Thesis – 67 pages – 17.919 words – July 2018
The effect of the franchisee’s
degree of local adaptation on its
business unit performance:
Enhancing great success or
terrible failure?
ROBIN VAN MILLIGEN
Oosterhamrikkade 27A
9713 KA, Groningen
robinvanmilligen@hotmail.com
S3204677
UNIVERSITY OF GRONINGEN
Faculty of Economics and Business
MSc Small Business & Entrepreneurship
Supervisor: Dr. E.P.M. Croonen
1
ABSTRACT
Finding the right balance between standardization and local adaptation is an ongoing and important challenge in the management and entrepreneurial literature, especially in the franchising context. Although this dilemma is quite common in the literature, there is still little empirical evidence that points in the direction of the right balance, especially not in the franchising domain. Therefore, this research focuses on finding the right balance in a franchising context. Since business performance is found to be “the time test of any strategy” (Venkataraman & Ramanujam, 1986, p. 802), taking business performance as “assessment criteria” for finding the right balance is logical and often used in other contexts. This means that the relation between local adaptation and business unit performance is the main interest of this research. An important contribution made to the literature is finding a measure for local adaptation that fits the franchising context. Moreover, this research is one of the first that takes specifically the perspective of the franchisee in consideration, what implies that the measure for local adaptation has to fit the unit level perspective instead of the more often used corporate level. Next to this objective, this research aims to prove the existence of the relation between local adaptation efforts and higher business unit performance. In order to find evidence for these research objectives, a quantitative way of data collection is used. A questionnaire is send to members of one fitness chain in The Netherlands and Belgium. The results of the analyses with this empirical data show that there are no significant relations between the multidimensional concept of local adaptation and the three different dimensions of business unit performance. However, a Mann-Whitney test shows a significant difference in sales performance between franchisees that conducted more frequently local adaptation activities than franchisees that conducted these activities less frequent. This test also shows a significant difference in local adaptation efforts, based on the local competition levels in the environment of the franchisees. It shows that more local competition means more local adaptation efforts. Last, a split up of the multidimensional concepts of local adaptation and business unit performance in single dimensions shows a significant relation between price and sales. Conclusively, this study shows that there is a significant relation between price and sales and that there are some significant differences found (with the Mann-Whitney test) in the relationship between local adaptation and business unit performance, but this study did not find evidence that supports the main relation. This is probably caused by a too small sample size. This means that this study provides “parts of the puzzle” in the local adaptation/standardization discussion and leaves room for further research on this debate.
2
TABLE OF CONTENT
Introduction 4
The challenge 4
Franchising and its members 4
The mismatch in interest 4
Franchising and entrepreneurial behaviour 5
The importance of local adaptation 6
Research gap 6
The franchisee’s perspective 7
Contributions to the literature 7
Main question 8
Implications 8
Literature review and hypotheses development 9
The franchising context 9
Previous literature on standardization/local adaptation 9
Explaining local adaptation 10
Assessing the degree of local adaptation 11
Relationship with performance 12
Dimensions of performance 13
Hypothesis local adaptation and performance 13
Heterogeneity in the local market of the franchisee’s business unit 14
Conceptual model 16
Methods 17
Sample and data collection 17
Measures 18
The degree of local adaptation 18
Business unit performance 19
Local market heterogeneity 22
Data analysis 23
3
TABLE OF CONTENT (continued)
Post hoc analyses 31
Sales 35
Profit 36
Customer attraction 37
Customer satisfaction and retention 38
The moderating variable 41
Discussion, conclusions and implications 42
Discussion and conclusions 42
Answer on main question 43
Implications 45
Limitations and future research 47
References 48
Appendix 54
Appendix 1: Measures of variables 54
Appendix 2: Origin of variables 57
Appendix 3: Regressions with sales (Post Hoc) 58
Appendix 4: Regressions with profit (Post Hoc) 60
Appendix 5: Regressions with customer attraction (Post Hoc) 62
Appendix 6: Regressions with customer satisfaction/retention (Post Hoc) 64
Appendix 7: Overview of literature on standardization/local adaptation 66
4
INTRODUCTION
The challenge
A key challenge described in the management literature is finding the balance between standardization of a firm’s overarching brand elements and adaptive behaviour of the individual business units to the local business environment (Schmid & Kotulla, 2011; Szymanski, Bharadwaj & Veradarajan, 1993). The franchising literature is one of the literature streams that has a growing interest into this challenge (Dada, 2016; Dant & Gundlach, 1999; Kaufmann & Eroglu, 1999). Franchising has become of growing interest in the academic literature in the past three decades (Combs & Ketchen, 2003). This is not surprising, seen the crucial role that franchising fulfils in most current economies. In the U.S., franchising accounts for more than 1 trillion dollars in sales a year and represents 17% of the GDP (Gilles, Combs & Ketchen, 2014). Besides this fact, Gilles et al. (2014) conclude that the number of franchise units in the U.S. market has made a tremendous growth of 40% between the years 2001 and 2008, making the growing importance of franchising literature a logical effect.
Franchising and its members
Franchising is an organisational form based on a long-term contract between the owner of a service, trademarked product or business format (franchisor) and the franchisee, who pays for the rights to distribute the product or service (Storholm & Scheuing, 1994). In this relationship, the franchisor has a centralized position and is responsible for implementing chain-wide standards for performance. On the other hand, the franchisee has a decentralized position and is responsible for the day-to-day operations in the local outlets and is concerned with local policies (Combs, Michael & Castrogiovanni, 2004). A very common theory that is used to explain conflicting perspectives in this franchisor-franchisee relationship is the agency theory. In this relationship, the franchisor (the principal) delegates authority to the franchisee (the agent) (Barthélemy, 2011; Combs et al., 2004). In general, the franchisor is more concerned with standardization to achieve economies of scale and a uniform image. Besides that, the franchisor has a main focus on system wide turnover, whereas the interest of the franchisee is often focused on local adaptation and unit profitability (Croonen, 2010). This creates a potential mismatch between the interests of the agents and the principal. To make sure that the agents keep acting in the interest of the principal, high monitoring and control costs for the franchisor are often an inevitable consequence (Combs et al., 2004).
The mismatch in interests
5 concepts. Achieving economies of scale and a uniform image (main interest of the franchisor) are objectives that benefit from standardization. This is very opposed to the concept of adaptation to the local market (main interest of the franchisee), which requires the ability to make local decisions (Bradach, 1997; Kaufmann & Eroglu, 1999). The main benefit of local adaptation is the ability to respond to the demand of the local customers, making the firm more customer-oriented. Teece (2010) describes this customer-centricity as more crucial than ever before, since the developments in the global economy have changed the customer-supplier balance in a way that the customers have more freedom of choice and therefore will choose for the supplier that serves their specific demand best. Dada (2016) found a similar field of tension as Kaufmann and Eroglu (1999) by researching the “extent to which franchisees are able to exercise independence of thoughts and actions to operate freely outside the standardized confines of the franchise system” (p. 2), something she conceptualizes as franchisee entrepreneurial autonomy. More autonomy for the franchisee increases the chance for non-alignment of the franchisee’s and franchisors goals (Cochet, Dormann & Ehrmann, 2008). To avoid that, the franchisor must use costly relational governance mechanisms, which increases agency costs and is therefore a negative outcome for the franchisor (Cochet et al., 2008; Dada, 2016). Bradach (1997), Dada (2016) and Kaufmann and Eroglu (1999) all describe it as one of the most difficult and persistent managerial challenges for franchisors to find a balance between these two objectives of franchising.
Franchising and entrepreneurial behaviour
6
The importance of local adaptation
Fortunately, in the last decades, the ‘local adaptation perspective’ is far more often used by researchers, resulting in a growing amount of theoretical arguments which show that a higher degree of local adaptation can lead to an increase in business performance (i.e. Dada, 2016; Kaufmann & Eroglu, 1999; Theodosiou & Leonidou, 2003). This means a turning point in the literature, where the role of local adaptation behaviour and the role of the franchisee gain momentum in their perceived importance for franchising business formats. An example of this evidence comes from Sorenson and Sørensen (2001), who argue on a theoretical basis that geographic dispersion of the different franchise units of the chain lead to variance in local market conditions. They continue by stating that local adaptation is required under these conditions to maximize chain wide performance and that franchised units are most suitable for this adaptation. This implies that not only the franchisees can benefit from local adaptation, but that also franchisors can expect benefits that serve the whole chain. In a systematic review on this discussion, Dada (2016) came with a similar conclusion about the role of the franchisee, by arguing that increased franchisee autonomy to adapt to the local market increases firm performance. The theoretical paper of Kaufmann and Eroglu (1999) also endorsed the importance of local adaptation in changing environments. These authors go a step further by stating that especially the franchisee is able to create innovations on the local level in this changing environment. Furthermore, Michael (1996) also came to a similar conclusion about the role of the franchisee, by stating that franchisees cannot fully use their human capital under standardization conditions, which include their knowledge about local conditions. However, although representing a small part of the literature, not every research presents positive results regarding local adaptation. An example is the paper of Winter, Szulanski, Ringov and Jensen (2012), who state that, in contradiction to the main thought, local adaptation activities (in this case selling nonstandard products) lead to an increased risk of failure for the business unit. These authors advice to stick with the “template” rather than adapt to the local circumstances.
Research gap
7 relation. The existence of this gap in the literature about the relation between the franchisee’s ability to locally adapt and business unit performance can partially be explained. First, the role of franchising in general, especially in the macro-economic system, has long been undervalued (Michael, 2014). Secondly, as mentioned before, franchising is long been assessed as a standardized business format, where any form of entrepreneurial behaviour could not play any role in unit performance (i.e. Dant & Gundlach, 1999).
The franchisee’s perspective
In this research, I will address the above stated empirical gaps in the literature. I take the perspective of the franchisee, suggesting that higher levels of local adaptation lead to superior franchisee unit performance. Local adaptation will be assessed based on the dimensions product, price, communication and promotion, adopted from the marketing mix (Theodosiou & Leonidou, 2003). Although the perspective of the franchisor concerning the downside of local adaptation for the economies of scale, brand uniformity and agency costs at the system level is also important to take into consideration, this research will focus on the perspective of the franchisees and their business units within one franchise system. The major reason to focus on the perspective of the franchisee is two folded. First, researchers like Shah, Rust, Parasuraman, Staelin and Day (2006) and Teece (2010) describe the growing importance of customer centricity and increasingly divergent customer demands, something that can be best addressed on a local level. However, this does not provide evidence that the ability to serve the customer demand locally results in actual financial or non-financial benefits. This is an important missing “piece of the puzzle” to find the right standardization/local adaptation balance. Second, the evidence for the existence of the downsides of local adaptation from a franchisor perspective like less brand uniformity and agency costs are well known in the franchising literature (i.e. Dant & Gundlach, 1999; Kaufmann & Eroglu, 1999).
Contributions to the literature
8 Although this relation between the degree of local adaptation and business unit performance seems already interesting enough, a couple of papers describe market heterogeneity as important environmental condition that can impact the proposed mechanism between the degree of local adaptation and business unit performance (i.e. Sorenson & Sørensen, 2001). The exact way how this moderator affects the proposed mechanism will be explained more extensively in the theory section. Considering the above proposed relation, the following question represents the backbone of this research: What is the effect of the franchisee’s degree of local adaptation on the performance of the franchisee’s business unit, while considering the differences in heterogeneity of the market?
Implications
9
LITERATURE REVIEW AND HYPOTHESES DEVELOPMENT
The franchising context
Franchising is an entrepreneurial activity, which is a crucial concept for economic expansion and new job creation (Barthélemy, 2011). In franchise constructions, the franchisor as chain operator collaborates with the franchisee to create economic value. Combs et al. (2004) define franchising as one firm (the franchisor) selling the rights to market services and goods using their brand name and business practices (the business format) to a second firm (the franchisee). Although the reasons to franchise from a franchisors perspective are found ambiguous, many researchers report franchising as important strategy to facilitate huge growth (Combs & Ketchen, 2003; Shane, 1996), as capital structure (Norton, 1995) or as distribution channel (Dant, Paswan & Kaufman, 1996). A common problem that exists in every context where two parties work together is the difference in goals that these parties pursue, a problem that is explained by the agency theory (Combs & Ketchen, 2003). As explained in the previous paragraph, franchisors are concerned with economies of scale, a uniform brand and low agency costs what benefits from standardization. On the other hand, franchisees are concerned with the local customer demands, an aspect that benefits from local adaptation. Of course, it must be noted that this separation of interests is not as clear as black and white. A franchisor will in the long term also benefit from serving the local demand and the franchisee will also benefit from standardized practices. However, when looking at the main interest, these parties are on opposite sides of the standardization/local adaptation scale. This paper aims to provide evidence, which can be used to find the optimal balance between these two extremes on the standardization/local adaptation scale.
Previous literature on standardization/local adaptation
10 adaptation strategy with the environment that the different units are located in. The focus of this fit is predominantly on marketing related aspects. This stream of research, which is the dominant one compared to domestic market studies, has its main origin in marketing journals and contributes to international marketing and international business literature. However, parts of this literature are not only published in marketing journals but rather in business/management journals, showing the importance of this issue for business scholars too (Schmid & Kotulla, 2011).
The second stream of literature examines this balance in domestic markets (sometimes in combination with subsidiaries in other countries) in a franchise context (i.e. Bradach, 1997; Cox & Mason, 2007, Mumdžiev & Windsperger, 2011; Winter et al., 2012). Within this literature stream, almost all papers take the perspective of the franchisor or corporate level as starting point to examine the advantages and disadvantages of standardization, which are mentioned in previous sections. This literature often connects this discussion to the entrepreneurial autonomy or decision rights of the franchisees (i.e. Cox & Mason, 2007; Dada et al., 2012; Dada, 2016). This second stream of literature contributes to more general management and the entrepreneurship and franchising literature. Conclusively, I can state that both streams of literature have the main focus on the perspective of the franchisor/corporate level. Although the role of the franchisee gets some attention, most literature does not take the perspective of the franchisee into consideration. This paper takes the viewpoint of the franchisee, what implies that this paper assesses the standardization/local adaptation dilemma from a new perspective compared to most previous literature.
Explaining local adaptation
11 in local adaptive behaviour, something Lumpkin et al. (2009) define as a form of firm-level entrepreneurial behaviour. Next to that, there exist differences in the level of entrepreneurial behaviour that the different franchisees express. The cause of these differences can be found in differences in personal characteristics of the individuals, something the Theory of Planned Behaviour (TPB) describes as antecedents of entrepreneurial behaviour (Ajzen, 1991). Although these antecedents are not part of this research, gaining understanding of how differences in entrepreneurial behaviour between the different franchisees come to existence and how this eventually results in differences in local adaptation efforts can be very relevant for the franchisor when assessing the local adaptation performance of his/her franchisees.
Conclusively, I can state that the level of local adaptation depends on both the autonomy granted to the franchisees in the contract with the franchisor and the antecedents that lead to differences in local adaptation efforts of the different franchisees. However, this study will not focus on the factors that cause the degree of local adaptation, but rather on the consequences of local adaptation. Theodosiou and Leonidou (2003) argue in their study about organisation wide performance that the decision to go for standardization or local adaptation is not black or white. Rather, they argue to assess both as the two ends of a continuum, on which combinations of both in different distributions are possible. It is one of the most difficult management issues for franchisors to find this right balance and achieve both a level of uniformity to obtain economies of scale and simultaneously avoid stifling efficient adaptive behaviour to the local market (Kaufmann & Eroglu, 1999). Theodosiou and Leonidou (2003) argue that finding the right balance should be assessed based on its impact on the entire organisational performance, but this relation with performance received unfortunately little empirical attention and only involves chain wide performance. Schmid and Kotulla (2011) make a similar statement by arguing that this balance must have the right performance-enhancing fit.
Assessing the degree of local adaptation
12 the marketing mix (McCarthy, 1975). Katsikeas et al. (2006) take a similar approach and also connect their measurement of local adaptation to a slightly different version of the marketing mix than Schmid and Kotulla (2011) use. Although this measure seems appropriate, it is not directly applicable for this research. The reason for this fact is that both papers use the 4 p’s in an international business context, where local adaptation is assessed on country level, rather than the franchise context where local adaptation is assessed on unit level, as used in this paper.
To make the 4 p’s marketing mix a proper measure for local adaptation in a franchising context, gaining understanding of the elements of this marketing mix tool is of critical importance. In its original form, the marketing mix consists of the elements product, price, promotion and place (McCarthy, 1975). Years later, Van Waterschoot and Van den Bulte (1992) introduced a revised version of this marketing mix, after some critique on the original one. This revised version forms the basis of most current research, like the research of Schmid and Kotulla (2011). It contains the basic elements product, price, distribution and communications, in which the latter is split in mass, personal and publicity communication. A secondary element is formed by promotion, which can strengthen one of the 4 mentioned elements for a short period of time (i.e. offering end-of-season sales as promotion mix for the dimension price) (Van Waterschoot & Van den Bulte, 1992). Following the previous studies that used the 4 p’s marketing mix of Van Waterschoot and Van den Bulte (1992) to measure the degree of local adaptation (Katsikeas et al., 2006; Schmid & Kotulla, 2011), this study will slightly modify the 4 p’s marketing mix tool with the elements product, price, communication and promotion to the local demand of the franchisee’s business unit to measure the degree of local adaptation that the franchisee conducts.
Relationship with performance
13 and business performance in an international context. Again, this leaves an important gap in the literature that can have important performance related consequences.
Dimensions of performance
Despite the fact that there are a lot of studies available that measure performance (as summarized in Schmid and Kotulla, 2011), the used performance dimensions are very divergent. Consequently, there is no clear measurement of performance in relation to the degree of local adaptation. Therefore, this study will use a common measure of business unit performance from the entrepreneurship literature, adopted from Kollmann and Stöckmann (2014). The use of this measure in this context is justifiable, since local adaptation is closely linked to, and according to some authors (i.e. Lumpkin et al., 2009) even part of, entrepreneurial behaviour. This measure uses dimensions of both financial and non-financial performance. The non-non-financial performance measures are included to take a broader perspective, since Venkatraman and Ramanujam (1986) conclude that only focussing on financial performance indicators assumes that financial goals are always dominant in the firms’ system of goals, while this is not always the case. To make this measure fit the local adaptation context, it will be slightly modified. Concerning the financial dimensions, both sales and profit will be used. These dimensions are, according to the literature (i.e. Carton & Hofer, 2010; Venkataraman & Ramanujam, 1987) part of the most important financial dimensions. The used non-financial performance dimensions, which are all related to the customer, are satisfaction, attraction and retention. These dimensions are not brought into relation with local adaptation separately, but rather as sub-dimensions of one performance dimension named non-financial performance. Although there are studies that use these dimensions as single-dimensional items in their analyses (i.e. Kahn, 2012), this study follows the reasoning of Ittner and Larcker (1998) and Rust and Zahorik (1993). They state that retention and attraction are both closely related to satisfaction or even the consequences of satisfaction and therefore can be measured together as one dimension. Next to that, there is no reason to expect that local adaptation will have a different relation with one of these three dimensions.
Hypothesis local adaptation and performance
U-14 shaped) concerning the multidimensional concept of local adaptation in a franchising context is not researched before and therefore new in the organisational, entrepreneurship and franchising literature. Since both the concepts of local adaptation and business unit performance are multidimensional, it cannot be stated that all relations between local adaptation and the dimensions of business performance follow a curvilinear course. A curvilinear relation is logical for profit, since this dimension depends on both the amount earned (where local adaptation can be beneficial) and the amount spent (where economies of scale can be beneficial). In other words, the lack of economies of scale will eventually diminish the benefits of local adaptation, making the relation follow the predicted inverted U-shaped curve. Furthermore, other research (Shoham and Kropp, 1998) that used profit as dependent variable also predicted an U-shaped relationship with the independent variable. However, this reasoning is less logical for the other dimensions of performance. Sales is concerned with the amounts of sold products which local adaptation should enhance according to the presented theory about customer centricity (i.e. Teece, 2010) and, therefore, less concerned with the benefits of standardization. The same counts for the non-financial measures customer retention, attraction and satisfaction that benefit most from their local demands being served. Therefore, I hypothesize that the mechanism between local adaptation and sales and the mechanism between local adaptation and the non-financial performance dimensions will have linear, positive relations instead of an inverted U-shaped relation (see figure 1).
H1: Local adaptation will have a positive effect on the sales performance of the franchisee’s business unit
H2: Local adaptation will have an inverted U-shaped effect on the profit performance of the franchisee’s business unit
H3: Local adaptation will have a positive effect on the non-financial performance dimensions of the franchisee’s business unit
Heterogeneity in the local market of the franchisee’s business unit
15 mechanism between local adaptation and performance stable. In other words, the strength of both mechanisms between the degree of local adaptation and the two forms of business unit performance (sales and non-financial) decreases in a more heterogeneous local market environment. This effect, which changes the strength of the proposed mechanism, can be noted as a moderating effect (Baron & Kenny, 1986). In case of profit, I expect that it will take longer to reach the optimum of the inverted U-shaped relation, meaning that the course of the graphs between point zero and the optimum will be less steep. Conclusively, it is expected that higher levels of local market heterogeneity moderate the strength of both the proposed mechanisms in a negative way (see figure 1 for the conceptual model).
H4a: Local market heterogeneity weakens the strength of the mechanism between local adaptation and sales performance
H4b: Local market heterogeneity weakens the strength of the mechanism between local adaptation and profit performance
16
Figure 1: Conceptual model. All dimensions in one grey square are measured as one concept. This model contains five concepts.
Degree of local adaptation
17
METHODS
Sample & data collection
The data needed to test the stated hypotheses is collected in one franchise chain named Fit Today1. It is
beneficial to collect this data within one franchise chain, to make sure that organisational or industrial differences do not interfere with the results. This means an improvement of the internal validity (Davies, Lassar, Manolis, Prince & Winsor, 2011). The health and fitness industry characterizes itself by high levels of switching customers and severe competition (Hurley, 2004). Companies in such environments tend to be more innovative, what makes it a good industry to analyse the existence of differences in local adaptation practices. These local adaptation practices are associated with entrepreneurial behaviour and its organisational outcomes (Kollmann & Stöckmann, 2014). The franchise chain Fit Today has units in multiple countries, with more than 4.000 units combined. The franchise chain allows their franchisees to deviate to some extent from the standard procedures of the franchisor, making differences in local adaptation behaviour possible. I focus on the franchisees’ business units that are located in the Netherlands and Belgium, which sample contains 74 units.
The data is collected with a digital survey among franchisees and therefore of a quantitative nature, like most empirical studies in franchising (Cox & Mason, 2009). The survey took place in the year 2018. First, the master franchisee, who has the franchise right for the Benelux, was approached to provide the names and mail addresses of all franchisees that fall under his jurisdiction. This master franchisee also encouraged the franchisees to collaborate, by making an announcement of this survey in their chain wide newsletter. The franchisees were approached with an email, containing personal login data for the digital survey. This approach resulted in a response rate of 28. Of this response rate, 3 responses were excluded from further analysis, since 2 of them were incomplete and 1 was filled in careless. This resulted in a net response rate of 25 franchisees. When excluding units that are owned by a franchisee that owns more than 1 unit, the sample size of 25 is equal to a response rate of 47%. I can already conclude that this sample size is rather low. Therefore, this study and the corresponding conclusions must be interpreted as the result of a study with an explorative nature.
Several authors describe that ownership of multiple franchise units can influence the research finding, since they can differ in behaviour and motivation concerning participation in franchising (i.e. Grünhagen & Mittelstaedt, 2005). In the sample of this franchise system, 28% of the respondents owned multiple units, what is equivalent to 7 respondents out of 25 (6 with 2 units, 1 with 4 units). Since this percentage is relatively low, the impact is minimal. In order to exclude any risk of interference with the dataset, I included a variable that asked the multiple unit owner franchisees to answer the questions of the survey based on the unit he/she considers as the most important one.
1
18
Measures
Below, the measures for each variable are explained in detail. For all variables, the reliability is calculated and presented in the form of Cronbach’s alpha’s. This study uses a cut-off point of 0.70, as used by Nunnally (1978), which is the most widely referenced reliability coefficient (Peterson, 1994). Therefore, measurements with values below 0.70 are not reliable. The reliability of two-item scales, which are used for the dimensions sales and profit, are also measured with Cronbach alpha. Although there is no agreed upon method to measure the reliability of two-item scales (Eisinga, Grotenhuis & Pelzer, 2013), many research does use the Cronbach alpha for this measurement (i.e. Cuijpers, Smits, Donker, Ten Have & De Graaf, 2009; Löwe, Kroenke & Gräfe, 2005).
The degree of local adaptation is the independent variable. To measure the degree of local adaptation
that the different franchisees conduct, I use the 4 p’s marketing mix developed by Van Waterschoot and Van den Bulte (1992) as basis. This mix exists of the elements product, price, distribution and communication. Promotion is a fifth element that can strengthen the four other elements. Distribution is excluded as element since it is nearly impossible for a franchisee in the fitness industry to adapt the distribution of the products and services to the local environment. The survey contains questions about to what degree the four elements of the marketing mix (price, product, promotion and communication) are used by the franchisee to adapt to the local market. In other words, to what degree the elements of the marketing mix are used for practices outside the standards that are implemented chain wide. The questions per element of the marketing mix (total of 12 items, see appendix 1) are based on the topics suggested by Katsikeas et al. (2006) and Schmid and Kotulla (2011) and measured on a five-point Likert scale, ranging from “Never” to “Very frequently”. An example is the question for product:
“How often do you change your assortment of products?”.
19 multiple factors. These variables are excluded from further analysis. After exclusion, the results show that all four factors from the marketing mix have greater factor loadings than 0.5, which is a common cut-off level (Govindarajan & Kopalle, 2006). Table 1 shows the loadings of the remaining factors and the variance explained, based on the rotated component matrix. The percentage of variance explained is 36.2, 20.6, 15.1 and 14.8 respectively. Conclusively, I can state that the results demonstrate a discriminant validity for this scale for local adaptation. The total scale with the remaining eight items has also proven to have a sufficient reliability (α = 0.70). In order to create one total value for the degree of local adaptation, the values for each dimension are summed and there after divided by the number of variables within that dimension. These values for each of the four dimensions are added up and divided by four, to create the total value for local adaptation.
Table 1 Loadings and explained variance EFA for local adaptation
Construct and item wording SL % variance
explained
Communication 36.2
Local adaptation 6 (promoting discounts) .86
Local adaptation 8 (advertising, current customers) .82
Local adaptation 9 (advertising, potential customers) .85
Local adaptation 10 (use of local press) .77
Product 20.6
Local adaptation 1 (change assortment) .87
Local adaptation 3 (change services) .90
Price 15.1
Local adaptation 5 (use of price reductions) .91
Promotion 14.8
Local adaptation 11 (use of extreme discounts) .93
Business unit performance is the dependent variable, which is influenced by the independent variable
20 available. In a validation study, Chandler and Hanks (1993) give an overview of the three most common methods to measure business performance. These are; 1) using broadly defined categories of performance, based on objective data, 2) using subjective measures to determine the owners’ satisfaction with performance, 3) using subjective measures to compare firm performance with competitors’ performance.
Although all three methods have both pros and cons concerning availability of data, consistency, reliability and validity, I use the second method. This method is named in various papers as the weighted satisfaction with performance index (i.e. Covin, Slevin & Covin, 1990) and is built on an instrument of Gupta and Govindarajan (1984). According to the study of Chandler and Hanks (1993), using the subjective weighted satisfaction with performance index ensures a high availability of data, a
good internal consistency and reliability. The choice to use only subjective data in this study is
two-folded. First, this studies’ field research suggested that the franchisees were unable or unwilling to hand over objective performance data, a fact that is also reported in other studies (i.e. Katsikeas et al., 2006, Tang, Kreiser, Marino & Weaver, 2010). In that respect, subjective performance measures create more complete information for research (Covin & Slevin, 1989). Second, there is empirical support that shows a proper reliability and validity for using exclusively subjective performance measures (Venkataraman & Ramanujam, 1987). Although the use of subjective data in combination with the weighted satisfaction with performance index has a lot of upsides, there is also one downside of this measure mentioned in the literature. This is an inadequate external validity, what means that respondents would sometimes interpret subjective and objective performance as two different things (Chandler & Hanks, 1993). However, later research refutes this downside and found support for the validity of this subjective performance measure by comparing the subjective data with objective data and proved that there exists no significant difference between these two measures (Kollmann & Stöckmann, 2014; Stam & Elfring, 2008).
21 The weighted satisfaction with performance index offers a solution. This index first asks respondents to score how important they perceive that the given different performance dimensions (sales, sales growth, profit, profit growth, customer satisfaction/attraction/retention) are to him/her. This is measured on a five-point Likert scale, ranging from “not important at all” to “very important” (Lomberg et al., 2017; Tang et al., 2010). The sum of these “scores” need to be equal to one, what means that each score is divided by the sum of all scores (Chandler & Hanks, 1993). Thereafter, using the same list of performance dimensions, the respondents are asked to report their satisfaction with these different performance dimensions. This measure is also based on a five-point Likert scale, ranging from “not satisfied at all” to “highly satisfied” (Lomberg et al., 2017; Tang et al., 2010). This score will be multiplied with the scores obtained from the importance measurements and transformed to numbers on a scale ranging from 0 to 1 in order to create the weighted satisfaction with performance index (Chandler & Hanks, 1993).
As stated above, the most representative performance measures are very context dependent and often not known beforehand. Therefore, this research has used a modified version of the dimensions used in the paper of Kollmann and Stöckmann (2014). For the financial performance measure, the sub-dimensions are sales, sales growth, profit and profit growth. The sub-sub-dimensions current sales and sales growth are combined in one dimension after measurement took place, which dimension is named sales (2 dimensions, 2 items (importance and satisfaction) per dimension, example importance: “Importance of annual growth of sales” and example satisfaction: “Satisfaction with annual growth of sales”). First, I measured the importance of sales. This scale has proven to have a reliability of α = 0.62. Since this value is below 0.70, it would imply that this scale is not reliable according to Nunnally (1978). However, a lot of research pointed out that levels above 0.60 are acceptable as well (i.e. Davis, 1964; Murphy and Davidshofer, 1988; Nunnally, 1967, as cited in Peterson, 1994). Therefore, I qualify this value as a below standard reliability, but continue further research with this scale since the above named authors do approve levels above 0.60. After this, I measured the satisfaction of sales. This scale has a higher reliability (α = 0.89). Profit is measured in the same way as sales, but exists out of the sub-dimensions current profit and growth in profit (2 dimensions, 2 items (importance and satisfaction) per dimension, example importance: “Importance of current profit” and example satisfaction: “Satisfaction with the annual growth in profit”). This scale has proven to have a high reliability in combination with importance (α = 0.86) and an exceptionally high reliability in combination with satisfaction (α = 0.99).
22 of Fit Today (3 dimensions, 2 items (importance and satisfaction) per dimension, example importance: “Importance of customer retention” and example satisfaction: “Satisfaction with customer attraction”). The dimensions are not split up in sub-categories, but represent together the dimension non-financial performance. Just like the measure of sales and profit, the respondents are first asked how important they perceive these dimensions and thereafter how satisfied they are with their performance on these dimensions. This scale has proven to have a sufficient reliability in combination with importance (α = 0.71) and an even higher reliability in combination with satisfaction (α = 0.84). Conclusively, I use subjective data in combination with the weighted satisfaction with performance index to measure the performance of the different franchisees’ business units, based on the dimensions within the categories financial and non-financial indicators.
Local market heterogeneity is the moderating variable. In order to measure the level of heterogeneity
in the local market of the franchisee, the measurement as published in a paper of Miller and Friesen (1982) is used. This measurement method exists of 4 items (example: “To what extent do you agree with the statement that the annual demand for my products/services varies strong per product group?”) to assess the level of heterogeneity in the local environment of the franchisee’s business unit, using a five-point Likert scale, ranging from “Strongly disagree” to “Strongly agree”. This scale has proven to have a not sufficient reliability (α = 0.63). Just like explained by the reliability of the importance with sales, I classify this as a below standard reliability, but use it in further analyses. Next to that, the items used in this measure are conceptually close to and based on the validated scale of environmental heterogeneity described by Miller and Friesen (1982), with a highly acceptable Cronbach alpha of 0.84. Therefore, it seems that the overall measure represents the construct adequately.
Age, size and floor space are used as control variables. Previous research has found that firm age and
23 like firm age, this measure also showed a skewed distribution. Therefore, I use the logarithm of this variable for the further analyses.
Data analysis
The collected data will be used to test the stated hypotheses. I test hypotheses 1, 3 and 4 separately, all with a simple linear regression analysis. This simple linear regression is used to find the effects of the independent variable local adaptation on the performance variables sales and non-financial dimensions, which relations are moderated by local market heterogeneity. For hypothesis 2, I use a curvilinear regression, because I hypothesised an inverted U-shape and therefore non-linear effect. Before these regressions can be executed, I have to make sure that the data is collected free from biases. Since I collected the data about all variables from the same person in an equivalent measurement context, method biases can have occurred (Podsakoff, MacKenzie, Lee, Podsakoff, 2003). These biases often cause measurement errors. These errors “threatens the validity of the conclusions about the relationship between measures” (Podsakoff et al., 2003, p. 879) and therefore can provide an incorrect explanation for the found relationship. This also counts for the relationships that are examined in the post-hoc analyses.
One common method bias that could have occurred in this study is social desirability. This bias refers to “the need for social approval and acceptance and the belief that it can be attained by means of culturally acceptable and appropriate behaviour” (Crowne & Marlowe, 1964, p. 109). In the case of this research, it is arguable that franchisees feel a form of pressure from their franchisor to achieve high financial performance results, since the income of the franchisor dependents highly on the royalty fees from the franchisee’s turnover. This could imply that the franchisees fill in high scores on all forms of performance importance, despite their real feelings about this importance of the different performance indicators. As suggested by Podsakoff et al. (2003), I reduced the chance for occurrence of this method bias by guarantying the respondents anonymity and asked to answer the questions as honest as possible.
24 (Podsakoff et al., 2003; Chang, Witteloostuijn & Eden, 2010). Therefore, I proceed with further analyses with the collected data.
RESULTS
Table 2 shows the descriptive statistics of the used variables. The descriptive statistics show that the average franchisee within the Fit Today chain does indeed engage in local adaptation, which has a mean score of 2.82 (on a scale ranging from 1-5). Another conclusion that can be drawn from these descriptive statistics is that the weighted satisfaction with performance index is most high for the non-financial performance indicators, with a mean value of 0.36 (other two values are 0.31 and 0.33). Last, the descriptive statistics show that the number of employees (size of business) contains a few outliers. The maximum of 15 employees is an exception, since most units have 2 or 3 employees.
Table 3 presents the correlations between the variables. When interpreting the results from this correlation table, I conclude that four correlations are found significant. Of these four correlations, two correlations between the control variable floor space and the dependent variables sales and non-financial show a significant correlation of moderate strength (values around 0.5/-0.5 with p < .05). The variables sales and profit both show a significant, strong negative correlation with the variable non-financial performance (values -.717 and -.854 with p < .01). This first prediction of possible relations implies that these variables are possibly related to each other, what deserves more attention. Before analysing the regressions, I conducted a Mann-Whitney analysis to check for significant differences between the respondents concerning the level of local adaptation. This test shows that there is a significant difference in the sales performance between the group franchisees who were local adaptive (mean score of 3 or higher) and the group who were not local adaptive (mean score below 3), (p < 0.05). Although this cannot be interpreted as evidence for a regressed relationship between the two variables, this Mann-Whitney test shows that there is at least some kind of relation between the two variables present.
Table 4 shows the results from the first linear regression analysis, in which local adaptation is regressed on sales performance. Model 1 contains the control variables. This regression analysis is
significant (R2 = .36, F = 3.99, p < .05). In model 2, the independent variable local adaptation is added.
This regression analysis is not significant (R2 = .40, F = 2.51, p > .05). Therefore, there is no
significant relationship found between local adaptation and sales performance (β = .00, p > .05) as hypothesis 1 suggested. So, hypothesis 1 is not supported. Model 3 shows the same regression as model 2, to which the moderator local market heterogeneity is added. This regression is not significant
(R2 = .40, F = 2.00, p > .05). The moderation effect of this moderator is found insignificant (β = -.00, p
25 Table 5 shows the results from the second linear regression analysis, in which local adaptation regressed on non-financial performance. Model 1 contains the control variables. This regression
analysis is not significant (R2 = .27, F = 2.53, p > .05). In model 2, the independent variable local
adaptation is added. This regression analysis is not significant (R2 = .30, F = 1.63, p > .05). Therefore,
there is no significant relationship found between local adaptation and non-financial performance (β =
-.01, p > .05) as hypothesis 3 suggested. So, hypothesis 3 is not supported. Model 3 shows the same
regression as model 2, to which the moderator local market heterogeneity is added. This regression is
not significant (R2 = .31, F = 1.33, p >.05). The moderation effect of this moderator is not found
significant (β = .01, p > .05). Therefore, hypothesis 4c is not supported either.
Table 6 shows the results from the curvilinear analysis, which is used to find evidence for hypothesis 2. In this hypothesised relation, local adaptation is regressed on profit performance, following an inverted U-shape. Model 1 contains the control variables. This regression analysis is not significant
(R2 = .10, F = .78, p > .05). In model 2, the independent variable local adaptation is added together
with a squared version of this variable. The squared variable is needed for the expected U-shaped
relationship. This regression analysis is not significant (R2 = .13, F = .56, p > .05) and it is therefore
not possible to interpret the relations any further (β = .02, p > .05). Therefore, hypothesis 2 is not supported. Model 3 and 4 show the same regression as model 2, to which the moderator local market heterogeneity is added and multiplied with local adaptation. This regression is not significant for
model 3 (R2 = .13, F = .44, p > .05), (β = .00, p > .05) and model 4 (R2 = .14, F = .39, p > .05), (β =
26
Table 2 Descriptive statistics
Variable Min. Max. Mean S.D. Scale
Local adaptation 1.50 3.63 2.82 0.52 1-5
Financial performance – Sales* Importance Satisfaction WSPI 3.50 2.00 0.26 5.00 5.00 0.37 4.44 3.98 0.33 0.46 0.78 0.02 1-5
Financial performance – Profit* Importance Satisfaction WSPI 3.00 2.00 0.23 5.00 5.00 0.36 4.36 3.90 0.31 0.53 0.84 0.03 1-5 Non-financial performance* Importance Satisfaction WSPI 4.00 2.00 0.31 5.00 5.00 0.48 4.72 4.16 0.36 0.38 0.71 0.04 1-5
Local market heterogeneity 2.00 4.00 3.13 0.48 1-5
Age of the business unit** 1.00 5.00 2.52 1.26 -
Size of the business unit 1.00 15.00 4.40 3.18 -
Floor space m2** 146 800 576 179 -
* After data collection, the importance and satisfaction scores are recalculated into a weighted satisfaction with performance index (WSPI)
WSPI example = (Importance sales/total importance)*(satisfaction sales/total satisfaction )
31
POST HOC ANALYSES
This post hoc analyses is designed to conduct further analyses with the collected data. In contrast to the main analyses of this paper, the post hoc does not analyse the data to find evidence for the relations that are hypothesised before data collection. Instead, it is designed to find relations that are based on the same theoretical arguments, but hypothesised after the main relations were analysed. This is a method that is often used in the literature when the main analyses show insignificant results and/or when the main analyses raise follow up questions that can be answered with the same dataset.
32 adaptation and dependent dimensions of performance are regressed with each other, but only when the possible existence of the relation is logical and collaborated by theoretical evidence.
33
Table 7 Descriptive statistics (post-hoc)
Variable Min. Max. Mean S.D. Scale
Independent variables Price 1.00 4.00 2.32 .90 1-5 Product 2.00 5.00 2.96 .73 1-5 Promotion 1.00 4.00 1.92 .86 1-5 Communication 1.00 4.00 3.09 .83 1-5 Dependent variables Sales Importance Satisfaction WSPI 3.50 2.00 0.26 5.00 5.00 0.37 4.44 3.98 0.33 0.46 0.78 0.02 1-5 Profit Importance Satisfaction WSPI 3.00 2.00 0.23 5.00 5.00 0.36 4.36 3.90 0.31 0.53 0.84 0.03 1-5 Customer attraction Importance Satisfaction WSPI 3.00 2.00 0.18 5.00 5.00 0.44 4.60 4.12 0.35 0.58 0.83 0.05 1-5 Customer satisfaction + retention Importance Satisfaction WSPI 4.00 2.00 0.30 5.00 5.00 0.52 4.78 4.18 0.37 0.38 0.73 0.06 1-5 Control variable Floor space m2 4.98 6.68 6.29 .40 -
* After data collection, the importance and satisfaction scores are recalculated into a weighted satisfaction with performance index (WSPI)
35
Sales
Expectations. The dependent performance dimension sales is regressed with the 4 dimensions of local
adaptation (figure 2). I expect that the relation with price is positive, since putting more effort in adapting the prices to the local willingness to pay, often means an increase in the sales. This could imply that the prices of the services of the Fit Today unit are lowered, a method that is empirically proven by Horgen and Brownell (2002) to increase sales in the restaurant industry. The relation with product is expected to be positive, since an assortment that is tailored to the customer demand will attract more customers, resulting in higher sales. This relation is supported by theoretical evidence of Shoham and Kropp (1998). The relations with the other 2 dimensions promotion and communication will be positive, since an increase in the amount of discounts or the communication about discounts and new equipment will attract more customers, resulting in higher sales. I conducted a set of simple regression analysis to find evidence for these described individual relationships.
Figure 2: Conceptual model sales
Results. Table 9 shows the results from the regression between the independent variable price and the
satisfaction with performance index of sales performance. Model 1 contains the control variables. In
model 2, the independent variable is added. This regression analysis is found highly significant (R2 =
.37, F(2, 22) = 6.40, p < .01). Next to that, the results show a positive significant effect between the two variables (β = .01, p < .05). Therefore, I can conclude that price has a positive influence on the sales of the business unit, like was proposed. The other relations are found insignificant. Due to space constraints, only the table with the significant regression is presented in the main text (table 9). The other regression tables of sales are presented in the appendix 3.1 to 3.3.
36
Table 9 Regression between Sales and price (Post Hoc)
Variable Model 1 Model 2 β S.E. β S.E. Intercept .15* .07 .15* .06 Control Floor space m2 .03* .01 .02* .01 Main effect Price .01* .00 R2 .22* .37** ∆R2 .31 ANOVA F 6.48 6.40
n = 25, * = p < .05 (two tailed), ** = p < .01 (two-tailed)
Profit
Expectations. For the second set of regressions, the dependent performance dimension profit is
37
Figure 3: Conceptual model profit
Results. The regression analyses show for all proposed relations insignificant results. Due to space
constraints, these relations will not been discussed. The regression tables are presented in appendix 4.1 to 4.4.
Customer attraction
Expectations. This third set of regressions is based on possible relations between the independent
variables of local adaptation and the dependent variable customer attraction (figure 4). For the relation with price, I expect a positive relation, because efforts to adjust the price to the willingness to pay, in general, will attract more customers. For the other four relations, I expect a positive effect. Both Henning-Thureau, Gwinner and Gremler (2002) and Keh, Nguyen & Ng (2007) present theoretical evidence for the relation between promotion and communication with the attraction of new customers. This seems logical, since both elements of the marketing mix aim on attracting more customers by offering high discounts and/or communicating relevant information about product and price offerings.
Figure 4: Conceptual model customer attraction
Results. The regression analyses show for all proposed relations insignificant results. The relation
between customer attraction and communication is found significant (β = .03, p < .05), but the model
is insignificant (R2 = .23, F(2, 22) = 3.24, p > .05). Although it is not scientifically appropriate to
38 interpret these results, it could imply that there is a relation between communication and attraction when regressed with a larger sample size or when more variation in the model is controlled for with control variables. If so, this confirms the theoretical evidence of Keh et al. (2007). Next to that, the significance level of the model is p = 0.06, what is according to some researchers still significant (p < 0.10). The results of this regression are presented in table 10. The other regression tables are presented in appendix 5.1 to 5.3.
Table 10 Regression between Customer attraction and communication (Post Hoc)
Variable Model 1 Model 2 β S.E. β S.E. Intercept .42* .16 .40* .15 Control Floor space m2 -.01 .03 -.02 .02 Main effect Communication .03* .01 R2 .01 .23 ∆R2 .16 ANOVA F .17 3.24
n = 25, * = p < .05 (two tailed), ** = p < .01 (two-tailed)
Customer satisfaction and retention
Expectations. This final set of regressions is based on possible relations between the independent
39 elements of the marketing mix aim on attracting more customers by offering high discounts and/or communicating relevant information about product and price offerings.
Figure 5: Conceptual model customer satisfaction/retention
Results. The regression analyses show significant results for the relation with communication. This
regression with communication is presented in table 11. Model 1 contains the control variables. In
model 2, the independent variable is added. This regression analysis is found highly significant (R2 =
.38, F(2, 22) = 6.73, p < .01). Next to that, the results show a negative significant effect between the variables communication and satisfaction/retention (β = -.03, p < .01). This is contradictory to the proposition, in which a positive relation was expected. There is no evidence in the literature that confirms this negative effect. Therefore, further research should take a closer look at the causes of this effect. The regressions that were found insignificant are presented in appendix 6.1 to 6.3.
40
Table 11 Regression between Customer satisfaction/retention and communication (Post Hoc)
Variable Model 1 Model 2 β S.E. β S.E. Intercept .83** .19 .86** .17 Control Floor space m2 -.07 .03 -.06 .03 Main effect Communication -.03* .01 R2 .21* .38** ∆R2 .32 ANOVA F 6.01 6.73
n = 25, * = p < .05 (two tailed), ** = p < .01 (two-tailed)
Table 12 Expected and found relationships (Post-Hoc)
Variable
Sales Profit Customer
attraction Customer satisfaction/ retention E F E F E F E F Price
+
+
Inverted U-shape NS+
NS+
NS Product+
NS+
NS+
NS+
NS Promotion+
NS Inverted U-shape NS+
NS+
NS Communication+
NS+
NS+
NS+
-
41
The moderating variable
The moderating variable “local market heterogeneity” has delivered no significant effects as mechanism on the relations as hypothesised in the main part of this research. I replaced local market heterogeneity as moderator for a closely related concept named local competition, that is originally intended as optional control variable. The below standard reliability of the local market heterogeneity scale was already an indicator of the possibility for non-significant results. Since the concept of local competition is close to heterogeneity but does exist out of only one question, it is expected to have better applicability in this research with a small sample size. As discussed by Croonen et al. (2016), high levels of local competition can threaten the revenues, what can motivate franchisees to put more effort into the success of their business unit. These efforts and revenues can be seen as local adaptation activities in relation to business performance dimensions, which relation can be moderated by local competition, like was hypothesised with local market heterogeneity as moderated.
This moderating variable local competition is measured with the statement “There is a lot of competition from other fitness clubs in my local environment” on a five-point Likert scale, ranging from “totally disagree” to “totally agree”. A simple linear analysis shows no significant differences between the effect of local competition and local market heterogeneity as moderating variables on the hypothesised relations, what implies that a more competitive environments does not lead to a need for more local adaptation. However, a Mann-Whitney analysis shows that franchisees who were operating in a highly competitive environment, conducted in fact significantly more local adaptation activities than the group who were competing in a low competitive environment (p < 0.10). It is worth noticing that I used a mean score of 3 (corresponding to conducting local adaptation activities “sometimes”) as cut-off point to split the group of respondents (score of 3 or higher means more conducting more frequent local adaptation activities). These results imply that, although not found significant in the conducted regression analyses, there is an effect of the local competitive environment on the amount of local adaptation activities that a franchisee conducts.
42
DISCUSSION, CONCLUSIONS AND IMPLICATIONS
Discussion and conclusions
Finding the right balance between standardization and local adaptation is an ongoing challenge in the literature (Dada, 2016; Schmid & Kotulla, 2011). This challenge has found its way to the franchising literature. Within this literature stream, the general line of thought is that the franchisor has a main interest in standardization and the franchisee in adaptation to the local market (Bradach, 1997; Kaufmann & Eroglu, 1999). Since both standardization and local adaptation seem to have benefits for the performance of the business unit, empirical evidence is needed to find the right balance. This empirical evidence is, till this day, largely missing. Before I draw conclusions based on the evidence for the relation between local adaptation and business unit performance, it is necessary to find evidence that support that franchisees do engage in local adaptation behaviour at all. After all, if franchisees do not conduct the local adaptation activities that theoretical research suggest (i.e. Kaufmann & Eroglu, 1999), the relation with performance becomes irrelevant. I measured this with an adjusted form of the marketing mix scale, adopted from international marketing research papers (i.e. Katsikeas et al., 2006).
I found evidence for the fact that franchisees do engage in local adaptation activities. This evidence, with a mean score of 2.82, suggests that franchisees engage in local adaptation behaviour with a frequency of “sometimes”. Moreover, when running a frequency analysis, I can conclude that some franchisees conduct a lot more local adaptation than others (never/seldom = 8%, seldom/sometimes = 44%, sometimes/often = 48% of respondents). This implies, as some research also concluded, that franchisees cannot be approached as one homogeneous group (Croonen et al., 2016; Kidwell, Nygaard & Silkoset, 2007). Instead, as already briefly mentioned in the literature review, gaining understanding of why some franchisees conduct more local adaptation activities than other franchisees under the similar external circumstances can be really valuable information for the franchisor to gain a strategic fit between the chains strategy and the characteristics of potential new franchisees. Although this conclusion does not directly contribute to finding the right local adaptation/standardization balance, it could feed the understanding of how a certain level of local adaptation efforts comes into existence in every franchise chain. Eventually, this can help selecting the right franchisees when the perfect local adaptation/standardization balance is determined based on the performance outcomes for a franchise chain.
