Strategic Planning in SMEs: A Case Study in Indonesia

Abstract

Hotels drive the growth and development of tourism. Despite their important role, many hotels are small and medium-sized firms (SME) that are struggling to survive against fierce competition. Experts agree that strategic planning is vital for SME survival, but it is not wholly applicable for SME managers. Meanwhile, Mintzberg's concept of crafting strategy offers a more productive insight into SME strategic planning, but its abstract nature has historically discouraged empirical research on its practical benefits. This study will be the first to empirically explore the operationalization of Mintzberg's crafting strategy characteristics, and analyze its influence on organizational learning using structural equation model. Using a sample of 50 hotels in Bali, Indonesia, this study reveals that managing pattern and stability, detecting discontinuity, and knowing the business have a positive but weak effect, whereas reconciling change and continuity proves to have a positive and significantly strong effect on organizational learning. This study has bridged the gap between the abstract concepts of crafting strategy, which is a potentially better approach for SMEs, with daily operational practices. This study also proves that Mintzberg's approach can be used to predict organizational learning. This relationship is crucial since previous studies concluded that organizational learning improves company performance.

1. Introduction

Hotels are one of the links that drive the growth and development of tourism business employment (Le et al., 2020), as well as the development and preservation of community arts and culture (Suwithi Ni Wayan, 2013). Approximately 95% of accommodation types in Bali are guest houses and bed & breakfast accommodations which are mostly small to medium-sized firms (SME). Between 2015 to 2019, the number of SME hotels grew above the industry rate although the growth in number of rooms is much lower (Indonesian Hotel and Restaurant Association Report, 2019). Small and medium-sized hotels in Bali struggle to survive. Preliminary interview with hotel owners and managers yielded the following facts:

• Competition is extremely fierce. International chain hotels are only next door, located even on small roads that were previously dominated by local hotels. The inability of local hotels to compete can be observed through the drop or stagnation of the average selling room rate.

• If this condition persists, it can eventually lead to bankruptcy where revenue can no longer cover operational costs. The last option will be to eventually close the business and sell the property to waiting international chain hotels.

Despite their many contributions, SMEs are plagued by high failure rates and poor performance levels (Jocumsen, 2004). While strategic planning (SP) is deemed essential for the survival of tourist destinations (Ritchie, 1999), along with other market stakeholders such as SMEs (Berry, 1998), it received relatively small attention in literature and empirical research (Phillips & Moutinho, 2014). Kraus et al. (2006) performed a literature analysis for the last 25 years and found a low number of empirical researches on strategic planning in small firms. This study aims to empirically explore the operationalization of Mintzberg’s crafting strategy concept on SME and whether it has a positive influence on organizational learning, which is also the novelty of this study.

2. Literature Review and Hypotheses

2.1. Classification of Strategy Approaches

In an attempt to understand the big picture, Mintzberg and Lampel (1999) catalogued existing strategy formation theories and classified them into ten streams of strategy approaches or the so-called Ten Schools of Strategy consisted of design, planning, positioning, entrepreneurial, cognitive, learning, power, cultural, environmental and configuration school. Each schools convey somewhat different point of view from their theorists.

The later and newer approaches to strategy formulation are nothing more than a blend of two or more of the Ten School’s concepts. Later, Whittington (2002) classified strategy approaches into four streams: the classical approach prioritizes profits as the highest goal of business through rational planning, the evolutionary experiments to achieve, the processual contends that firms consist of a coalition of different ideas instead of pursuing a single objective, systemic perspectives believe that strategy depends on the strategist’s social background. The four approaches differ fundamentally along two dimensions: outcomes and the processes by which they are achieved (Phillips et al., 1999).

2.2. Strategic Planning

There are several overlapping and consistent definitions of strategic planning in literature. Wolf and Floyd (2017) summed up strategic planning (SP) as less formalized, periodic process that provides a structured approach to strategy formulation, implementation, and control. SP serves to influence an organization’s strategic direction for a given period and to coordinate and integrate deliberate as well as emerging strategic decisions.

Studies into SMEs suggest that, ceteris paribus, a key determinant of business success lies in the absence or presence of SP (Wang et al., 2007). The rationalist perspective – supported by authors like Ansoff, Chandler, Taylor, and Sloan – claim that strategic planning shall be a formal, rational process, carried out by the top management and a staff of strategic planners, who ultimately deliver a plan that is to be implemented straightforwardly as it is formulated (Ghezzi et al., 2010). However, Henry Mintzberg introduced a Crafting Strategy (CS) concept that tackles SP from a different perspective. He emphasized that an organization may have a pattern of steps taken (or strategies implemented) without realizing it. Strategies can either be brought upon by change or be formulated beforehand and implemented later (Mintzberg, 1987). Consider the analogy of the potter: “The manager is the craftsman and strategy is the clay. Like the craftsman, the manager sits between past organizational capabilities and future market opportunities. If the manager is a true craftsman then he will bring into their work a deep knowledge of the materials in their hands. That is the essence of Crafting Strategy. In practice, of course, all strategy-making walks on two feet: one deliberate, and the other emergent. The combination of deliberation and control with flexibility and organizational learning produces the most effective strategies” (Mintzberg, 1987). Here, CS diverts away from pre-formulated processes and spotlights day-to-day operations with five key characteristics: managing stability, detecting discontinuity, knowing the business, managing pattern, and reconciling change and continuity.

There are two important yet contrasting SP theorists whose concepts are still very relevant in today business schools: Michael Porter’s strategy as a deliberate formulation, and Henry Mintzberg’s emergent strategy. Which one is more relevant today? In most industries, one would look out for strategic flexibility because it is getting extremely difficult to factor in variables into business decisions. With this in mind, Mintzberg’s emergent strategy is simply more relevant to the world we live in – it reflects the fact that our plans will fail (Moore, 2011). Strategy shall be mainly emergent, made of a set of informal single management decisions taken in response to external, unpredicted changes (Ghezzi et al., 2010). While hypotheses explaining the formality of a company’s planning process were well accounted for, no relationship between formal planning process and subjective company performance was observed (Falshaw et al., 2006). On the other hand, Bamberger (1983) noted that there is a positive relationship between the existence of a more or less formal strategic planning system and the firms’ growth and between business strategy and firm’s competitive advantages (Lestari et al., 2020).

Practitioners and academicians debate on the merits of strategic planning although increasing evidence has confirmed that strategic planning has a positive effect on the performance of small businesses (Balasundaram, 2009). Perhaps the more essential question to ask is what form of SP will be a better fit for smaller firms to implement. The findings of this research need to be integrated with prior and future research to illustrate the extent to which SP can generate healthy outcomes for its stakeholders (Phillips & Moutinho, 2014). Due to scarcity of resource, SMEs are unable to devote toextensivecorporate planning style research and strategy development activities (Bellamy et al., 2019). Small enterprises are not little big enterprises, and thus, suggestions for strategic planning that were developed in the context of large firms might not apply to their smaller counterparts. In the future, specifically tailored concepts of strategic planning in smaller enterprises may emerge (Kraus et al., 2006).

2.3. Organizational Learning

2.3.1. Organizational Learning as Strategic Planning Outcome

DiBella et al. (1996) defined organizational learning (OL) as the capacity or processes within an organization to maintain or improve performance based on experience. This activity involves knowledge acquisition (the development or creation of skills, insights, relationships), knowledge sharing (the dissemination to others of what has been acquired by some), and knowledge utilization (integration of the learning so that it is assimilated, broadly available, and can also be generalized to new situations). Research that focuses on exploring the performance effects of planning may need to include OL as a critical contingency variable (Flores et al., 2008). The landscape of SP research, mapped by Wolf and Floyd (2017) also placed OL as one SP’s outcomes. On the other hand, Bartlett and Ghoshal (1998) argue that the age of SP is fast evolving into the era of OL, since managers have deprioritized forecasting and planning for the future and started building sensitivity to emerging changes. Previous studies show that OL is a major component in any effort to improve organizational performance (Hindasah & Nuryakin, 2020; Kittikunchotiwut, 2020) and strengthen competitive advantage (Aragón-Correa et al., 2007; Flores et al., 2008; Lopez et al, 2005; Senge, 1997).

Understanding which types of tools might be useful for different learning contexts, information sources, and processes may assist practitioners in developing their knowledge management strategy and systems, and enable them to “manage” their learning (Jenkin, 2013). Emergent strategy, which is the essence of crafting strategy, implies that an organization is learning what works in practice (Moore, 2011). SP involves sensemaking, which implies OL (Versiani et al., 2018) that reduces the likelihood of a firm’s competencies becoming outdated, enabling the competencies to remain dynamic and thus favoring improvement in performance (García-Morales et al., 2012).

2.3.2. The Impact of Organization Learning on Business Sustainability

Business sustainability refers to the fulfilment of an organization’s profit-making and social development goals without jeopardizing the environment (Galpin et al., 2015). In contrast, the present economic activity has contributed towards a rather serious environmental crises, such as intolerable climate change, scarcity of clean water and food, diminishing fuel, as well as economic instability across nations (Yusoff et al., 2019). Faulkner (2001) warned that the increasing number of disasters and crises will heavily affect the tourism industry, ranging from natural to human influenced incidents. This puts increasing pressure on industry players to minimize the environmental impact of economic activity by focusing on sustainability solutions in their business processes (Yusoff et al., 2019).

The changing business environment and a growing perceived mandate to address sustainability intersects with the reality of resource constraints that is a feature of SMEs (Missimer, 2015). SMEs need to incorporate ‘sustainability’into their strategy in order to achieve business stability and prevent business discontinuation (two crafting strategy characteristics). Previous empirical study on SMEs found that organizational learning capability (Yusoff et al., 2019) and intellectual capital (Srikalimah et al., 2020) influences business sustainability. This relationship indicates that in order for businesses to sustain in a competitive market, organizations should give full attention on learning capability that facilitate learning in the organization. This approach will guide the organizations to react for present and upcoming business problems.

2.4. Hypotheses

Literature has established that strategic planning influences organizational learning. Based on the five characteristic of crafting strategy, five causal hypotheses were identified.

H1: Managing stability positively influences organizational learning.

H2: Detecting discontinuity positively influences organizational learning.

H3: Knowing the business positively influences organizational learning.

H4: Managing pattern positively influences organizational learning.

H5: Reconciling change and continuity positively influence organizational learning.

3. Research Methods and Sample

3.1. Defining Variables

A crafting strategy concept which emphasizes on emergent strategy and less formal strategy planning as opposed to formal and long-term planning might be easier for smaller firms to adopt. Investigating this concept, requires operationalization for quantifiable results. The characteristics of crafting strategy: managing stability, detecting discontinuity, knowing the business, managing pattern, and reconciling change and continuity are the independent variables of this study.

The novelty of this study lies in the development of measurable attributes for each variable based on literature review and in-depth interviews with hotel owner-managers. The dependent variable is organizational learning and measurable attributes were based on the essential factors that determine organizational learning capability (Chiva et al., 2007).

3.2. Data Collection

This study adopts snowball sampling to accommodate questions on sensitive information such as strategy, internal and financial information. Hence, referral from someone that the respondent trusts is crucial for a good response rate. The researcher had to maintain data accuracy by ensuring that the respondents who answered the questionnaire are those heavily involved in day-to-day operations, direction-setting and decision-making. The owner and general manager had to be local. Finally, the hotel unit must have complete independence in strategy and decision making. Respondent and hotel profiles are shown in Table 1 and Table 2.

In the introductory letter, the researcher explained that the analysis would be restricted to an aggregate level to prevent the identification of any individual or hotel names. The questionnaire was constructed based on literature reviews and formalized during the interviews; this further increases the understanding of hotel management while ensuring that the questions were well-understood. Data was collected using a 1 to 5 Likert Scale, where the respondent selects to what extent they agree on the provided statements.

Table 1: Respondent Profiles

Table 2: Hotel Profiles

Initially, the researcher aimed for approximately 250 respondents to complete this study; however, the COVID-19 pandemic disrupted the data-collection process. The steep decline in tourism forces accommodation providers to temporarily close. With no flights heading for Bali, this world-renowned destination was practically empty; the lack of revenue has been sending employees and hotel managers home since March 2020. Questionnaires were sent only to those who fulfilled the requirements and verbally agreed to participate. Seventy questionnaires were distributed and 52 were filled and returned (74% response rate). There were 2 incomplete questionnaires that were taken out of this study.

3.3. Operational Measurements

Managing stability (MS). strategy imposes stability on an organization, there will be no strategy without stability; no direction to the future, and no pattern from the past. Organizations must simultaneously maintain the stability needed for members to be able to make sense of their experiences, yet also achieve fundamental changes that is necessary for effectiveness (Gustafson & Reger, 1995). It is therefore imperative that today’s managers embrace stability and learn to manage continuity if they want to survive (Sturdy & Grey, 2003). Out of the five operational attributes tested, only two are accepted (Table 3).

Table 3: Accepted and Rejected Attributes based on First Round Reliability and Validity Test

Detecting Discontinuity (DD). is one of the most important capabilities for a company to have – the knowledge to detect what might discontinue the business. Strategies form a new set of approaches in organizing and managing innovation – for example, how the firm searches for weak signals about potential discontinuities, how it makes strategic choices in the face of high uncertainty (Bessant, 2008). There are five operational attributes tested, and one is rejected (Table 3).

Knowing the Business (KB). Note the kind of knowledge involved; not intellectual knowledge, not analytical reports or abstracted facts and figures (though these can certainly help), but personal knowledge, intimate understanding, equivalent to the craftsman’s feel for the clay (Mintzberg, 1987a). “Risk comes from not knowing what you are doing -Warren Buffet” (Hagstrom, 2013). There are five operational attributes tested, and one is rejected (Table 3).

Managing Pattern (MP). It is essential for the managers to be able to detect an emerging pattern. Patterns that are proven useful can be incorporated into the intended or formulated strategies. The idea is what’s happening already. A pattern recognition perspective helps explain why some persons, but not others, identify specific opportunities (Baron, 2006). There are four attributes tested and three are rejected (Table 3).

Reconciling Change and Continuity (CC). Management writers have triedto draw attentiontowardsthe need tomanage change and continuity together for better performance; but very few have actually attempted to present a framework for doing so. Managing the forces of continuity and change simultaneously (i.e. interaction of both forces) lead to higher organizational performance (Nasim & Sushil, 2011). Finding the right attributes to measure this paradoxical view, change and continuity, is not an easy task. There are four attributes tested in the final round and three are accepted (Table 3).

Organizational Learning (OL). This refers to measurable attributes based on the essential factors that determine organizational learning capability: experimentation, risk taking, interaction with the external environment, dialogue and participative decision making (Chiva et al., 2007). There are 6 attributes tested; of which, four are accepted (Table 3).

3.4. Partial Least Square

Many of those that use Partial Least Square (PLS) do so because of the limited theoretical and substantive knowledge in their domains. While one might believe that CBSEM is superior to PLS for establishing theoretical models, PLS is incredibly appropriate for exploratory studies where theoretical knowledge is relatively scarce. In fact, there are other instances beyond initial exploratory stages that PLS is well suited for, such as in the case where an incremental study builds on a prior model by developing both new measures and structural paths (Chin, 2010). PLS eminence for structural equation model (SEM) is preferable according to one of the most cited experts in multivariate data analysis, Hair et al. (2011). They contend that PLS-SEM path modelling can indeed be a ‘silver bullet’ for estimating causal models in many theoretical model and empirical data situation.

Criticism of PLS-SEM often focuses on past abuses while overlooking its many benefits that is not offered by CBSEM. PLS-SEM can also be used for confirmatory theory testing (Hair et al., 2011). Its flexibility and its comparatively high statistical power make the PLS particularly adequate for SEM applications that aim at prediction or theory building such as in studies that focus on identifying critical success drivers (Höck & Ringle, 2010).

Other researchers also viewed PLS-SEM as a ‘silver bullet’ or panacea for dealing with empirical research challenges such as smaller sample sizes (Marcoulides & Saunders, 2006; Sosik et al., 2009). Although Marcoulides also warned PLS users to make sure that sample size must be adequate to support the conclusion. As rule of thumb the minimum sample size should be equal to the larger of the following: (1) ten times the largest number of formative indicators used to measure one construct or (2) ten times the largest number of structural paths directed at a particular latent construct in the structural model (Hair et al., 2011; Marcoulides & Saunders, 2006; Sosik et al., 2009)

This study uses PLS-SEM based on Hair’s suggested criteria:

(1) The research is exploratory. Based on researcher knowledge, this will be the first empirical study to develop attributes for SP based on crafting strategy and the first to observe their influence on organizational learning

(2) The goal is to predict key target constructs or identify key ‘driver’ constructs. Aside from testing the hypotheses, this study will also uncover which crafting strategy characteristic is the key ‘driver’ on organizational learning

(3) With sample size in consideration, this study’s use of 50 hotel samples meets the first criteria. In the final model, OL has largest number of indicators, which is five (50 is 10 multiplied by five). There are four independent variables in the final model, directed at one dependent variable; hence the second criteria’s minimum sample is 40, which this study also satisfies.

4. Results

4.1. Evaluation of the Reflective Measurement Model

In PLS-SEM the indicator for reliability is the result of the outer loading value where 0.70 or higher is preferable; however, for exploratory research, loading values between 0.4 to 0.6 or higher is acceptable (Bagozzi & Yi, 1988; Hulland, 1999; Wong, 2013). Although Cronbach’s alpha value is presented to test the internal consistency reliability, Bagozzi and Yi (1988); Hair et al. (2011) suggested the use of composite reliability (CR) as replacement. CR with 0.6 or higher means high levels of internal consistency reliability have been demonstrated.

Validity test is closely related to whether we measure what should be measured or not. The validity test evaluates the degree of accuracy of the research measuring instrument against the actual content or meaning being measured (SUGIARTO, 2017). Reliability test was performed by examining the average variance extracted (AVE), which is an indicator for convergent validity. To show good validity, AVE should be 0.5 or higher (Bagozzi & Yi, 1988). Another indicator to look at is the discriminant validity. Fornell and Larcker (1981) suggest that the square root of AVE of each latent variable should not be greater than the correlation among the latent variables to demonstrate validity.

Table 3 shows the remaining of 29 from originally 37 operational attributes out of six latent variables. The final test result rejected another 11 attributes that did not fulfill the quality criteria. The attributes of MP and MS have significantly reduced to one and two attributes. The researcher decided to merge these two variables in order to prevent identification and estimation problems as well as to ensure satisfactory construct reliability. Each latent construct should be defined by a sufficient number of psychometrically sound indicators: “Two might be fine, three is better, four is best, and anything more is gravy” – Kenny,1979 (Mueller & Hancock, 2008).

The final attributes are 18 and 5 latent variables. This produces implications to the defined hypotheses; at this phase, H1: Managing stability positively influence organizational learning and H4: Managing pattern positively influence organizational learning, are rejected. However, since managing pattern and managing stability both promote past experiences of the organization, then it will make sense to combine the remaining attributes to one characteristic variable, managing pattern and stability (MPS). Aside from rejecting H1 and H4, a new hypothesis, H6: Managing pattern and stability positively influence organizational learning, is added.

Table 4 presents the reliability for five latent variables and 18 attributes. Outer loadings for each attribute are higher than 0.6. All five variables performed a high reliability test result. CR, CA and rho_A were all above 0.7 and mostly above 0.8. Similarly, the composite reliability test indicates interrelated non-homogenous components; all latent variables have both CRs, rho_A and Cronbach’s alpha greater than 0.7, which indicates the measures are reliable (Pratono & Mahmood, 2015). The discriminant validity also proven by all indicators.

Table 4: Final Reliability and Validity Test (A)

AVE for all variables are above 0.5 and Fornell-Larcker criteria in Table 5 suggested that the square root of AVE of each latent variable is not greater than the correlation among the latent variables. Outer loading and cross loading show that each set of attributes strongly relate to their own construct and did not have a stronger connection with other construct. If this is found to be the case, the claim can be made for discriminant validity at the item level. The loadings value for each construct are also showing narrow range which result in a greater confidence that all item help in estimating the underlying construct (convergent validity) (Chin, 2010).

Table 5: Final Reliability and Validity Test (B)

The outer measurement of reliability and validity confirm that the sets of attributes belong to each of the four characteristics do reflect their own variable. The cross loadings are congruent with what was stipulated by Chin (2010), that each item loads more highly on their own construct than on other constructs and that all constructs share more variance with their measures than with other constructs. Therefore, the researcher concludes that all five variables (four exogenous, one endogenous) and 18 attributes are reliable and valid to continue to next measurement: the structural model evaluation.

4.2. Evaluation of Structural Model (Inner Path Model)

As suggested by Hair et al., (2011) the bootstrapping subsamples used is 5,000. The t-statistics and p-value is generated using two tailed tests with significance level of 5%. The structural model evaluation in Table 6 indicated that there is a positive and significant structural relationship between reconciling change and continuity (CC) and organizational learning (OL) with t-value of 5.211 (> 1.96) and P value of 0.000 (< 0.01). Whereas, detecting discontinuity (DD), knowing the business (KB) and managing pattern and stability (MPS) although positive are not significant (t-value < 1.96 and p-value > 0.01). CC has a path coefficient value of 0.614 which is significantly higher than 0.1 and much higher compared to path coefficient belong to DD, KB and MPS.

Table 6: Structural Model Measurement

The computation of size f² approximates the four variables’ predictive power over OL. According to Chin (2010), f² values of 0.02, 0.15, and 0.35 specify whether the exogenous variable have small, medium or large effect. In compliance with preceding result, CC has large effect on OL while DD, KB and MPS have small effect on OL. Another assessment of the structural model involves the model’s capability to predict. The predominant measure of predictive relevance is the Stone–Geisser’s Q² which postulates that the model must be able to adequately predict each endogenous latent construct’s indicators. The Q² value is obtained by using a blindfold procedure. The blindfolding procedure is only applied to endogenous latent constructs that have a reflective measurement model specification. Although Q² comes in two forms, the cross-validated redundancy and communality; Hair et al. (2011) recommend using the cross-validated redundancy, which unlike the cross-validated communality, uses the PLS-SEM estimates of both the structural model and the measurement models for data prediction and, thereby, perfectly fits the PLS-SEM approach. If an endogenous construct’s cross-validated redundancy measure value (i.e., Q²) for a certain endogenous latent variable is larger than zero, its explanatory latent constructs exhibit predictive relevance. The cross-validated redundancy Q² for OL is 0.441 (> 0).

The PLS path modelling estimation explained the following:

Explanation of the target endogenous variable variance. The coefficient determination R2 is 0.663 for organizational learning (OL) endogenous latent variable. This means that all together the four independent variables explained 66.3% of the variance in OL.

Inner model path (arrow between latent variables) coefficient sizes and significance.

The inner model suggests that amongst exogenous variables, CC express the largest explanatory share on OL (0.614), followed by DD (0.143), KB (0.133) and MPS (0.115). The hypothesized path relationship between CC and OL is strongly significant while for each DD, KB, MPS and OL is much less significant. The standardized path coefficient need to be above 0.1 (Wong, 2013).

The four characteristics of crafting strategy have predictive power towards organizational learning (OL). Together they explain 66.3% of variance in OL. However not all characteristics share the same explanatory power over OL. Reconciling change and continuity (CC) dominantly influence OL while the other three characteristics managing pattern & stability (MPS), detecting discontinuity (DD), knowing the business (KB) are weak predictor of OL. This conclusion is drawn from several indicators t-value, p-value, f² value and path coefficient. The result of Q² cross-validated redundancy confirmed that the explanatory latent constructs (CC, DD, KB and MPS) exhibit predictive relevance towards OL.

The PLS-SEM measurement results supported and not supported following hypotheses:

Not supported H1: Managing stability positively influences organizational learning.

Supported H2: Detecting discontinuity positively influences organizational learning.

Supported H3: Knowing the business positively influences organizational learning.

Not supported H4: Managing pattern positively influences organizational learning.

Supported H5: Reconciling change and continuity positively influence organizational learning.

Supported H6: Managing pattern and stability positively influence organizational learning.

5. Discussion and Conclusion

Crafting strategy through its four characteristics prove to have a positive and significant influence on OL, although the degree of influence varies for each characteristic.

Managing pattern and stability, which relates to things that have been done inside the company – past experiences, repetition of pattern, repetition of strategy that used to bring success – is proven to reflect a weak influence towards OL. Schilling and Kluge (2009) developed a theoretical foundation to describe and explain impediments to OL based on the expanded 4i model (intuiting, interpreting, integrating and institutionalizing). These following barriers are highlighted during the integrating phase: (1) values and assumptions of senior managers and managers’ desire to retain a positive self-image and (2) defensive routines of other departments (not invented here-syndrome). These samples of OL barriers could be closely related with managing pattern and stability. From this study, MPS still contributes to a positive relationship, but if overlooked by the strategist, the glory in the past can make the company reluctant to learn new things which could potentially impair performance.

Detecting discontinuity is a knowledge or effort to prevent discontinuation of business. Contrary to managing pattern and stability, this characteristic – when overdone – could drive an organization to perpetual instability. Strategies and directions keep changing as repercussions of ‘protecting the organization from discontinuation’. Every time the market changes or a competitor launches a new campaign, these companies will immediately react leaving little time and opportunity to learn. Remember, there will be no learning without stability.

The third characteristic, knowing the business, involves the personal knowledge of company senior managers superseding intellectual knowledge or analytical reports, facts and figures. This may present several barriers to organizational learning: (1) over-confidence of managers in existing practices, rigid and outdated core beliefs, (2) fear of loss of ownership and control of knowledge (Schilling & Kluge, 2009). Just knowing the business might have been enough to run a business successfully two decades ago, but it is a different story today. Knowing the business without any motivation to distribute or pass down to other organization members prevents learning. One of Peter Senge’s five learning disciplines is personal mastery: people with a high level of personal mastery live in a continual learning mode, they never ‘arrive’ (Senge, 1997).

The fourth characteristic, reconciling change and continuity, has a significantly strong effect over organizational learning. The continuity part of the fourth characteristic contrasts the previous detecting discontinuity characteristic. An organization that is preoccupied with avoiding discontinuity might struggle with instability due to constant strategic adjustments. On the other hand, an organization that focuses on continuity might understand maintaining stability while adjusting to change is essential for sustainable growth. Nasim and Sushil (2011) pay special interest in studying the paradox of managing continuity and change. In the last two decades, however, the increasing pace of change has compelled businesses to evolve a more flexible approach in managing change and transformation, thus calling on them to strike a balance. Forces of continuity can contribute to inertia in an organization, therefore need to be managed for effective change. Although, not all continuity forces obstruct change, some for example, core ideology and strong culture, may even be vital or desirable to leverage change (Nasim & Sushil, 2011).

The researcher advises that OL is not the only proximal outcome influenced by SP. OL is also not the only factor that influences business performance. Reconciling change and continuity have proven to significantly affect OL, but other characteristics that has only a weak effect on OL could have a major effect on other proximal outcomes. In summary, the four characteristics of crafting strategy need to be in balance – too much or too little of one ingredient will jeopardize an organization’s chances of survival.

6. Contribution

Mintzberg’s crafting strategy is still widely thought in today business schools, yet the number of empirical studies conducted on this seminal strategic planning theory is close to none. The researcher did not find any previous empirical studies that attempt to operationalize the crafting strategy.

6.1. Theoretical Contribution

This explorative study has successfully introduced four characteristics of Mintzberg crafting strategy along with each sets of measurable attributes. In doing so, this study has bridged the gap between an abstract concept of craftmanship in doing SP, which is a potentially much better approach for SMEs, with daily operational practices. This study also proves that Mintzberg’s approach can be used to predict organizational learning. This relationship is crucial since previous studies concluded that organizational learning improves company performance.

6.2. Managerial Implication

With a deeper understanding on how to do a well-grounded but less complicated SP, SME managers will be motivated to spend more effort in SP. The confirmation of a positive relationship between crafting strategy and OL will motivate managers to start evaluating the effectiveness of their SP against organizational learning. Hopefully, as previous studies have concluded, the improvement of organizational learning will strengthen business performance hence leading to greater chances of survival.

7. Limitation and Future Research

This investigation has several limitations. First, this study only focuses on one outcome construct, organizational learning. The effect of each crafting strategy characteristics could be different for other outcomes, which presents a promising domain to explore future research. Secondly, the data collection is based on the perception of individual respondents, which implicates a certain degree of subjectivity. Thirdly, this study has a limited number of respondents, restricted by pandemic conditions, geography and industry.

Owing to the fact that the perimeters of this study end at measuring structural relationship between crafting strategy characteristics and organizational learning as just one of the proximal outcomes, fellow researchers are invited to further improve on this first attempt at operationalizing crafting strategy. A more comprehensive model could be built upon by adding other constructs and analyzing the impact towards other proximal outcomes such as organization effectiveness, or distal outcomes such as business performance and sustainability. Future research can also verify its application in different sectors of the tourism industry, or even different industries, with a bigger sample, at different locations, larger organizations, and with more or different outcomes as intervening, mediating or independent variables. Ultimately, the author hopes that this study may bring back an enthusiasm for empirical research study in strategic planning, especially those that will benefit small and medium-sized organizations.