very unlikely to have occurred by chance alone. This is not necessarily an absolute value, but rather a result of a rounding error with a value lower than 0.000.

In the Appendix (Table XVII-XVIII), multiple comparisons are carried out to examine the differences by country. These results suggest that the highest difference exist between Finland and China, at a significance of 0.000 for both the ‘need’ and the ‘unnecessary’-information. Also, China has significant differences with all other countries, ranging between 0.000-0.017 (apart from the U.S. for the ‘need’-information). These differences are further described in the means plots in figure X-XI below, also found in the Appendix (Figure X- XI).

Internal Marketing Communication

Hanna Sloberg Sara Nilsson