PERFORMANCE DEVELOPMENT, WHERE TO FOCUS?
STUDY OF THE RELATIONSHIP BETWEEN MATURITY LEVELS AND
FINANCIAL PERFORMANCE
Master thesis
MSc Supply Chain Management
University of Groningen, Faculty Economics and Business
April 30, 2017 WOUTER BREEN Student number: 2227096 Email: w.p.s.breen@student.rug.nl Supervisor University Dr. X. Zhu Co-assessor University Dr. J. Veldman Company Supervisor Ir. M. Fiksinski Scenter, Driebergen-Rijsenburg
ABSTRACT
Organizations that engage in business process management will at some stage ask themselves how much they benefit from this engagement. Organizational maturity models have surfaced as a measure of an organization’s capability evaluation. Previous research suggests that a high organizational maturity level score leads to better organizational performance. However, insight into higher organizational maturity that leads to better financial performance has not been investigated yet. Furthermore, the relation between individual dimensions of the organizational maturity model and financial performance are investigated. This thesis investigates the abovementioned claims by means of a survey of eleven organizations in the Dutch construction industry. The result of this study indicates that higher organizational maturity results in improved financial performance. A relation was found, yet interestingly, this study found no linear relation between the variance of maturity levels and financial performance. In addition, this study found no proof of the relation between a dominant individual maturity model dimension and financial performance either. These findings imply that organizations that want to increase their financial performance need to invest in all maturity model dimensions to improve the organizational maturity level. Organizations can use these findings as a benchmark and incentive to increase the maturity of their organization.
1. INTRODUCTION
Managing continuous process improvement requires the management of both people and processes (Gillies and Howard, 2003). For organizations to survive in today’s rapidly changing environment, continuous improvement should be a core organizational process. Continuous improvement (CI) is an interesting topic for businesses and literature. Frequently cited papers, such as those written by Paulk et al. (1993), Lockamy and McCormack (2004) and Estampe et al. (2013), show continuous improvement as a part of maturity models. Maturity models today are applicable to over 20 domains, of which the majority consists of software development and software engineering maturity models (Wendler, 2012), which will be further elaborated upon in the theoretical background.
Literature on Supply Chain Management (SCM) has also altered maturity models in the last couple of years, according to Tarhan et al. (2016). Chan and Qi (2003) note that a steady stream of articles has been published about SCM literature, but SCM’s relation to financial performance, especially in the construction industry, does not receive adequate attention. Beelaerts van Blokland et al. (2012) measure financial performance in the aerospace and automotive industries (Beelaerts van Blokland et al., 2010) and state that, for further research, financial measures in the construction industry should be investigated as well. This interest in researching the construction industry relates to the – since the 1990s – increased interest in SCM theories in order to improve the coordination of the many subcontractors and suppliers involved in the construction supply chain (Segerstedt and Olofsson, 2010). Vaidyanathan and Howell (2007) add to this that one of the unique characteristics of the construction industry is - unlike in any other industry - the need for a strong inter-firm collaboration across the entire Construction Supply Chain (CSC) in order to be able to complete a project. This combination is what makes the collaboration between firms and supply chain alignment paramount. Vrijhoef and Koskela (2000) and Vaidyanathan and Howell (2007) regard the construction industry as the black sheep of the industries because it is seen as unable to adopt SCM techniques which other industries are capable of implementing (Dubois and Gadde, 2000). This is contrasted by Saad et al. (2002), which noticed an increasing number of companies in the construction industry that are beginning to adopt supply chain management techniques to improve performance. The unique supply chain characteristic of the construction industry, the necessity of inter-firm collaboration, provides the author of this thesis the possibility to compare maturity to financial performance measures. To enrich the theory and practical implications that result from the above-mentioned text, this thesis focuses on the relation between the maturity level of organizations in the construction industry and their financial performance.
This finding is supported by the Supply Chain Operations Reference (SCOR) model (Lockamy and McCormack, 2004); benchmarking schemes (Simatupang and Sridharan, 2004); significantly higher product quality (Harter et al., 2000); Supply Chain information integration (Trkman et al., 2007) and by Business Process Maturity (BPM) models that focus on their practical applicability and their usefulness (Röglinger et al., 2012). However, Lapide (2006) contests the link between supply chain performance and maturity levels. The authors stated above all name various performance indicators, but surprisingly little research has been done on the different levels of a continuous improvement maturity model and financial performance measures.
Much has been written about maturity models, which typically include a logical ordering of levels from the present state to maturity (Becker et al., 2009, Gottschalk, 2009, Kazanjian and Drazin, 1989). At first, the present level of maturity of a company is represented by the capabilities as regards a specific class of objects and the application domain (Rosemann and de Bruin, 2005). A maturity model typically describes a pathway aimed at systematically advancing business processes besides the maturity continuum (Skrinjar, 2008). In their paper, authors Lockamy and McCormack (2004) compare organizational maturity with qualitative performance metrics. They found an insignificant relation between SCM maturity and business performance in their research and suggested that the relation should be refined. Yet, despite the considerable number and increased reach of maturity models (Wendler, 2012) and the encouraging achievements of using maturity models in other domains (Paulk et al., 1993; Harter et al., 2000), the use of BPM models has not gained wide acceptance in research or practice yet (Tarhan et al., 2016). This is in contrast with research performed by Beelaerts van Blokland et al. that showed a connection between financial performance indicators (FPIs) (2012) and maturation in the aviation industry and automotive industry (2010). In the near future, these researchers will compare the construction industry to the aviation industry.
This paper strives to scientifically complement the findings of the aforementioned research on the connection between supply chain performance and maturity levels. Furthermore, this research will add to existing literature by researching the relation between financial performance and maturity levels of organizations in the construction industry. By proving that higher organizational maturity positively influences financial performance, one can conclude that investing in organizational maturity would pay off financially for a company and its supply chain partners.
increased financial performance. Additionally, results show that organizations that want to increase their maturity level should focus on all maturity model dimensions. This paper is not able to test if a dominant dimension exists that has more influence on financial performance than other dimensions due to the limited sample size of this thesis.
The structure of this research is as follows: chapter 2 gives the theoretical background followed by hypotheses. Chapter 3 describes the used methodology, and the results are listed in chapter 4. Chapter 5 provides the discussion and chapter 6 presents the final conclusions, the theoretical and managerial implications, the research limitations, and recommended areas for further research.
2. THEORETICAL BACKGROUND AND HYPOTHESES
This chapter will discuss the theoretical background of the thesis by, firstly, showing a perspective on continuous improvement as part of maturity models. Secondly, financial Key Performance Indicators (KPIs) (Beelaerts van Blokland et al., 2012) are introduced and the relationship between maturity models and KPIs is explored. And finally, the hypotheses that shape this study are provided.
2.1 Continuous improvement as part of maturity models
Business Process Management (BPM) is a contemporary management technique that focuses on managing an organization’s operations in terms of ‘business processes’ (Dijkman et al., 2016). CI can be seen as a part of these processes that are measured and structured sets of activities designed to produce a specific output for a particular customer or market (Davenport, 1993). BPM handles the methods, tools and techniques to identify, analyze, execute, monitor and change these business processes, resulting in a cycle of CI (Davenport, 1993). This is in contrast to the more traditional function-based management that focuses on customers and relations between activities with the goal of aiming to achieve higher customer satisfaction and a better collaboration between business functions (Dumas et al., 2013).
2.1.1 Maturity model explanation
linked to better performing processes and, more specifically, resulting in a better quality outcome. A large number of studies have supported this claim: Herbsleb and Zubrow, (1997); Herbsleb and Goldenson, (1996); Jiang et al., (2004); Krishnan and Kellner, (1999); and Krishnan and Kriebel, (2000). However, the main focus of most of these studies is on the maturity, and thus performance, of one single process of a business function (Dijkman et al., 2016), while BPM focuses on the collection of all business processes of the organization in general.
The concept of single process maturity proposes that a process has a lifecycle that is assessed by the extent to which the process is explicitly defined, managed, measured and controlled (Lockamy and McCormack, 2004). Additionally McCormack (2001) and Skrinjar et al. (2008) claim to provide some initial evidence that organizations focusing on BPM improvements perform better as a whole. A maturity level symbolizes a threshold which, when reached, institutionalizes a total systems view that is necessary for achieving a group of process goals (Dorfman and Thayer, 1997). The achievement of each maturity level establishes an increased level of process capability in an organization (McCormack et al., 2009). Furthermore, as processes mature they move from an internally focused departmental perspective to an externally focused supply chain system perspective.
Tarhan et al. (2016) and Rosemann et al. (2006) note that there is no scarcity of process maturity models in the current business environment. When talking about maturity models most people first think of the Capability Maturity Model Integration (CMMI) developed by Software Engineering Institute (SEI) in 2004. The CMMI (Figure 1) is claimed to be the best known (Estampe et al., 2013; Wendler, 2012) and most widely adopted maturity model (Eadie, et al., 2011). As one can see in Figure 1, the model has five maturity level stages. Its main goal is to provide guidance in developing or improving processes in order to meet the business goals of an organization (SEI, 2004). As the successor of the CMM, the generalization of improvement concepts makes the CMMI extremely abstract, which in turn makes it not as exclusive to software engineering as its predecessor.
--- Insert Figure 1 about here ---
The second component of maturity refers to the measured objects: the capabilities. This means that, according to Wendler (2012), maturity models have to define criteria for measurement, such as processes, conditions or application targets. Maturity models focusing on one criterion are so-called one-dimensional models. These days, most of the maturity models are multi-dimensional, including affected processes, organizational units and problem domains (Röglinger et al., 2012). To clarify the statements above, maturity models describe and determine the state of completion of certain capabilities. The application of this concept is not limited to a specific domain. Maturity models therefore define simplified levels or stages, which measure the completeness of the analyzed objects via different sorts of criteria. This explanation is related to the definition of Becker et al. (2009): “A maturity model consists of a sequence of maturity levels for a class of objects. It represents an anticipated, desired, or typical evolution path of these objects shaped as discrete stages. Typically, these objects are organizations or processes.” This definition reflects the idea of maturity models and serves as the foundation of the maturity models mentioned in this thesis.
2.1.2 Overview of wide range of maturity models
Over the past years, various authors (Röglinger et al., 2012; Becker et al., 2009; Estampe et al., 2013; Van Looy, 2014; Tarhan et al., 2016) studied diverse aspects and differences of the models and they compared numerous maturity models in their papers. Röglinger et al. (2012), Becker et al. (2009) and Estampe et al. (2013) agree on the usefulness of a maturity model: its characteristics, its applicability besides theory and the requirements needed. In her book, Van Looy (2014) constitutes a comparative study on the practical relevance of 69 business process maturity models (BPMM) and the historical background in order to become aware of the underlying practical implications. She acknowledges four tracks in BPMM history that are widely accepted and applied in various sectors. Among these four is the above-mentioned SEI track, including the CMM (Paulk et al., 1993) as well as the CMMI (SEI, 2004) and the OMG track focusing on BPMM (OMG, 2008). Furthermore the International Organization for Standardization (ISO) which includes the SPICE project as explained in sector 2.2.3 and at last the fourth track: the Federal Aviation Administration integrated Capability Maturity Model (FAA-iCMM) (Van Looy, 2014). Additionally, the objective of Tarhan et al. (2016) was to better understand the research on maturity models by conducting a systematic literature review of 61 BPMM studies between 1990 and 2014. Their research concluded that the validity and practical usefulness of these models is scarce (Tarhan et al., 2016) as most of the supply chain maturity models focus on specific, tangible characteristics that can easily be measured in processes, such as the BPM model that identifies five maturity levels (Lockamy and McCormack, 2004). In their paper, they complement earlier research of McCormack, namely McCormack and Johnson (2001). Lockamy and McCormack (2004) state that each level contains different characteristics associated with process maturity such as predictability, capability, control, effectiveness and efficiency. From the view that business process improvement requires the management of both people and processes (Gillies and Howard, 2003), it becomes evident that theoretical BPM maturity model aspects and BPM in practice are still out of alignment.
2.1.3 Maturity models in the construction industry
Supply chain relationship models were first introduced in other sectors before being introduced to the construction industry at a later stage. The SPICE project (Standardized Process Improvement for Construction Enterprises) is one of the first significant research efforts to adopt the CMM model for the construction industry (Sarshar et. al. 1999). After conducting a questionnaire, a case study and an expert panel survey, the researchers concluded that industry participants agreed on the need to develop process maturity models for the construction industry. The researchers concluded that, even though there was a lot of similarity between the software and construction industries, the CMM model could not be directly applied to the construction industry, mostly because CMM is applicable only for a single enterprise and it did not capture the multi-enterprise supply chain aspects of the industry (Vaidyanathan and Howell, 2007). One of the unique characteristics of the construction industry is that, unlike in other industries, a strong inter-firm collaboration is required across the entire construction supply chain in order to finalize a project. It requires internal alignment of the focal company whilst making use of the expertise of the buyers and suppliers in the supply chain (Segerstedt and Olofsson, 2010). Vrijhoef and Koskela (2000) characterize the supply chain in the construction industry as a typical make-to-order supply chain, with every project creating a new product or prototype.
As concluded from the SPICE study, there is a need to take into account the multi-company supply chain for the construction industry. Vaidyanathan and Howell (2007) support this view on multi-firm collaboration and name the three dimensions - functional, project and firm - a process needs to go through in order to gain maturity and to have complete operational efficiency. The authors propose a Construction Supply Chain (CSC) maturity model of four stages and a maturity level assessment that provides a strategy to progress alongside the model (Vaidyanathan and Howell, 2007). The desired state per stage is mentioned; hence no tangible process measurements are present.
relationships, which is in contrast to the unique construction characteristic of inter-firm collaboration with multiple firms, and therefore this attribute shows the weakness in those last three models. Meng et al. (2011) therefore propose a model that includes what is missing in the six models, which is an assessment method for the construction organizations. A matrix with four levels and eight dimensions was proposed, and it includes assessment criteria in a hierarchical structure. Meng et al. (2011) separate their individual dimensions in three sub-dimensions per dimension to propose the assessment method in greater detail to increase the practical applicability.
This likewise separation is also present in Scenter’s Performance Development model1, the successor of the Performance Improvement model by Spitsbaard and Fiksinski (2014). Their maturity model - visible in Figure 2 - entails six individual performance dimensions, where the level of internal dimension alignment determines the level of one of the five maturity level stages. The dimensions Process Optimization, Steering, Orientation, Teams, Leadership and Engagement respectively strike an organization’s hard and soft values and can be measured on a scale of five maturity levels. Röglinger et al. (2012) note that most maturity models provide limited guidance to identify maturity levels and to implement measures for improvement. Tarhan et al. (2016) support this by stating that measures to evaluate maturity are lacking when it comes to the distinction of a maturity model and an assessment model. The model proposed by Scenter (2017) includes these characteristics. The process performance dimension is described in other maturity models (Lockamy and McCormack, 2004) while steering is a dimension in the model of Vaidyanathan and Howell (2007) and the orientation dimension focuses on mission, vision and strategy. Berger (1997) suggests that improvement tasks can be integrated into the regular work of individual employees and teams, while Manor (2016) suggests that the type of leadership can have a major influence on individual employees and teams within the supply chain of the organization. At last, Singh and Power (2009) state that the engagement dimension is a strong, deep and meaningful part of aligning a company with customers and suppliers in order to improve performance.
The diversity of dimensions on which the performance level in the maturity model (Figure 2) is measured can be divided between hard (Process Optimization, Steering and Orientation) and soft dimensions (Teams, Leadership and Engagement). Each dimension offers a clear insight into which processes the company needs in order to focus on improving its maturity. The internal alignment of the six individual dimensions determines the maturity level of the organization. The author will use this maturity model because of the wide variety of variables and practical applicability. Furthermore, the model’s easy accessibility provides a well-defined insight into dimensions of companies that are subject to this research.
1
The model is adopted from the book that is currently being finalized and which will be published in
--- Insert Figure 2 about here --- 2.2 Financial key performance indicators
Business performance can be measured by a variety of different criteria. Tseng et al. (2009) state that a proper separation for measuring business performance can be made between financial measures and non-financial measures. Tseng et al. (2009) name competition performance, manufacturing capability and supply chain relationships as measures of non-financial performance. This paper focuses on FPIs used in the study done by Beelaerts van Blokland et al. (2012). They introduced these indicators to measure the Average Value Leverage (AVL) of an organization through a 3C-model (Figure 3) in the aviation industry. The selection of these indicators will be elaborated upon briefly.
The 3C-model is a result of papers written by Beelaerts van Blokland et al. (2010) and Beelaerts van Blokland et al. (2012). The model aims at integrating theories into one of value creation. The combined performance of each C (configuration, conception, and continuation) is seen as a positive driver for value creation. Configuration is seen as the way the supply chain is structured and how chain partners cooperate. Conception is the process of developing and designing products that are of value to the customer - this therefore relates to company innovation. At last, Continuation concerns the adherence of products to customer needs in order to create a company competitive advantage that leads to improved financial results, which allow the company to continue to exist. The 3C-model ideally entails sharing of innovation, production efforts and investments with supply chain partners, leading to increased gains for everyone involved (Beelaerts van Blokland et al., 2010). Together the three C’s drive for better innovation.
2.2.1 Configuration
implementing, despite this, theory argues that supply chain collaboration has proven to be difficult to implement (Sabath and Fontanella, 2002). Efforts should nevertheless be made, according to Clark et al. (1995), because suppliers are the co-investors and co-developers of new designs. Clark et al. (1995) identified project strategy variables that express the contribution of value by suppliers in man-hours. The employee emerges from this research as a basic indicator expressing value-leverage for co-development and co-production. The turnover per employee (T/E) emerged from the research by Clark et al. (1995) as indicator to measure value leverage on supply chain configuration.
2.2.2. Conception
Chesbrough (2003) found that the boundaries of companies become more open in order to increase their competitive advantage. Companies that still use closed innovation, which represents the outdated way of researching and development, fall behind on companies that use open innovation. As Chesbrough (2003) suggests, many corporations that once spanned the traditional routines of firm innovation have been replaced by supply chains, where other firms do some or nearly all of the production of goods and services (Petrick, 2007). Beelaerts van Blokland et al. (2012) call this open innovation and split it in two types, inbound and outbound open innovation. Outbound innovation seeks for external companies that are able to commercialize a technology rather than relying on the internal paths of innovation. Inbound innovation suggests that companies should use the research of others and their own marketing channels instead of relying on their own research and development department. According to the paper of Teece (1998), innovations and their learning opportunities are closely related. They state that when aspects for organizational learning change constantly these learning opportunities becomes severely restricted. Gassmann and Enkel (2005) come with a third type of open innovation, which combines both, outbound and inbound innovation and which transfers the knowledge through the company. They state that open innovation should be one of the key characteristics of an integrator in a supply chain network. Research done by Dubois and Gadde (2002) describes the construction industry as unable to adopt techniques that have improved performance in other industries. It has been argued by Gann (1996) that the short-term perspective of the construction industry stimulates sub-optimization and obstructs innovation and further technical development (Dubois and Gadde, 2000). Since technology has such an enormous impact on society and the economy, a company’s ability to continuously innovate its products and business model is essential for its future success and thus survival (Menzel et al., 2007). The research and development per employee (R&D/E) emerged from this view as an indicator for the conception of new products or techniques by innovations.
2.2.3 Continuation
supply and demand side. The value of integrating members throughout supply chains has been identified and studied in various industries (Simatupang et al., 2002). Hakansson and Snehota (1989) state that the effectiveness within a business organization thus depends on the capacity to obtain resources through trade with other parties. It is therefore important to mention that the configuration of the network system is between the organization and the other parties rather than within the organization, which are, according to Hakansson and Snehota (1989), the determinants of the bargaining/competitive position and the overall effectiveness of the organization in achieving its goals. Moller and Svahn (2003) argue that, by developing specific networking capabilities, firms are able not only to transfer complex knowledge but also to co-create new resources through intentional business nets. These capabilities arise from the coordinated activities of groups of people that pool their individual skills and assets – this way, an employee emerges as a factor that influences value leverage. According to the research of Lovell (2007), the employee represents the value driver to develop and establish relations and exchange knowledge for continuous customer value creation. From the value network perspective, Lovell (2007) identified the effectiveness of people/employee in terms of profit per employee (P/E).
Concluding these three C’s, the following three FPIs were discovered: 1. Turnover per employee (T/E)
2. Research and development per employee (R&D /E) 3. Profit per employee (P/E)
These key performance indicators, or value leverage indicators as Beelaerts van Blokland et al. (2012) call them, allow for a comparison of how companies perform individually from a financial perspective and to investigate the relation to the level of organizational maturity model. The above-mentioned FPIs are used in the aerospace industry and are thus seen as sufficient to measure financial performance in the construction industry. As in the paper written by Beelaerts van Blokland et al. (2012), the 3C value leverage model is sustained by financial oriented variables. These include (1) turnover, (2) operating profit, (3) R&D expenses and (4) employees to calculate the AVL. These measures are further elaborated upon in chapter 3.
--- Insert Figure 3 about here --- 2.3 Hypotheses
illustrated in this paper’s research framework (Figure 4). As stated before in chapter 2.1, continuous improvement can be seen as an underlying mechanism of a maturity model. The order of the model’s levels forms a logical sequence from the present state to organizational maturity. Organizations that score low on maturity levels have their focus on figuring out day-to-day issues and on tasks that need to be performed at that moment. Higher maturity levels focus on integration throughout the company and on intercompany collaboration to bundle buyer and supplier expertise in order to continuously improve. By knowing the level of maturity, a company can organize the efficiency of its dimensions in order to improve internal performance. In contrast, low maturity levels work uncoordinated and reactive without having a future goal in mind. To conclude, a company has to focus on internal alignment first before gaining organizational maturity.
Previous research by Beelaerts van Blokland et al. (2012) shows a connection between FPI and maturity in organizations in the aviation industry. Their research suggestion to conduct future research with FPIs in the construction industry will be carried out. One can understand from the lean perspective that waste elimination and fewer inventories, whilst producing top quality products, lead to a better financial performance. The internal alignment while using buyer and supplier expertise for the focal company has been proven to function in other industries (Segerstedt and Olofsson, 2010). Hence Dubois and Gadde (2000) see the construction industry as unable to adopt techniques that other industries are capable of executing. Multiple studies have shown that a higher maturity model level leads to increased performance; as a reduction in development cycle time (Lockamy and McCormack, 2004), higher development productivity (Jiang et al., 2004), improved project performance (Lockamy and McCormack, 2004) and organizational maturation furthermore reduce conflict and encourage cooperative behavior (Vaidyanathan and Howell, 2007). Tseng et al. (2009) split performance in financial and non-financial measures and, as the above-mentioned authors formulate improvements in non-financial measures due to a higher maturity level, this assumes that it also holds for financial performance. Due to this assumption, the suggestion is made that a higher maturity level would lead to substantially better financial performance, which would be in line with findings in the aviation industry (Beelaerts van Blokland et al., 2012). Or, in other words:
Hypothesis 1: A level difference in the maturity model ranking of a company would lead to a higher AVL financial performance score.
the business process management maturity model of Rosemann et al. (2005), has different dimensions that would imply that dimensions do not matter for the level of maturity performance. In the paper of Tarhan et al. (2016), 61 papers were investigated, each containing different maturity models in the area of business process management. Nesensohn et al. (2015) formulate eleven key attributes in their paper’s maturity model, while Vaidyanathan and Howell (2007) only use three dimensions for their so-called construction maturity model, and Van Looy et al. (2013) focuses on six capability areas. This suggests that dimensions are submissive to the maturity model and there is not a dominant dimension. This research will put all dimensions to a test and states that a dimension from Figure 2 does not have more influence on the maturity level and the relation to financial performance than the figure’s other dimensions. This can be formulated as a hypothesis:
Hypothesis 2: An individual dimension of the organizational maturity model does not have more influence on the AVL financial performance score than other dimensions.
Where one could easily believe Hypothesis 1 to be true, no evidence for this has been found. After thorough literature research on the subject of individual dimensions in relation to financial performance, it seems that no study has done research on Hypothesis 2. The newness of this study therefore lies in providing support for Hypothesis 1 and, while doing so, exploring Hypothesis 2. The hypotheses are shown alongside information in the research framework below (Figure 4).
--- Insert Figure 4 about here ---
3. METHODOLOGY
This chapter will discuss the methodology of this thesis by, firstly, showing a perspective on the research design of this thesis. Secondly, the data collection process is explained. Thirdly, the scale of measurements used in this thesis is explained. And finally, the data analysis plan that shapes the results of this study is provided.
3.1 Research Design
this research. The advantages of a survey method are clear: (1) obtain structured primary data with (2) a high degree of reliability, which allows for testing the hypothesized relationships or the differences between other contexts (Karlsson, 2009). By using a survey instead of semi-structured interviews a larger range of respondents can be approached. The survey approach is less time-consuming for the respondent and the required data that relate to maturity level per company can be obtained and used to distinguish respondents. The structured data, in contrast to data obtained through case studies, shows a better fit with the testing nature of the hypotheses of this research. The choice for structured questionnaires will also lead to high reliability since it is easy to repeat the research.
3.2 Data Collection Process
At the beginning of the data collection process, companies in the construction industry were contacted and they were asked to participate in the present study. Some supervisors were contacted directly and they received information about this study’s goals. An email was also sent to the company’s general e-mail address in order to reach other supervisors that could not be contacted directly. A company-specific introduction was provided to each participant to explain the usefulness for the company to participate in the study. If they agreed with participating in this study, the supervisors had to complete online questionnaires. All participants received the same online questionnaire. The participants answered questions about the dimensions displayed in Figure 2 (see page 11). All the responses on the questionnaire have been treated confidentially.
The top 50 companies in the construction industry in the Netherlands have received the questionnaire and four questionnaires were sent to companies in the construction industry known through a personal network. The initial response was low with eleven responses three weeks after sending the questionnaires. After sending a reminder this number increased to fourteen responses. From the fourteen received questionnaires, twelve were filled in completely. Of the two remaining questionnaires more than 75% of the data was missing. Methods of missing data imputation, such as mean replacement, would significantly bias the outcome results. In such cases, Hair et al. (2010) recommend to only include complete cases in the analysis. At last, one company was not willing to provide the annual company report, which is necessary to compare the maturity level to the AVL financial performance score. Therefore, this study continued with the eleven completed questionnaires.
average response rate can be due to the usage of an online survey, which typically shows a lower response rate compared to a survey by phone or by mail. Furthermore, not all email addresses of individual contacts were available in which case the survey was addressed to a company’s general e-mail address. Because of this, the questionnaire might not have found the right person or it got lost organizations. The average response rate of 20% is not uncommon for Internet questionnaires. Shih and Fan (2009) conducted an analysis on response rates of Internet studies; from the 35 surveys used for their study, twelve surveys were found to have an equal or lower response rate.
The number of incomplete surveys (4%) of this study can be a threat to the internal validity of the questionnaire. There is a possibility that respondents did not complete the survey because the questions were not understood, which raises the question if the respondents who did finish the survey actually understood the questions. Experts did not encounter any problems understanding the survey when they checked the questions before the questionnaire was sent. Furthermore, the questionnaire was addressed to members of the higher management or managers who can be expected to have knowledge of and be familiar with the concepts that this survey addressed. To ensure whether it was the actual manager who filled in the survey, questions were asked to obtain information on the function of the interviewee and the amount of years he/she is active in the company. At last, in order to triangulate the data obtained, a report with the outcomes of the survey was sent to the respondents with additional questions from the researcher. Altricher et al. (2008) formulate that triangulation provides the author with a more complete view on the situation. The respondents were able to change mistakes and ask further questions. There were no respondents that asked study-related questions. The entire questionnaire can be found in Appendix B.
To investigate the relation between the maturity level and the financial KPIs, all respondents were asked to provide their annual company report of the fiscal year 2015. For most of the companies the annual reports were available through the Internet. The reports that were not available online were received via email once the survey was completed. Data needed to be subtracted from the annual report in order to triangulate configuration, conception and continuation, which led to variable 4 shown below (see chapter 3.2.4 Average Value Leverage financial indicator). The AVL indicator originates through the first three variables introduced below. These first three variables are integrated in the fourth variable, AVL, which will be used as the KPI of financial performance in this study. 3.2.1 Configuration
leverage its assets across the supply chain. As construction companies heavily rely on their supply chain partners, the turnover generated through the supply chain is used as a measure for leverage in the supply chain. Configuration can be calculated by using Variable 1. For the denominator employee it needs to be noted that only full-time employees, and not part-time employees or interns, are taken into account. This holds for all the employee denominators of Variable 1, Variable 2 and Variable 3.
(Variable 1)
3.2.2 Conception
The average investment expenses in the variable Research and Development per employee are formulated as the conception of a company. The expenses on R&D divided by the number of employees provide insights into the innovative focus of a company (Pam, 2010). Variable 2 is formulated below.
(Variable 2) 3.2.3 Continuation
This indicator shows the extent to which employees of a firm are able to generate customer value. An employee is seen as the value drive that develops relations, builds networks with supply chain partners and exchanges knowledge to create value for the customer. The variable profit per employee indicates to what extent a company is able to attract demand from the market for the products they produce. Lovell (2007) states the importance of the profitability and continuity of a company which is determined by customer needs on one the hand, while network building in the supply chain can, on the other hand, be measured by the indicator P/E. Operating profit is used to formulate the continuation outcome for all analyzed companies.
(Variable 3)
3.2.4 Average Value Leverage financial indicator
The relation between the three above-mentioned variables expresses the capability of a company to leverage value (Beelaerts van Blokland et al., 2012). This is done through the average of demand, supply and new value creation through R&D to ensure the continuity of company operations. The AVL (Variable 4) as used in the paper of Beelaerts van Blokland et al. (2012) of the analyzed construction companies will be calculated and compared with the maturity level of the company retrieved from the questionnaire.
These variables are used to measure financial performance of an organization. The financial figures are compared to the outcomes of the questionnaire in order to test the relation between financial performance and the maturity model. Section 3.3 provides more information on the data from the questionnaire.
3.3 Survey Measurement Scale
As indicated before, a structured questionnaire was the basis of data collection for this research (Appendix B provides the complete questionnaire). At first, general questions were asked to indicate the function of the respondent and the active working years for that specific company. Then questions were formulated that result from the dimensions of the Scenter model (Scenter, 2017) shown in Figure 2. Each dimension consists of multiple questions and was measured on a five-point ordinal Likert-type scale. The answer options of the ‘hard’ dimensions (Process Optimization, Steering and Orientation) were formulated based on SMART principles (Doran, 1981). The options were formulated specifically for the respondent to select a predefined and measurable outcome for the specific question rather than a decision made on gut feeling. Wendler (2012) suggests predefining options, which helps define criteria for measurements like processes or targets. This formulation therefore also increased the construct validity for this research. For the ‘soft’ dimensions - Team, Leadership and Engagement - a Likert-type scale of (1) Totally Disagree to (5) Totally Agree was used since the outcomes on these dimensions were too intangible to be quantified. The output of this questionnaire is analyzed accordingly with the procedure of the data analysis described in section 3.4.
3.4 Data Analysis Plan
4.90-5.00. The author believed that scores higher than .90 were seen as sufficient for a higher maturity level. The author inserted this boundary because of the definition of Becker et al. (2009) mentioned in section 2.1.1. This definition formulates that the maturity model builds on the previous level and follows an evolutionary path to the desired maturity stage. To test the first hypothesis the maturity level per respondent is taken as the independent variable. To test the second hypothesis the individual dimension scores per company were taken as the independent variable.
Unfortunately, due to the limited sample size of eleven organizations, real testing is not possible. The author, therefore, will provide conjectures and assumptions regarding the analyzed data. The sample size is too small to perform a regression analysis and therefore a descriptive statistics analysis will be performed. The descriptive statistics analysis will provide the author with information regarding the spread of samples. The outcome of the descriptive statistics analysis provides the author with data on means, standard deviation, median, minimum, maximum, range and coefficient of variation. Additionally, an exploratory data analysis will be performed. Due to the limited sample size conjectures are formulated regarding the relation between the outcome of the dependent and independent variables. Data plots are provided for the visualization of the reader and provide additional input for the conjectures formulated in sections 5 and 6, discussion and conclusion, respectively.
4. RESULTS
This section summarizes the results of this research. Figure 5 provides a visualization of the AVL performance score in euros on the y-axis and the maturity level of the companies in the sample group on the x-axis.
Figure 5: Scatter spread on respondents AVL score per maturity level
Figure 5 shows a spread between the maturity and the AVL performance score. Additionally, the author is aware that due to the limited sample size a conclusion regarding a relation between variables cannot be quantified. The AVL-score spread of all respondents is further visualized in the boxplot of Figure 6 below. Figure 6 informs one on the wide spread of AVL score that is obtained by the respondents’ maturity levels two, three and four. Additionally one can observe that there is a larger variance between the higher AVL scores after the median in comparison to the lower AVL scores.
Figure 6: Boxplot on AVL Score All Respondents
Additionally, Figure 5 shows the respondents AVL-score for their obtained maturity level. From this scatter plot one can identify the spread per maturity level and furthermore a group of respondents that are close to each other. At last, to formulate conjectures on Hypothesis 1 a descriptive statistics analysis was used, as previously explained in the data analysis plan. The outcome is presented in Table 1.
50000 75000 100000 125000 150000 175000 200000 225000
AVL Financial Score
H1 Descriptive Statistics
Mean
131086
Standard Deviation
43909
Median
111896
Minimum
57411
Maximum
200328
Range
142917
Coefficient of Variation (CV)
0.33
Table 1: Statistics on respondents.From the descriptive statistics in Table 1 additional information regarding the sample size can be derived. The CV tells us that there is a limited variability in the sample size, due to the fact that the CV < 1. What also is worth noting, from Figure 5, is the increasing AVL score per maturity level. As mentioned before, the sample size restrains the author for testing the first hypothesis. Yet the author conjectures that due to the increasing AVL score per maturity level one can observe from Figure 5, there is a relation that a higher organizational maturity level leads to an increased AVL score. By taking into account the figure and the table above, the author assumes the first hypothesis - A level difference in the maturity model ranking for a company would lead to a higher AVL financial performance score – to be true.
To formulate conjectures regarding Hypothesis 2, a descriptive analysis was performed, as mentioned in the data analysis plan in section 3.4. The outcome of this second analysis is summarized in Table 2, which is presented below. Due to the previously mentioned constraint of the sample size no regression could be performed. This means that the influence of a specific dimension cannot be compared to the dependent variable AVL Performance, implying that no reasonable statement can be provided that one dimension has more influence on the AVL performance score than other dimensions. Therefore a conclusion on Hypothesis 2 - that a single dimension does not have more influence than the other dimensions - cannot be accepted nor rejected. A boxplot presented in Figure 7 including the average score per respondent per dimension is provided and will be further discussed in the conclusion and discussion section.
Optimization Steering Orientation
Process
Teams
Leadership Engagement
Mean
2.76
3
3.07
3.32
3.48
3.86
Standard Deviation
0.78
0.76
0.69
0.56
0.58
0.44
Median
3.16
3
3
3.33
3.44
3.97
Minimum
1.44
1.22
2.22
2.33
2.89
3.07
Maximum
Range
3.56
2.12
2.78
4
1.78
4
4.22
1.89
4.56
1.67
4.67
1.60
Coefficient of Variation
0.28
0.25
0.22
0.17
0.17
0.11
From Table 2 one can conclude that the Mean of the ‘soft’ dimensions Engagement, Leadership and Teams obtain a higher score than a ‘hard’ dimension like the Process Optimization dimension. Yet, due to the testing nature of this research no conclusion regarding the relation between an individual dimension and AVL performance can be delivered. This observation will be further elaborated upon in section 5.3.
Figure 7: Boxplot variance of respondents average dimension score
As expected, the abovementioned graphs and tables predict a positive relation between maturity levels and the AVL performance indicator. Unfortunately, due to the testing nature of the descriptive statistics, no evidence or predictions have been found between the relational influences of a single dimension having a more dominant influence on the AVL performance indicator than the other dimensions.
5. DISCUSSION
5.1 Organizational Maturity and AVL
First of all, the results presented in Figure 5 indicate an increasing outcome between a higher maturity level and AVL performance. Due tot the limited sample size this cannot be tested but this can be observed from Figure 5. This is in line with the outcomes that were found in other studies, such as the confirmed relation that companies with a higher organizational maturity level outperform companies that have not reached a higher level (McCormack and Johnson, 2001). Research and surveys by McCormack and Lockamy (2004) proved that processes have a maturity life cycle and that there is a correlation between improving process maturity and non-financial business performance. Additionally, analyses by Harter et al. (2000) suggest that higher levels of maturity are associated with significantly better quality output performance. Furthermore, Dijkman et al. (2015) contributed to theory a relation between innovativeness and performance as a property of organizational maturity. All these findings of previous research are in line with the findings of this study: a higher organizational maturity level leads to better performance.
However, in this study, a prediction on the link between organizational maturity and organizational AVL financial performance is provided, whereas previous research shows relationships between maturity and different sorts of process improvements. Thus, this research provides an original and valuable contribution to the present frame of literature by adding financial performance to the spectrum of performance-investigated studies, which conjecture to be positively related to organizational maturity.
finding on the information sharing ability of the construction industry provides a more valuable and different insight into the relation between maturity levels and financial performance than other industries.
Unfortunately, none of the companies that participated in this study scored a maturity level of 1 or 5, therefore the relations of these levels to AVL performance still remains unknown. From the data in the figures, one can conclude that organizations that succeed in improving their organizational maturity level result in an increased AVL performance. For organizations to achieve a higher organizational maturity level, their focus needs to be on increasing all organizational maturity model dimensions. Furthermore, the descriptive statistics from Table 1 show a large variance between the range of minimum and maximum observed AVL-score. Additionally, the statistics from the CV inform one on the small variability in this sample and the outcome of 0.33 derived from Table 1 explains the 33% variability outcome of the respondents. The small variability in the sample provides us with a better understanding of the first hypothesis of this paper. As mentioned above, the differences in the outcomes shown in Figure 5 allow readers to comprehend different maturity levels and the respondents’ respective AVL performance. Due to the limited sample size, this relation cannot be tested, allowing the author to use the observations of the data from Figure 5 to provide conjectures on Hypothesis 1 that a positive relation exists.
5.2 Individual dimensions and AVL
that are ahead of the bottleneck. An organization should focus on improving the least performing dimension in order to evolve to the next maturity level, which is assumed to relate to a higher AVL financial performance.
5.3 Additional Findings
To investigate the research from a different perspective, the author inspected the individual data one more time. What can be observed from the AVL data, and what is clearly shown in Figure 8, are the outliers in the dataset, shown in the red circles, and especially the outlier of maturity level 4. Most of the respondents per maturity level are located close to each other and the outlier of maturity level 4 with the red circle in Figure 8 appears to be around half the AVL score of the outcome of the other two respondents that scored maturity level 4. This outlier has a certain influence on the variation in the respondents’ outcome variation and spread per maturity level. To analyze this outlier even further, the author took a closer look at the outcome of the general questions of this respondent. The CEO of the company completed the questionnaire, which could imply two different situations. Firstly, the CEO filled out the questionnaire honest and without the bias of his own company. Secondly, the CEO is likely to have an optimistic view of his company and answered the questions with a bias where a better score was obtained than if another employee had filled out the questionnaire. The author is aware that this is a limitation of this study and will elaborate more on this subject in section 6.2.
--- Insert Figure 8 about here ---
score than the hard dimensions. The second scenario implies that the respondents filled out higher answers over the course of the questionnaire, which would explain why the average score of the soft dimensions is higher than the score of the hard dimensions. These conjectures cannot be supported by a calculation, yet the author can provide recommendations to bypass these scenarios for future research. Recommendations regarding future work on these scenarios are given in section 6.2.
6. CONCLUSION
The survey conducted for this paper provides conjectures on the two research questions of this research: (1) Is there a difference in the relation between levels of company maturity and financial performance? And (2) Does an individual dimension of a maturity model have more influence on financial performance than other dimensions? The conjectures to these questions were found through a survey held among eleven organizations in the Dutch construction industry.
6.1 Findings
Nowadays, organizations are dealing with a rapidly changing environment. To keep up with these fast changes, organizations need to focus on continuous process improvement to try and stay ahead of the competition. BPM is a contemporary management technique that focuses on managing an organization’s operations in terms of ‘business processes’ (Dijkman et al., 2016). When organizations engage in business process management, the management will ask itself at some stage if the organization benefits from it.
Maturity models first appeared as a measure for evaluating the capabilities of an organization in a particular discipline. Over the course of years, many different maturity models were created that not only focused on specific disciplines but also covered the maturity of an entire organization. An organizational maturity model, as proposed by Scenter (2017), typically represents an anticipated, desired or evolutionary path where objects shaped as discrete stages evolve into the desired maturity stage. An analysis in the construction industry was performed on the levels of this maturity model in relation to financial performance indicators introduced on page 11. The organizational Conception, Configuration and Continuation were seen as measures able to indicate financial performance. The AVL performance indicator that was derived from these three measures was used as the dependent variable in this research.
outcome of the conjectures of this study provides insights into the theory that a company that performs at maturity level 3 performs on average better on AVL financial indicators than a company at maturity level 2. Moreover, as shown in Figure 5, a company that performs at level 4 on average has a better AVL financial performance than a company at maturity level 2 and 3, respectively. None of the respondents were found to have maturity level 1 or 5, and therefore no conclusions of the hypotheses can be related to these maturity levels. These findings in the construction industry are similar to those in the aviation industry in providing a conjecture that a positive relation between maturity and financial performance exists. The main differences that were found were financial differences between the aviation industry and the construction industry. A good reason for this may be the difference in initial investments of an order and the nature of supply chain integration that is different between the aviation industry and the construction industry. The unique make-to-order characteristic of the construction industry is an example of one of these differences. Furthermore, next to the theoretical conjectures these findings offer practical insights for managers. The outcome of this thesis provides one with the conjecture that it pays off financially for a company to develop the maturity model dimensions in order to evolve to the next maturity level.
This research also investigates the hypothesis “Does an individual dimension of the maturity model have more influence on AVL performance than other dimensions?” The findings of this research cannot provide a conclusion regarding this hypothesis, which tells us that the conjecture that there is not a dominant dimension organizations should focus on to improve AVL performance. While this study focused on the 50 largest construction companies in the Netherlands, it is acknowledged that smaller construction companies are affected by the findings in this paper as well. For future research, smaller organizations should consequently be included as well. This is further elaborated upon in section 6.2.
6.2 Limitations and future work
construction companies beyond the Dutch border. This will expand the sample size of the respondents and additionally provide the possibility to compare the construction industries in different countries. An additional limitation of this study is that for each company only one respondent was questioned. It, therefore, is possible that the answers contained a bias as explained in the additional findings of section 5.3. For future research, it is recommended to provide a questionnaire to multiple members of the organization to reduce this bias. While it is unknown if a bias occurred it is hard for one respondent to answer questions on behalf all employees of the respondents’ organization. Especially since, as mentioned in section 4, the active workforce in the respondents’ organizations ranged between 158 and 15,154 for the smallest and largest organizations respectively.
This study focuses on the absolute financial measures used in the Dutch construction industry and their relative AVL performance change related to various maturity levels. A limitation of this study is that this research did not investigate the construction industry in other countries than the Netherlands. As mentioned above, for future research it can be valuable to provide insights into similarities and/or differences between Dutch and foreign construction industries. Potential differences could be significant to investigate because the hard and soft dimensions are most likely differently shaped in different countries. This relation can be investigated in greater depth in future research.
Furthermore, a limitation of this study is the absence of control variables, which typically are variables that are related to the dependent variable. What can be of interest for future research is adding control variables such as company age, organization size, stock market listed organization and industry/sector specific characteristics like housing or dredging. These examples of control variables provide an additional insight into the specific outcome of AVL performance. Moreover, industries change constantly and fluctuate due to economic opportunities and setbacks. For future research it could be of interest to investigate the AVL financial performance per company for a number of years and conclude if a trend occurs due to external characteristics.
6.3 Implications
Nevertheless, the findings in this study provide an insight into the practice that these efforts, for companies in the Dutch construction industry, are likely to pay off.
For Scenter, the founding fathers of the maturity model (Figure 2) presented in this study, these insights have significant implications. Their guidance and consultation on changes for each dimension can help companies to advance through a series of maturity levels in order to achieve a better AVL performance. For organizations, these first conjectures imply that it is financially rewarding to invest in the supervision of the route to improvement of the maturity model dimensions. The initial question, “Can proof be provided that an increase in maturity level leads to increased financial performance?” has hereby not been answered yet completely. This means that, by delivering this paper, the experts of Scenter can provide an initial conjecture to companies that their guidance in improving process dimensions pays off financially for their organization. The goal for Scenter should be on providing proof for this conjecture by investigating this topic alongside the suggestions for future work presented in section 6.2.
7. REFERENCE LIST
1. Akintoye, A. McIntosh, G. & Fitzgerald, E. (2000). “A survey of supply chain collaboration and management in the UK construction Industry.” Journal Purchasing & Supply Management. 6(3): 159–168.
2. Altrichter, H. Feldman, A. Posch, P. & Somekh, B. (2008). Teachers investigate their work;
An introduction to action research across the professions. Routledge. p. 147. (2nd edition). 3. Arnold, U. (2000). “New dimensions of outsourcing: a combination of transaction cost
economics and the core competencies concept”. European Journal of Purchasing & Supply Management. 6: 23-9.
4. Becker, J. Knackstedt, R. & Pöppelbuß, J. (2009). "Developing Maturity Models for IT Management – A Procedure Model and its Application". Business & Information Systems Engineering (BISE). 1(3): 213-222.
5. Beelaerts van Blokland, W. Elferink, N. & Curran, R. (2010). Setting up a Company Performance Measurement Methodology for the Aerospace Industry: Deduction from the Automotive Industry. 10th AIAA Aviation Technology, Integration, and Operations (ATIO) Conference, 1(September), 1–15.
6. Beelaerts van Blokland, W. B. Fiksiński, M. Amoa, S. Santema, S. Silfhout, G. V. & Maaskant, L. (2012). Measuring value-leverage in aerospace supply chains. International Journal of Operations & Production Management. 32(8): 982-1007.
7. Berman, S. L. Wicks, A. C. Kotha, S. & Jones, T. M. (1999). Does stakeholder orientation matter? The relationship between stakeholder management models and firm financial performance. Academy of Management Journal. 42(5): 488-506.
8. Briscoe, G.H. & Dainty, A.R.J. (2005). “Construction supply chain integration: An elusive goal?” Supply Chain Management: An International Journal. 10(4): 319–326.
9. Briscoe, G.H. Dainty, A.R.J. Millett, S.J. & Neale, R.H. (2004). “Client-led strategies for construction supply chain improvement”. Construction Management and Economics. 22(2): 193-201.
10. Bruin, T. De Rosemann, M. Freeze, R. & Kulkarni, U. (2005a). Understanding the main phases of developing a maturity assessment model. In Australasian Conference on Information Systems. 8–19.
11. Chan, F. T. S. & Qi, H. J. J. (2003). An innovative performance measurement method for supply chain management. Supply Chain Management: An International Journal. 8(3): 209– 223.
13. Choi, T.Y. & Krause, D.R. (2005). “The supply base and its complexity: implications for transaction costs, risks, responsiveness, and innovation.” Journal of Operations Management. 24(5): 637-652.
14. Crosby, P. (1979). Quality is free. New York: McGraw-Hill.
15. Davenport, T. (1993). Process innovation: reengineering work through information technology. Boston: Harvard Business School Press.
16. Dijkman, R. Lammers, S. V. & de Jong, A. (2015). Properties that influence business process management maturity and its effect on organizational performance. Information Systems Frontiers. 18: 717-734.
17. Doran, G. T. (1981). "There's a S.M.A.R.T. Way to Write Management's Goals and Objectives". Management Review. 70(11): 35-36.
18. Dubois, A. (2003). “Strategic cost management across boundaries of firms”. Industrial Marketing Management. 32: 365-74.
19. Dubois, A. & Gadde, L.E. (2000), “Supply strategy and network effects - purchasing behaviour in the construction industry”. European Journal of Purchasing & Supply Management. 6(3-4): 207-15.
20. Dubois, A. & Gadde, L.E. (2002). The construction industry as a loosely coupled system: implications for productivity and innovation. Construction Management and Economics, 20(7): 621–631.
21. Dumas, M. La Rosa, M. Mendling, J. & Reijers, H.A. (2013). Fundamentals of business process management. Berlin-Heidelberg: Springer.
22. Eadie, R. Perera, S. & Heaney, G. (2011). Key process area mapping in the production of an e-capability maturity model for UK construction organisations. Journal of Financial Management of Property and Construction. 16(3): 197- 210.
23. Ellinger, A.E. (2002). Improving marketing/logistics cross-functional collaboration in the supply chain. Industrial Marketing Management. 25: 85-96.
24. Estampe, D. Lamouri, S. Paris, J. L. & Brahim-Djelloul, S. (2013). A framework for analysing supply chain performance evaluation models. International Journal of Production Economics. 142(2): 247-258.
25. Gann, D. (1996). Construction as a manufacturing process? Similarities and differences between industrialized housing and car production in Japan. Construction Management and Economics. 14: 437–50.
26. Gassmann, O. & Enkel, E. (2006). Towards a theory of open innovation: three core process archetypes. R&D Management Conference.
