Entrepreneurs’ Cognitive Schemas and New Venture Business Model Innovativeness: The Mediating Effect of Environmental Scanning

Abstract

Sustainable innovation now plays a leading role in the transformation and upgrading activities of enterprises, and the new business model is increasingly a typical form of innovation. Therefore, based on neo-institutional theory of organization, this study mainly analyzes the impact of cognitive schemas on new ventures’ business model innovativeness by selecting new service industry enterprises as the research object and referring to data from the China Statistical Yearbook and “China Innovation and Entrepreneurship Index”. In addition, we entrusted professional market research companies and business incubators to collect 142 valid questionnaires. The results show that general cognitive schema negatively affects business model innovativeness, while specific cognitive schema positively affects it. Specifically, general cognitive schema reduces the level of effort and sustainability of entrepreneurial environmental scanning activities, which is not conducive to the design of highly innovative business models; specific cognitive schema not only improves the level of effort and sustainability of entrepreneurial environmental scanning activities, but also drives the entrepreneurs to pay more attention to new information outside the industry, which is conducive to the design of a highly innovative business model.

1. Introduction

New ventures are the most active group in designing new business models, some of which are even keener on designing highly innovative business models and exploring the huge value behind them. This enables them to survive a crisis and then enter the fast track of development, becoming a force that cannot be ignored to promote economic transformation and upgrading. As indicated in the “2016 China Unicorn Enterprise Development Report”, jointly released by Torch High Technology Industry Development Center, Ministry of Science and Technology, and Greatwall Enterprise Institute, the unique business models designed by entrepreneurs are largely responsible for the rapid growth of new ventures such as Didi and Mobike into unicorns, the emergence of new industries and the transformation of traditional industries. Still, there are rare native, highly innovative business models that can overturn traditional business rules. Most entrepreneurs tend to copy the traditional business model [1], simply change partial content [2] or barely change it. They are committed to ensuring new ventures can obtain excess returns from the traditional business model by setting complementary solutions, fixing partner relationships and improving efficiency [3]. This indicated that it is not easy to design highly innovative business models. Therefore, “why some entrepreneurs can design highly innovative business models while others cannot” is a practical problem for debate.

This practical problem belongs to the subject research of “business model design”. Regarding business model design, scholars, most “process-consequence”-oriented at present, think the business model is gradually established by entrepreneurs learning by trial and error when they constantly develop, test and adjust a variety of business model assumptions [4]. However, the related research does not analyze the underlying reasons that lead to different innovative business model designs, which only shows the process and steps of trial-and-error learning.

Discussing business model design from the perspective of antecedents can remedy such deficiency, which can be called “antecedent-consequence”-oriented business model design research, explaining the problem from two different perspectives of “positioning” and “cognition”. Most early research adopted the perspective of positioning and believed that external environmental requirements determined the direction of business model design [5], while differences in enterprise resource reserves would further lead to different innovative business model designs [6]. This view assumes that entrepreneurs can obtain and interpret external environmental information equally and completely, which is often impossible and has been greatly questioned in subsequent research. In recent years, research from a cognitive perspective has been gradually emerging. Scholars believe that business models are conceived and formed by entrepreneurs, with the existing cognitive schema about business models in their minds providing a reference for such conceiving activities [7]. The differences among the cognitive schemas are the root cause of why entrepreneurs design different innovative business models [8], and cognitive flexibility has a positive direct effect on entrepreneurship competence [9]. Compared with research from the positioning perspective, research from the cognitive perspective, with decision logic of bounded rationality, loosens presuppositions, which is more in line with the actual situation of decision making. Therefore, it is more reasonable to discuss the mechanism of business model design from the perspective of cognitive schema. In its infancy, research from cognitive perspective, however, lacks a clear logic to define cognitive schema categories and analyze their influence mechanism [10] and also ignores the role of the external environment it plays as the “key information material provider” in the process of business model design. Therefore, it could not systematically explain why different business model design results in different levels of innovation by mistaking this process for an entrepreneur’s fantasy process. Therefore, it is necessary to integrate viewpoints from positioning perspectives in further research, of which the key is to deal with the relationship between “entrepreneurs’ cognitive schema” and “external environmental information”.

Based on such cognition, from the cognitive perspective, there is significant empirical evidence on the relationship between cognitive schemas and new ventures’ business model innovativeness. According to the views of Sarma and Sun (2017) [11], a business model is an institution, so it is appropriate to use the analytical logic of the neo-institutional theory of organization for reference to solve the theoretical research problem in this paper. The neo-institutional theory of organization focuses on the research of “How institutions form and change” [12]. Scholars advocate the core status of organizational decision-makers’ cognitive schemas while emphasizing the important role of the external environment and carry out analysis by introducing mature cognitive theories from other fields [13]. There is a representative research point that the organizational decision-makers’ cognitive schemas affect how they understand the social environment, and then determine the institution they eventually form [14]. We further use sense-making theory to expand this view, believing that the environmental scanning activity is an important means for entrepreneurs to understand the characteristics of the business environment [15]. Different cognitive schemas will drive entrepreneurs to carry out different environmental scanning activities, which results in business model designs with levels of innovation [8,16].

Based on the above analysis, this paper will, following the theoretical logic chain of “cognitive schema-environmental scanning-business model innovation”, explore the impact that entrepreneurs’ cognitive schemas impose on business model innovation of new ventures, which can be defined as the following: the mediating effect that environmental scanning has on the impact that entrepreneurs’ cognitive schemas impose on business model innovation in new ventures. In terms of theoretical contribution, the analysis from the cognitive perspective in this paper is not only a powerful supplement to the “process-consequence”-oriented research, but also conducive to the development of “antecedent-consequence”-oriented research by revealing the influence that cognitive schemas, as the antecedent of business model design, impose on the designs.

2. Theoretical Analysis and Research Hypothesis

2.1. Institutional Attributes and Category Definition of Business Model

A business model is usually defined as a trading system through which enterprises deal with their stakeholders [3,17]. Such a system, with institutional attributes characterized by business rules, norms and creeds [11], can guide and restrain the business operations of enterprises to ensure that they can continuously and repeatedly acquire specific customer value and enterprise value. Scholars of neo-institutional theory of organization believe that there are three pillars, namely regulation, standard and cognition, in the institution, of which cognition is the most striking feature of research in this field. Cognition, in this case, refers to the organizations’ common understanding about the properties of the social environment, which, from the individual level, is the cognitive schema in the mind of organizational decision makers used to construct social significance [18]. In recent years, the cognitive quality of business models has gained more and more attention. Zhou directly defines the business model as a cognitive schema explaining how enterprises acquire value from business transactions [1]. Therefore, cognitive schemas are the essence of the institutional attributes of a business model, and are also the key entry point to discuss, based on neo-institutional theory of organization, the design of a business model.

Old institutions influence the formation and change in new institutions [18]. Therefore, from the micro-cognitive level, the current cognitive schema of organizational decision-makers about the properties of the social environment determines whether they abide by or reform the dominant institutional arrangement in the field [13]. To distinguish different cognitive schema categories is the first task to explain such a decisive effect [19]. In the research of business models, Amit and Zott (2015) [7] also clearly pointed out that the existing cognition differences about business models in entrepreneurs’ minds are an important antecedent that leads to different kinds of business model designs.

The field is considered as the core analytical level of research on the neo-institutional theory of organization, which is a recognized institutional life field composed of organizations that are aggregated or clustered together [18]. In the same field, the organizational decision-makers’ cognitive schemas about the properties of the social environment are homogeneous because of the same institutional arrangement, while in different fields, each other’s cognitive schemas about the properties of the social environment are unique because of the different institutional arrangements [12]. In business model research, the field can be regarded as the industry in which the enterprise is located [7]. A similar process is also carried out in many enterprise management types of research based on the neo-institutional theory of organization [20]. According to these understandings, the entrepreneurs’ cognitive schema on business models, taking the target industry of entrepreneurial activities as a reference, can be divided into general cognitive schema and specific cognitive schema (Figure 1).

As shown in Figure 1, Entrepreneur A and Entrepreneurship B belong to entrepreneurship in the same industry. It means that they have both been working in the target industry of entrepreneurial activities, and that they have a common understanding of the process and rules of trading activities in this industry. This understanding is no different from customers, suppliers, investors, government and other stakeholders. Their cognitive schemas about the business model, therefore, reflect the dominant business model generally familiar to individuals and organizations in this industry [21], which can be referred to as “general cognitive schema”. Entrepreneur C and Entrepreneurship D belong to entrepreneurship in different industries. It means that they have both been working outside the target industry of entrepreneurial activities and that their understandings about the process of trading activities are different from the dominant business model in this industry. Therefore, the cognitive schemas about business models of Entrepreneur C and Entrepreneur D are very unique to Entrepreneur A and Entrepreneur B, which are unfamiliar to individuals and organizations in this industry and may lead to cognitive conflicts [2], which can be referred as to “specific cognitive schema”. Different cognitive schemas will enable entrepreneurs to have different understandings of “what the business model should be”, which is the root cause of the different levels of innovation in the new ventures’ business model design [22].

Although research on business models does not directly refer to the above-mentioned two cognitive schemas, some scholars have similar definitions. For example, Chesbrough and Rosenbloom analyzed the business model design process of six new technology enterprises from the Xerox research and development center. they found that, at the beginning [3], these enterprises all took the business model of Xerox as the dominant logic to develop business activities, meaning that their original business models were almost the same as Xerox’s. Such dominant logic is the embodiment of the general cognitive schema. Martins analyzed the business model design process of the founder or top managers of six companies [22], such as Tesla and Starbucks. They found that they formed a unique understanding of the business model through learning from the business operation models in other industries and designed a highly innovative business model. Such a unique understanding is the embodiment of the specific cognitive schema.

2.2. Entrepreneurs’ Cognitive Schema and New Ventures’ Business Model Innovativeness

According to the neo-institutional theory of organization, if a new venture wants to enter the market smoothly and attain the incumbents’ recognition, it must win the recognition of legality. For this, entrepreneurs may implement robust design strategies to imitate the dominant business models within the control of incumbents in the target industry of entrepreneurial activities to effectively avoid the strong resistance that may be encountered because of the application of highly innovative business models [23]. Cliff believed that entrepreneurs who strongly follow and maintain the dominant business logic of the industry which they are in are likely to become imitative entrepreneurs and that their strategic decisions (such as business model design [20], made under the influence of such logic, usually tend to be convergent. In addition, resource acquisition is the key to ensure the smooth growth of new ventures. If entrepreneurs can form a very close relationship with the incumbents in their industry, they can directly obtain resource support from the incumbents and establish a high industry status. Therefore, general cognitive schema will drive entrepreneurs to stick to and maintain the dominant business model in the industry to ensure a long-term good relationship between themselves and the incumbents [24].

At a deeper level, if they can conceive, based on the familiar cognitive schema, a business model which fits right into the industry, entrepreneurs will feel settled and comfortable. Such feelings could be more enjoyable if they can benefit from the dominant business model. This means that the dominant business model will bring entrepreneurs positive emotional feelings and that entrepreneurs will enhance their recognition of it on the emotional level as feedback [25]. Deep emotion is an important guarantee to maintain institutional stability. If they have high emotional recognition, actors will show extremely high spiritual support towards the current institutional arrangements and will not make decisions deviating from the current institutional arrangements [26]. Therefore, the entrepreneur will not abandon the dominant business model to pursue a highly innovative model full of uncertainty when there are various general cognitive schemas and the dominant business model still has the potential to make a profit.

Accordingly, a hypothesis is proposed in this paper as follows:

Hypothesis 1a.

General cognitive schemas negatively affect business model innovativeness.

According to the neo-institutional theory of organization, institutions are never fixed and actors with rich experience outside the field will be keen to challenge the existing institutions [20]. The change in the dominant business model in the industry depends on the entrepreneurs’ reflexivity, the level of which is determined by the uniqueness of previous experience [27]. The specific cognitive schema is derived from the previous experience accumulated by entrepreneurs, and it is different from that in the current industry. Such novelty makes it difficult to form an apparent match between the specific cognitive schema and the target industry environment of entrepreneurial activities [28]. However, for entrepreneurs from outside the industry, the specific cognitive schema is the main reference for business model design, and they will continue to make decisions based on this schema when there is no suitable substitute. Therefore, entrepreneurs have to make a deeper analysis of the structural matching degree between the specific cognitive schema and entrepreneurial environment of the target industry [28], which is a higher level of reflexivity that can enable them to find opportunities to reform the dominant business model of the target industry easily.

What is behind this reflexivity is the entrepreneurs’ demand for profits. The dominant business model of the target industry may conflict with the dominant value system of entrepreneurs, leading to cognitive dissonance, which is an important reason for business model innovation [20]. To address cognitive dissonance requires a feasible alternative to the dominant business model of the target industry. Entrepreneurs from outside the industry do not consider the dominant business model of the target industry as a necessary norm. They can develop feasible alternatives under the lead of various specific cognitive schemas. Therefore, they will strongly question the legality of the model from the perspective of performance and morality when they feel that the dominant model is not conducive to their entrepreneurial activities [29]. Such questions will further drive entrepreneurs from outside the industry to actively reveal the deficiencies in the dominant business model and develop new business models in line with their interests based on the structured match [30].

Accordingly, a hypothesis is proposed in this paper as follows:

Hypothesis 1b.

Specific cognitive schemas positively affect business model innovativeness.

2.3. The Mediating Effect of Environmental Scanning

When there is a large amount of information in the external environment but limited cognitive resources for entrepreneurs, their environmental scanning activities show the following characteristics [31]: entrepreneurs will selectively scan environmental information in different fields; alternatively, entrepreneurs will appropriately allocate cognitive resources to environmental scanning activities, which is reflected in the entrepreneurs’ effort and sustainability level compared with other work when they carry out environmental scanning activities. Both of these characteristics will be analyzed in the following paragraphs.

2.3.1. The Mediating Effect of Environmental Scanning Field

The dominant business model with institutional attributes is the authority in the entrepreneurs’ minds, which means that the dominant model is representative of the correct model [8], making environmental scanning activities containing cognitive biases based on this appropriate. If there are more various general schemas, the entrepreneur will pay more attention to the familiar information related to the dominant model when designing the business model and think less about whether the information is valuable. Unfamiliar information means high uncertainty [28]. Entrepreneurs cannot judge whether it is a threat or an opportunity according to the existing cognitive schema. Therefore, they tend to ignore unfamiliar information that is difficult to effectively control. In addition, the institutional attributes of the dominant business model require entrepreneurs to carefully understand the model to ensure that the design competition pursued is “legal” [7], which drives entrepreneurs to widely collect information about the dominant model. When they have rich-enough general schemas, they will find that the dominant business model is not fixed, and, on the contrary, it should be partially adjusted continuously to maintain its superiority [14]. Such partial adjustment presents a tipping point in cognitive change at the cognitive level. Before reaching the tipping point, entrepreneurs need to consider how to better adapt to the dominant model, which requires them to collect information within the industry to continuously enrich the information base of the dominant model and improve its effectiveness [21].

Accordingly, a hypothesis is proposed in this paper as follows:

Hypothesis 2a.

General cognitive schemas negatively affect unfamiliar field scanning.

Hypothesis 2b.

General cognitive schemas negatively affect outside industry field scanning.

The business model designs that specific cognitive schemas bring to entrepreneurs are not from, or even may not be suitable for, the current entrepreneurial activities in the target industry. These designs also have institutional attributes, and entrepreneurs are very familiar with and have benefited from them for a long time so that they will not give them up easily [8]. They have to work hard to find the entry point to effectively apply these designs through collecting unfamiliar information from the industry. Entrepreneurs collect information about the industry just to know about the industry, but they do not agree with the dominant business model in the industry. Cliff argued that to question the dominant business model is the beginning of the development of new models, which is usually initiated by entrepreneurs with experience outside the industry [20]. However, despite knowing that they need to “get rid of the constraint of the dominant business model”, entrepreneurs do not have a specific ultimate goal to “create a new business model” [29]. It is difficult for them to precisely locate the key information and determine how to use the information from different sources. Therefore, they have to refer to the business model design logic of multiple industries to conduct experimental activities, and to validate relevant hypotheses requires more industries to be scanned [4].

Accordingly, a hypothesis is proposed in this paper as follows:

Hypothesis 2c.

Specific cognitive schemas positively affect unfamiliar field scanning.

Hypothesis 2d.

Specific cognitive schemas positively affect outside industry field scanning.

New information and knowledge from unfamiliar and outside industry fields provide entrepreneurs with a new perspective on business models, which provides them with the opportunity to extensively know about the business models of other industries, identify the deficiencies in the dominant business model in the target industry and the opportunities for correction question the rationality of the dominant model and remove its “mythical” status [20]. With the input of new information and knowledge, entrepreneurs tend to use creative methods, such as analogical reasoning and the connectivity concept, to connect the new information and knowledge to the existing knowledge structure to build a highly innovative business model that can replace the dominant business model in the target industry [22,31]. However, when entrepreneurs focus on the familiar information within their industry, it is difficult for them to learn information about business models in other industries, so that they cannot develop a feasible alternative to the dominant business model in the target industry. Moreover, with the continuous increase in experience in the industry, entrepreneurs will excessively rely on and apply the familiar cognitive schemas related to the dominant business models in the industry and form cognitive inertia, which will restrict the willingness and ability of entrepreneurs to design highly innovative business models [30].

Accordingly, a hypothesis is proposed in this paper as follows:

Hypothesis 3a.

Unfamiliar field scanning positively affects business model innovativeness.

Hypothesis 3b.

Outside industry field scanning positively affects business model innovativeness.

Combined with the above analysis, from the perspective of cognitive schemas, entrepreneurs with more general cognitive schema are more likely to design low innovative business models, and tend to carry out fewer environmental scanning activities in unfamiliar and outside industry fields, while entrepreneurs with more specific cognitive schema are more likely to design highly innovative business models and tend to carry out more environmental scanning activities in unfamiliar and outside industry fields. From the perspective of environmental scanning, entrepreneurs who tend to carry out environmental scanning activities in unfamiliar and outside industry fields are more likely to design highly innovative business models and vice versa. Based on this, the mediating effect hypothesis is proposed in this paper as follows:

Hypothesis 4a.

Unfamiliar field scanning plays a mediating role between general cognitive schema and business model innovativeness.

Hypothesis 4b.

Outside industry field scanning plays a mediating role between general cognitive schema and business model innovativeness.

Hypothesis 4c.

Unfamiliar field scanning plays a mediating role between specific cognitive schema and business model innovativeness.

Hypothesis 4d.

Outside industry field scanning plays a mediating role between specific cognitive schema and business model innovativeness.

2.3.2. The Mediating Effect of Environmental Scanning Intensity

General cognitive schema prevents entrepreneurs from investing a lot of time and energy in environmental scanning activities, which is mainly due to the isomorphism mechanism of institutional attributes of dominant business models. On the one hand, combined with the mimetic isomorphism mechanism, new ventures usually face high uncertainty in the early stage of establishment. Imitating the business model of the incumbents is an effective means to reduce uncertainty because the incumbents’ business model is the dominant model in the industry, which means entrepreneurs do not have to collect additional information to confirm its legality. On the other hand, combined with the coercive isomorphism mechanism and normative isomorphism mechanism, the business model designed by entrepreneurs should meet the expectations of stakeholders in this industry and social culture [21]. It means that there is a common standard, namely the dominant business model in the industry, to measure whether the new ventures’ business model is suitable or not. This means that it is safe to develop a business based on the dominant business model. General cognitive schemas make entrepreneurs more willing to accept the standard and not too proactive in collecting new information to challenge the dominant business model.

Based on the above analysis, a hypothesis is proposed in this paper as follows:

Hypothesis 5a.

General cognitive schema negatively affects the effort level of scanning.

Hypothesis 5b.

General cognitive schema negatively affects the sustainability level of scanning.

A specific cognitive schema makes entrepreneurs disagree with the dominant business model in the industry and try to revise and even subvert the model. Changing the dominant business model is an institutional change action, which is not easy. Entrepreneurs need to collect and understand a large amount of information inside and outside the industry to form new knowledge to prove to stakeholders the disadvantages of the dominant model and the advantages of the new model [30]. For example, Denicolai believed that to realize business model innovation, enterprises should not only pay attention to the accumulation of their own experience and knowledge but also need to constantly acquire new knowledge from the outside, and then connect all of this in novel ways [32]. According to the neo-institutional theory of organization, institutional change cannot be achieved overnight, and it will go through crucial links, such as de-institutionalization, pre-institutionalization, theorization, diffusion and strengthening institutionalization. Actors need to pay attention to different environmental information in different links and carry out different reform activities [29]. According to such a view, the highly innovative business model design process also includes more sub-processes, which are usually known as the different phases of learning in business model research. Each phase has a specific task in knowledge accumulation, and the lack of any phase will produce huge damage to the overall design process, which means that environmental scanning activities must be carried out for a long time.

Accordingly, a hypothesis is proposed in this paper as follows:

Hypothesis 5c.

Specific cognitive schemas positively affect the effort level of scanning.

Hypothesis 5d.

Specific cognitive schemas positively affect the sustainability level of scanning.

The high level of effort and sustainability of entrepreneurial scanning means that they will allocate more attention resources to environmental scanning activities, which can reduce the cognitive bias due to limited attention resources [31]. The reduction in cognitive bias makes it possible for entrepreneurs to develop a variety of feasible business model designs after systematically analyzing information from different sources. With an increase in available options, entrepreneurs are more likely to design highly innovative business models [4]. For example, research from Chesbrough (2010) [8] suggests that highly innovative business models come from the experimental process. Because they do not figure out “whether to innovate” or “how to innovate”, and take the uncertainty of innovation results and the resistance stakeholders possibly develop into consideration, entrepreneurs usually experiment with varieties of business models at the same time and then choose the new final one according to the experimental results. Further, continuous environmental scanning activities also help entrepreneurs improve their capabilities to creatively process information, making them more likely to design highly innovative business models.

Accordingly, a hypothesis is proposed in this paper as follows:

Hypothesis 6a.

The effort level of scanning positively affects business model innovativeness.

Hypothesis 6b.

The sustainability level of scanning positively affects business model innovativeness.

Combined with the above analysis, from the perspective of cognitive schemas, entrepreneurs with a more general cognitive schema are more likely to design low innovative business models and tend to devote much less effort to environmental scanning activities, and the scanning activities are also not sustained. Entrepreneurs with a more specific cognitive schema are more likely to design highly innovative business models and will be very diligent and persistent in carrying out environmental scanning activities. From the perspective of environmental scanning, entrepreneurs who tend to carry out environmental scanning activities diligently and persistently are more likely to design highly innovative business models and vice versa. Based on this, the mediating effect hypothesis is proposed in this paper as follows:

Hypothesis 7a.

The effort level of scanning plays a mediating role between general cognitive schema and business model innovativeness.

Hypothesis 7b.

The sustainability level of scanning plays a mediating role between general cognitive schema and business model innovativeness.

Hypothesis 7c.

The effort level of scanning plays a mediating role between specific cognitive schema and business model innovativeness.

Hypothesis 7d.

The sustainability level of scanning plays a mediating role between specific cognitive schema and business model innovativeness.

The research model is shown in Figure 2.

3. Research Design

3.1. Measurement of Variables

3.1.1. Independent Variable: Cognitive Schema

This paper, with the reference of viewpoints of upper echelons theory, takes entrepreneurs’ working experience as a proxy variable to measure their cognitive schemas. The operation is as follows: First, ask the interviewee to list all of their work experience, corresponding industries and the time when it started and ended in detail. Second, based on the research of Toft-Kehler et al. (2014) [33] and the code distance of Industrial Classification for National Economic Activities (GB/4754-2011), calculate the industry similarity between the industry where they worked and the current entrepreneurial industry. The industry similarity index is used to reflect the internal and external attributes of the entrepreneurs’ work experience. To be specific, the industry similarity between the industry where they worked as No. $i$ work and the current entrepreneurial industry can be calculated by Formula (1):

$$Similarity(i)=d_i$$

(1)

In this formula, $d_i$ refers to the code distance of industrial classification for national economic activities between the industry where they worked as No. $i$ work and the current entrepreneurial industry. When the top three-digit codes of the two industries are the same, it means that they belong to the same subcategory with a strong correlation. At this point, $d_i$ = 3. When the top two-digit codes of the two industries are the same, it means that they belong to the same subcategory with a less-strong correlation. At this point, $d_i$ = 2. When the top two-digit codes of the two industries are the same, it means that they belong to the same category with a non-obvious correlation. At this point, $d_i$ = 1. When no codes of the two industries are the same, it means that they belong to different industries with no correlation. At this point, $d_i$ = 0. Third, combined with the research of Spanjer and Witteloostuijn [34], work out the weighted accumulative scores of entrepreneurs’ inside industry and outside industry experience, which are the richness scores of entrepreneurs’ general cognitive schema and specific cognitive schema, respectively. The formula is as follows:

$$IE={\displaystyle \sum _{i=1}^{+\infty }{e}_i{d}_i}$$

(2)

$$OE={\displaystyle \sum _{i=1}^{+\infty }{e}_i(3-d_i})$$

(3)

In these formulas, $IE$ and $OE$ refer to the weighted accumulative scores of entrepreneurs’ inside industry and outside industry experience; $e_i$ refers to the number of years of working experience of their No. $i$ work; $d_i$ refers to the same meaning as in Formula (1), reflecting the industry similarity. Since the maximum value of $d_i$ is 3, 3-$d_i$ refers to the industry difference between the industry where they worked as No. $i$ work and the current entrepreneurial industry. When $d_i$ = 0, it means that the No. $i$ work experience completely belongs to outside industry experience.

3.1.2. Dependent Variable: Business Model Innovativeness

With reference to the novel business model scale developed by Zott and Amit [24] and the research of Guo et al. [35]. the measurement of business model innovativeness maintains nine items after deleting some items and adopts a 5-point Likert scale (1 = completely inconsistent, 5 = completely consistent) to measure.

3.1.3. Mediating Variable: Environmental Scanning Field and Environmental Scanning Intensity

The measurement of environmental scanning field and environmental scanning intensity refers to the measurement method of Li et al. [31]. The environmental scanning field includes two variables, namely scanning of unfamiliar fields and scanning of outside industry fields, which are measured according to a score distribution table; environmental scanning intensity includes two variables, namely scanning effort level and scanning sustainability level, which are measured according to the 5-point Likert scale with 4 items (1 = completely inconsistent, 5 = completely consistent).

3.1.4. Control Variable

In this paper, the following variables are controlled in the analysis: variables at the individual level include the gender, age and education level of entrepreneurs; variables at the enterprise level include the age of the enterprise, the industry in which the enterprise is located, the asset size and the region in which the enterprise is located. All control variables are virtualized in the analysis by being assigned either 0 or 1.

3.2. Samples and Data Collection

This paper selects 260 new service ventures as the research object after learning from previous business model research and chooses Tianjin and Shandong Province where there is a developed service sector or higher innovation as the research region, with reference to the China Statistical Yearbook (2016) and “China Innovation and Entrepreneurship Index”. Based on the above work, we entrust professional market research companies and visit business incubators to issue and collect questionnaires. After screening, there are 142 valid questionnaires in the final official database.

Furthermore, after the descriptive statistics analysis of skewness and kurtosis of the measured variables, it is found that the absolute value of the skewness of each measured variable is less than 2, and the absolute value of the kurtosis is less than 5. Therefore, it can be determined that the data of the measured index conform to the normal distribution(as shown in

Table 1. Descriptive statistics of sample characteristics (N = 142).

Variable	Type	Quantity and Proportion	Variable	Type	Quantity and Proportion
Gender	Male	82 (57.7%)	Region	Tianjin	96 (67.6%)
	Female	60 (42.3%)		Shandong	46 (32.4%)
Age	Age 25 and below	8 (5.6%)	Asset size	Less than 500,000	23 (16.2%)
	Age 26~30	33 (23.2%)		500,000~990,000	10 (7.0%)
	Age 31~35	57 (40.1%)		1,000,000~4,990,000	63 (44.4%)
	Age 36~40	30 (21.1%)		5,000,000~9,990,000	22 (15.5%)
	Age 41 and above	14 (9.9%)		10,000,000 and above	24 (16.9%)
Educational level	Senior high school/technical secondary school and below	5 (3.5%)	Age of enterprise	1 year	28 (19.7%)
				2 years	47 (33.1%)
	College	42 (29.6%)		3 years	39 (27.5%)
				4 years	28 (19.7%)
	Bachelor Degree	76 (53.5%)	Industry	Information transmission, software and information technology service	26 (18.3%)
	Master Degree	17 (12.0%)		Wholesale and retail service	29 (20.4%)
				Leasing and business service	27 (19.0%)
	Doctor Degree	2 (1.4%)
				Others	60 (42.3%)

).

Since the data collected in this paper are all self-reported from the respondents, this may lead to common method bias. In order to reduce this problem, this paper controls the common method bias from both programming design and statistical testing. Among them, the program design adopts the following methods: first, many objective variables are set in the questionnaire; second, inversion questions and verification questions are set; third, in the guidance part, it is pointed out that the investigation is only for academic research, and there is no right answer. We will keep the information strictly confidential to avoid deviation of the respondents for the protection of privacy or publicity.

The statistical test adopts Harman’s single factor test, putting all variables together for unrotated exploratory factor analysis. According to the results of factor analysis, if the explanatory power of a single factor or a common factor is very large, it can be determined that there is a serious common method bias in the data. The result shows that the KMO is 0.867, and there are five factors obtained in total when it is not rotated. The explanatory power of the first factor is 35.646%, which does not account for the majority, indicating that the common method bias in this paper is acceptable.

4. Discussion

Reliability and Validity Tests

For the reliability test: the α coefficient of business model innovativeness and scanning sustainability levels is 0.876 and 0.751, respectively, both higher than 0.7; the α value of scanning effort level is 0.690. However, through analysis of the CITC value, it is found that the overall reliability significantly improved to 0.745 after deleting the item “I take different methods to find information sources to obtain relevant information”, which, therefore, is deleted in the formal analysis.

Based on that, Mplus 7.0 software was used to test the construct validity of each variable scale. It is found that in business model innovativeness scale: χ2/df = 1.163, RMSEA = 0.034, CFI = 0.991, TLI = 0.988, SRMR = 0.037; scanning effort and sustainability level scale: χ2/df = 1.353, RMSEA = 0.050, CFI = 0.989, TLI = 0.979, SRMR = 0.032. In addition, from the perspective of standardized factor load values, the factor load between each variable measured item and construct is significantly at the level of 0.01, and all of them are higher than 0.5. Therefore, the fitting degrees of different measurement models of variables are all good, indicating that the construction validity of these scales is relatively ideal.

Finally, this paper builds a comparative nested measurement model and uses Mplus 7.0 software to compare and analyze the fitting degree of various nested measurement models to test the discriminant validity among the research variables. As shown in

Table 2. Discriminant validity analysis of research variables (N = 142).

Model	χ2	df	χ2/df	RMSEA	SRMR	CFI	TLI
Benchmark model	153.969	99	1.555	0.063	0.056	0.941	0.929
Two-factor model	197.932	103	1.922	0.081	0.062	0.899	0.882
Single-factor model	296.133	104	2.847	0.114	0.080	0.795	0.763

Notes: the benchmark model is a combination of business model innovativeness, scanning effort level and scanning sustainability level; the two-factor model is a combination of scanning effort level and scanning sustainability level; and the single-factor model is a combination of all variables.

, the benchmark model (three-factor model) has the best fitting effect when compared with the two competitive measurement models. This means that the three research variables measured by the 5-point Likert scale in this paper have high discriminant validity and can represent three different constructs.

5. Empirical Analysis and Results

5.1. Correlation Analysis

The mean value, variance and correlation analysis results of research variables are shown in

Table 3. Mean value, standard deviation and correlation coefficient of research variables (N = 142).

	Variable	1	2	3	4	5	6	7	8	9	10	11	12	13	14
1	Business model innovativeness	1
2	General cognitive schema	−0.565 **	1
3	Specific cognitive schema	0.462 **	−0.236 **	1
4	Scanning of unfamiliar fields	0.229 **	−0.196 *	0.244 **	1
5	Scanning of outside- industry fields	0.324 **	−0.182 *	0.231 **	0.632 **	1
6	Scanning effort level	0.613 **	−0.377 **	0.441 **	0.180 *	0.192 *	1
7	Scanning sustainability level	0.533 **	−0.210 *	0.360 **	0.188 *	0.117	0.691 **	1
8	Gender	−0.004	0.059	0.036	−0.039	−0.155	0.120	0.045	1
9	Age	−0.013	−0.225 **	−0.299 **	0.111	0.139	−0.129	−0.146	0.074	1
10	Educational level	−0.001	0.013	−0.019	0.134	0.085	0.051	0.012	0.095	−0.051	1
11	Age of the enterprises	−0.055	0.116	−0.122	0.093	−0.045	−0.081	−0.002	0.220 **	0.221 **	0.085	1
12	Industry	0.127	0.092	−0.005	−0.024	0.117	0.012	−0.036	0.184 *	0.081	0.023	0.192 *	1
13	Asset size	−0.162	0.127	−0.160	−0.027	−0.075	−0.154	−0.110	0.069	−0.116	0.109	−0.086	0.045	1
14	Region	0.149	0.020	0.190 *	0.181 *	0.311 **	0.077	0.084	−0.379 **	−0.203 *	−0.327 **	−0.323 **	−0.139	−0.131	1
Mean		3.751	9.880	14.289	45.845	46.014	2.745	3.868	0.578	0.690	0.669	0.528	0.183	0.768	0.676
SD		0.597	9.374	12.477	17.040	19.321	0.588	0.662	0.496	0.464	0.472	0.501	0.388	0.424	0.470

Notes: (a) Variable 1 refers to the mean; variables 2–3 are calculated values based on objective data; variables 4–5 are subjective evaluation values; variables 6–7 refers to the mean; variable 8 (if male) =1; variable 9 (if age 35 and below) =1; variable 10 (if educational level bachelor degree and above) =1; variable 11 (if age of enterprises is 1 or 2) =1; variable 12 (if information transmission, software and information technology services) =1; variable 13 (if the asset size is one million and above when established) =1; variable 14 (if Tianjin) =1. (b) * refers to p < 0.05; ** refers to p < 0.01.

. The results show that there is a certain correlation between the core study variables, which is not strong, indicating that there is no serious problem of multicollinearity between them.

5.2. Hypothesis Test

Hypothesis test analysis is carried out based on data standardization processing in this paper.

Table 4. Regression analysis results of entrepreneurs’ cognitive schemas on environmental scanning (N = 142).

Dependent Variable	Scanning of Unfamiliar Fields		Scanning of Outside Industry Fields		Scanning Effort Level		Scanning Sustainability Level
	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7	Model 8
Gender	0.030	0.002	−0.060	−0.086	0.187 **	0.164 **	0.094	0.066
Age	0.165 *	0.206 **	0.228 ***	0.269 ***	−0.122	−0.123	−0.151 *	−0.117
Educational level	0.239 ***	0.231 ***	0.243 ***	0.235 ***	0.088	0.076	0.037	0.027
Age of the enterprises	0.151 *	0.191 **	0.004	0.036	−0.090	−0.010	0.032	0.082
Industry	−0.029	−0.029	0.165 **	0.164 **	0.025	0.040	−0.030	−0.027
Asset size	0.025	0.089	−0.021	0.035	−0.187 **	−0.091	−0.122	−0.049
Region	0.352 ***	0.322 ***	0.435 ***	0.408 ***	0.102	0.081	0.091	0.062
General cognitive schema		−0.130		−0.098		−0.340 ***		−0.179 **
Specific cognitive schema		0.255 ***		0.229 ***		0.289 ***		0.271 ***
R2	0.127	0.216	0.217	0.283	0.087	0.317	0.049	0.167
ΔR2	0.127 **	0.089 ***	0.217 ***	0.065 ***	0.087 *	0.230 ***	0.049	0.118 ***
Value F	2.778 **	4.032 ***	5.320 ***	5.785 ***	1.824 *	6.792 ***	0.984	2.940 ***
VIF (max)	1.463	1.506	1.463	1.506	1.463	1.506	1.463	1.506

Notes: * refers to p < 0.1, ** refers to p < 0.05, *** refers to p < 0.01.

is the regression analysis results of entrepreneurs’ cognitive schemas on environmental scanning: according to Model 2, the general cognitive schema does not have a significant effect on scanning of unfamiliar fields, so Hypothesis 2a is not supported. However, the specific cognitive schema has a significantly positive effect on scanning of unfamiliar fields (β = 0.255, p < 0.01), so Hypothesis 2c is supported. According to Model 4, the general cognitive schema does not have a significant effect on scanning of outside industry fields, so Hypothesis 2b is not supported. However, the specific cognitive schema has a significantly positive effect on scanning of outside industry fields (β = 0.229, p < 0.01), so Hypothesis 2d is supported. According to Model 6, general cognitive schemas have a significantly negative effect on scanning effort level (β = −0.340, p < 0.01), so Hypothesis 5a is supported. A specific cognitive schema has a significantly positive effect on scanning effort level (β = 0.289, p < 0.01), so Hypothesis 5c is supported. According to Model 8, general cognitive schemas have a significantly negative effect on scanning sustainability level (β = −0.179, p < 0.01), so Hypothesis 5b is supported. A specific cognitive schema has a significantly positive effect on scanning sustainability level (β = 0.271, p < 0.01), so Hypothesis 5d is supported.

Table 5. Regression analysis results of the mediating effect of environmental scanning (N = 142).

Dependent Variable	Business Model Innovativeness
	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7	Model 8
Gender	0.051	0.034	−0.039	0.034	0.045	−0.027	0.011	0.000
Age	0.004	−0.043	0.050	−0.048	−0.077	0.003	−0.001	−0.013
Educational level	0.070	0.056	−0.016	0.050	0.026	0.028	0.046	0.019
Age of the enterprises	−0.062	0.052	−0.031	0.047	0.047	0.056	0.023	0.045
Industry	0.159 *	0.187 ***	0.125 *	0.188 ***	0.166 **	0.172 ***	0.197 ***	0.159 ***
Asset size	−0.162 *	−0.043	−0.052	−0.046	−0.048	−0.010	−0.026	−0.011
Region	0.173 *	0.164 **	0.034	0.155 *	0.111	0.133 *	0.142 **	0.099
General cognitive schema		−0.530 ***		−0.526 ***	−0.517 ***	−0.403 ***	−0.466 ***	−0.417 ***
Specific cognitive schema		0.293 ***		0.286 ***	0.264 ***	0.186 ***	0.197 ***	0.158 **
Scanning of unfamiliar fields			−0.011	0.027				−0.090
Scanning of outside industry fields			0.181 **		0.129 *			0.154 *
Scanning effort level			0.420 ***			0.372 ***		0.190 **
Scanning sustainability level			0.229 **				0.355 ***	0.250 ***
R2	0.075	0.487	0.466	0.488	0.499	0.582	0.593	0.622
ΔR2	0.075	0.412 ***	0.391 ***	0.001	0.012 *	0.095 ***	0.105 ***	0.135 ***
Value F	1.558	13.950 ***	10.308 ***	12.489 ***	13.065 ***	18.245 ***	19.054 ***	16.205 ***
VIF (max)	1.463	1.506	2.100	1.638	1.738	1.516	1.511	2.443

Notes: * refers to p < 0.1, ** refers to p < 0.05, *** refers to p < 0.01.

is the regression analysis results of the mediating effect of environmental scanning: according to Model 2, the general cognitive schema has a significantly negative effect on business model innovativeness (β = −0.530, p < 0.01), so Hypothesis 1a is supported. A specific cognitive schema has a significantly positive effect on business model innovativeness (β = 0.293, p < 0.01), so Hypothesis 1b is supported. According to Model 3, the three mediating variables, namely scanning of outside industry fields, scanning effort level and scanning sustainability level, have a significantly positive effect on business model innovativeness, with coefficients β = 0.181 (p < 0.05), β = 0.420 (p < 0.01) and β = 0.229 (p < 0.05), respectively, so Hypotheses 3b, 6a and 6b are supported. Scanning unfamiliar fields does not have a significant effect on business model innovativeness, so Hypothesis 3a is not supported.

For the mediating effect: according to Model 4, Model 5, Model 6 and Model 7 in Table 5, general cognitive schemas still have a significantly negative effect on business model innovativeness after adding four mediating variables, respectively, whose absolute value of the standardized coefficient is lower than that of Model 2 in Table 5, which is −0.530, and such an effect is still robust in Model 8 in Table 5. However, as in Table 4, the relationship between general cognitive schema and scanning of unfamiliar fields as well as the relationship between general cognitive schema and scanning of outside industry fields fails to pass the test. Therefore, it can only be determined that scanning effort level and scanning sustainability level play mediating roles between general cognitive schema and business model innovativeness; namely, Hypothesis 7a and Hypothesis 7b are supported while Hypothesis 4a and Hypothesis 4b are not supported.

According to Model 4, Model 5, Model 6 and Model 7 in Table 5, specific cognitive schema still has a significantly positive effect on business model innovativeness after adding four mediating variables, respectively, whose absolute value of the standardized coefficient is lower than that of Model 2 in Table 5, which is 0.293, and such an effect is still robust in Model 8 in Table 5. However, as in Table 3, the relationship between scanning of unfamiliar fields and business model innovativeness fails to pass the test. Therefore, it can only be determined that scanning of outside industry fields, scanning effort level and scanning sustainability level play mediating roles between specific cognitive schema and business model innovativeness; namely, Hypothesis 4d, Hypothesis 7c and Hypothesis 7d are supported while Hypothesis 4c is not supported.

5.3. Robustness Test

5.3.1. Screening Samples to Test the Robustness of Conclusions

Selecting some samples from the formal investigation samples for regression analysis is a common method to test the robustness of research conclusions. In this paper, the research object is mainly new ventures established by experienced entrepreneurs, while a small number of samples with no prior experience inside and outside the industry are also included. the score of general cognitive schema and specific cognitive schema is calculated based on the entrepreneurs’ previous work experience. Some studies have pointed out that whether entrepreneurs have previous experience or not often leads to significant differences in decision logic [28]. To avoid the interference of entrepreneurs’ experience, 10 samples with no prior experience inside and outside the industry are eliminated. a robustness test on the above-mentioned research conclusions is conducted based on the remaining 132 valid samples.

Table 6. Screening samples to test the robustness of the relationship between cognitive schemas and environmental scanning (N = 132). Regression analysis results of control variables to dependent variables are omitted here, and similar processing is also done in Table 7 and Table 8 below.

Dependent Variable	Scanning of Unfamiliar Fields		Scanning of Outside Industry Fields		Scanning Effort Level		Scanning Sustainability Level
	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7	Model 8
General cognitive schema		−0.164 *		−0.135		−0.377 ***		−0.200 **
Specific cognitive schema		0.214 **		0.187 **		0.271 ***		0.265 ***
R2	0.127	0.217	0.210	0.276	0.065	0.335	0.033	0.171
ΔR2	0.127 **	0.090 ***	0.210 ***	0.066 ***	0.065	0.271 ***	0.033	0.137 ***
Value F	2.574 **	3.760 ***	4.700 ***	5.166 ***	1.223	6.836 ***	0.608	2.789 ***
VIF (max)	1.430	1.433	1.430	1.433	1.430	1.433	1.430	1.433

Notes: * refers to p < 0.1, ** refers to p < 0.05, *** refers to p < 0.01.

presents the regression analysis results of the relationship between entrepreneurs’ cognitive schemas and environmental scanning after screening samples. The table shows that general cognitive schema have a significantly negative effect on the scanning effort and sustainability levels, so Hypothesis 5a and Hypothesis 5b still pass the test, which is consistent with the conclusion of previous researches. specific cognitive schema have a significantly positive effect on scanning of unfamiliar and outside industry fields and the scanning effort and sustainability levels, so Hypothesis 2c, Hypothesis 2d, Hypothesis 5c and Hypothesis 5d still pass the test, which is consistent with the conclusion of previous researches.

Table 7. Screening samples to test the robustness of mediating effect of environmental scanning (N = 132).

Dependent Variable	Business Model Innovativeness
	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7	Model 8
General cognitive schema		−0.561 ***		−0.564 ***	−0.544 ***	−0.437 ***	−0.497 ***	−0.452 ***
Specific cognitive schema		0.263 ***		0.267 ***	0.240 ***	0.174 **	0.178 **	0.153 **
Scanning of unfamiliar fields			−0.057	−0.016				−0.148 *
Scanning of outside industry fields			0.221 **		0.126 *			0.189 **
Scanning effort level			0.418 ***			0.328 ***		0.157 *
Scanning sustainability level			0.229 **				0.322 ***	0.241 ***
R2	0.069	0.527	0.468	0.527	0.538	0.598	0.612	0.644
ΔR2	0.069	0.458 ***	0.399 ***	0.000	0.012 *	0.071 ***	0.086 ***	0.118 ***
Value F	1.312	15.086 ***	9.606 ***	13.477 ***	14.106 ***	18.010 ***	19.123 ***	16.438 ***
VIF (max)	1.430	1.433	2.086	1.529	1.619	1.563	1.501	2.349

Notes: * refers to p < 0.1, ** refers to p < 0.05, *** refers to p < 0.01.

presents the regression analysis results of the mediating role of environmental scanning that plays between entrepreneurs’ cognitive schemas and business model innovativeness after screening samples.

The table shows that general cognitive schema and specific cognitive schema have significantly negative and positive effects on business model innovativeness respectively, so Hypothesis 1a and Hypothesis 1b still pass the test, which is consistent with the conclusion of previous researches. scanning of outside industry fields and the scanning effort and sustainability levels have a significantly positive effect on business model innovativeness, so Hypothesis 3b, Hypothesis 6a and Hypothesis 6b still pass the test, which is consistent with the conclusion of previous researches. Scanning of outside industry fields plays a mediating role between specific cognitive schema and business model innovativeness, so Hypothesis 4d still passes the test. scanning effort level and scanning sustainability level play a mediating role between two kinds of cognitive schemas and business model innovativeness, so Hypothesis 7a, Hypothesis 7b, Hypothesis 7c and Hypothesis 7d pass the test.

In a word, the regression analyses after the screening samples found that the hypothesis relationships between variables that passed the test in previous research are still significant in the robustness analysis, which is consistent with the conclusion of previous research, so the conclusions are robust.

5.3.2. Replacing Key Variables to Test the Robustness of Conclusions

Using different methods to re-measure the independent or the dependent variable, and then repeating the regression analysis based on the new measurement results, is another commonly used method to test the robustness of research conclusions. Therefore, this paper further adopts different methods to measure business model innovativeness to conduct a robustness test. Many domestic scholars have used the scale of Zott and Amit (2007) [24] to measure business model innovativeness. Most of their scales only include part of the items on the original scale, and the scale adopted in this paper can cover these items. Regarding this research, the item of “enterprises continuously promote business model innovation” is deleted in this paper. The remaining eight items are used to measure business model innovativeness for a robustness test. The regression results of the robustness test for the mediating effect of environmental scanning are shown in

Table 8. Replacing key variables to test the robustness of mediating effect of environmental scanning (N = 142).

Dependent Variable	Business Model Innovativeness
	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7	Model 8
General cognitive schema		−0.529 ***		−0.525 ***	−0.517 ***	−0.411 ***	−0.471 ***	−0.422 ***
Specific cognitive schema		0.282 ***		0.273 ***	0.253 ***	0.182 **	0.193 **	0.154 **
Scanning of unfamiliar fields			0.005	0.036				−0.074
Scanning of outside industry fields			0.173 *		0.129 *			0.147 *
Scanning effort level			0.413 ***			0.348 ***		0.181 **
Scanning sustainability level			0.205 **				0.328 ***	0.227 ***
R2	0.074	0.478	0.436	0.479	0.489	0.560	0.567	0.594
ΔR2	0.074	0.403 ***	0.362 ***	0.001	0.012 *	0.083 ***	0.090 ***	0.117 ***
Value F	1.532	13.407 ***	9.137 ***	12.024 ***	12.560 ***	16.691 ***	17.182 ***	14.421 ***
VIF (max)	1.463	1.506	2.100	1.638	1.738	1.516	1.511	2.443

Notes: * refers to p < 0.1, ** refers to p < 0.05, *** refers to p < 0.01.

. The analysis results are consistent with the previous research conclusions, so the conclusions are robust.

6. Conclusions, Contribution and Prospects

6.1. Research Conclusion and Discussion

6.1.1. The Relationship between Entrepreneurs’ Cognitive Schemas and Business Model Innovativeness

Entrepreneurs’ cognitive schema has a very important influence on entrepreneurial activities, which not only depends on the richness of cognitive schema but depends more on the type of cognitive schema. Different types of cognitive schemas often lead to great differences in the direction and effect of entrepreneurial activities. Cognitive schema comes from previously accumulated experience. Different types of experience shape different cognitive schemas. Previous research in entrepreneurship has distinguished experience types according to multiple criteria. Some scholars distinguished experience types according to work functions [36]. Some scholars distinguished them according to geographical or industrial similarities [33]. Some scholars distinguished them into management experience and entrepreneurial experience [37]. This paper mainly distinguishes the experience types according to the industrial similarity and further defines general cognitive schema and special cognitive schema based on the neo-institutional theory of organization.

As the empirical analysis results show, general cognitive schemas make entrepreneurs less likely to design highly innovative business models, while various specific cognitive schemas make entrepreneurs tend to design highly innovative business models. This conclusion is consistent with the findings of institutional entrepreneurship research, which emphasizes that organizations at the edge of the field are keener on institutional change [38]. From the perspective of individual institutional entrepreneurs, Cliff et al. (2006) [20] also found that if entrepreneurs previously worked in organizations at the edge of the field or outside the field, they will question the current popular business model in the field and be keen to adopt a new one. If the entrepreneur has previously worked in the core organization in the field, they will still stick with and maintain the prevailing business model in the field.

6.1.2. The Relationship between Environmental Scanning Activities and Business Model Innovativeness

It has been widely recognized in innovation research that new information leads to more opportunities for innovation, which is conducive to success. New information is collected by the entrepreneurs’ unique environmental scanning activities, which require them to remove the inherent bias and pay attention to the unusual and even seemingly unreasonable environmental information. New information can be obtained directly from unfamiliar or outside industry fields, and such a possibility can be improved by enhancing the level of effort and sustainability of scanning. The above viewpoints have been fully confirmed in fields, such as product innovation and technological innovation [31]. In the field of business model research, scholars have also found that entrepreneurs can design disruptive business models by referring to and modifying business models from outside the industry [22].

As the empirical analysis results show, entrepreneurs are more likely to design a more innovative business model when they scan the information outside the industry with a higher level of effort and sustainability. These findings are consistent with previous research in the field of innovation. However, although many pieces of research in the field of innovation have confirmed that the innovation level of products or technology is more likely to improve as more unfamiliar information is scanned, the relationship between scanning of unfamiliar fields and business model innovativeness has not been tested in this paper. The specific reason may be that entrepreneurs who scan information outside of the industry usually have clear target industries they want to focus on, which is difficult for entrepreneurs who scan unfamiliar fields for information. As a result, there will be information chaos, which is not easy to extract for feasible and highly innovative business models.

6.1.3. The Relationship between Entrepreneurs’ Cognitive Schemas and Environmental Scanning Activities

Unique environmental scanning activities are important for highly innovative business model design. Therefore, what kinds of entrepreneurs tend to carry out such activities? Research in the field of innovation seldom pays attention to this issue, but there is in-depth exploration carried out in the field of entrepreneurship. Related research has focused on the analysis of how expert entrepreneurs with rich work or entrepreneurial experience view the external environment and the results of consequent entrepreneurial activities. This research found that expert entrepreneurs view the external environment in different ways based on the similarity between the cognitive schemas derived from their experience and the external environment. When they have high similarity, expert entrepreneurs mostly rely on intuition to make decisions, and they will pay little attention to external environmental information with the low level of effort of scanning. When they have low similarity, the intuition does not work. Expert entrepreneurs will invest a lot of time and energy in environmental scanning activities, and further interpret environmental information through methods, such as categorical reasoning and structural matching, to make the final decision [28,39,40].

Based on the above viewpoints, the empirical analysis results are consistent with this research. Specifically, the general cognitive schema will lead to a decrease in the entrepreneurs’ scanning effort and sustainability; the specific cognitive schema will drive the entrepreneurs to pay attention to the information in the unfamiliar and other industry fields, requiring a high level of effort and a long time to carry out environmental scanning activities. The negative relationship between general cognitive schema and scanning of unfamiliar and other industry fields is not verified. The reason may be that, even within the industry, attention should also be paid to the information that is generated by information technology and could reverse the current order from outside industries though it is unfamiliar to the entrepreneurs because such information could have a great influence on the industry’s traditional business rules.

6.1.4. The Mediating Effect of Environmental Scanning Activities

As the empirical analysis results show, the effort and sustainability levels of scanning play a partial mediating role between the general cognitive schema and business model innovativeness. General cognitive schemas lead entrepreneurs to be eager to simplify decisions and unwilling to invest time and energy to focus on external information, which ultimately leads to their failure of highly innovative business model design. This finding is consistent with research conclusions related to strategic cognition. Scanning of outside industry fields, effort and the sustainability level of scanning play a partial mediating role between the specific cognitive schema and business model innovativeness. A specific cognitive schema can drive entrepreneurs to design a highly innovative business model. The partial reason is that they attach importance to environmental scanning activities, willing to invest a lot of time and energy in them and pay close attention to new information and new knowledge outside the industry, which all make them easier to develop new thinking for business model design.

6.2. Theoretical Contribution

First of all, based on the “antecedent-consequence” orientation, this paper intrinsically explores the reason why new ventures can design business models of innovativeness. There are typical cases of rapid growth under highly innovative business models, such as Alibaba, Apple and Airbnb. Such cases have triggered scholars’ thinking about “why these enterprises can design highly innovative business models”, and relevant research results will help deepen the research on business model design. The currently popular “process- consequence”-oriented research in business model design focuses on describing the business model design process [4]. In contrast, the “antecedent-consequence”-oriented research focuses on discovering the internal and external reasons why new ventures can design different business models, which is more suitable to solve the above problems. However, the current research on design antecedents of business models is chaotic and the mechanism of design antecedents is poorly understood. Based on the “antecedent-consequence” orientation, this paper focuses on the antecedent of business model design, the entrepreneurs’ cognitive schema and discusses its influence on the innovation of business model design results of new ventures, which can make up for the deficiency in existing research in business model design.

Secondly, this paper distinguishes the types of entrepreneurs’ cognitive schemas about business models and analyzes the different roles that different types of cognitive schemas play in the process of business model design. Cognitive schemas guide people to understand the external environment to carry out various activities, which will greatly affect the results of business model design. Although scholars in the field of business models have long noticed the importance of entrepreneurs’ cognitive schemas, the theoretical basis for relevant research is relatively weak. They only emphasize the hindrance of cognitive inertia of cognitive schemas to the new business model design, with no detailed definition of cognitive schema category [8]. Different entrepreneurs have different cognitive schemas. Some cognitive schemas bring entrepreneurs cognitive inertia, while some cognitive schemas bring cognitive flexibility, leading them to design highly innovative business models [7,41]. Because of this, entrepreneurs’ cognitive schemas about business models are divided into the general cognitive schema and specific cognitive schema in this paper, with reference to the neo-institutional theory of organization. Empirical data are used to verify the differentiated impact of the two cognitive schemas on the innovativeness of new ventures’ business models. This can not only overcome the lack of theoretical understanding of the category of cognitive schemas in current business model research from the cognitive perspective but also help understand the relationship between cognitive schemas and business model design results more comprehensively.

Thirdly, integrating the viewpoints from both positioning and cognitive perspectives, the mediating effect of entrepreneurs’ cognitive schemas on the relationship between “cognitive schemas—business model innovativeness” is systematically revealed in this paper. Combined with the above analysis, it is more reasonable, based on the “antecedent-consequence” orientation, to discuss the business model design from the perspective of cognitive schema. However, the analysis process of existing research is too microscopic, ignoring the influence of the external environment [22]. Both environmental and cognitive factors partially determine the results of business model design [7]. Only emphasizing environmental factors and ignoring cognitive factors will easily make the research deviate from the decision-making practice due to the excessive pursuit of rational assumptions. Only emphasizing cognitive factors and ignoring environmental factors will make the research fully idealistic. Therefore, with reference to the neo-institutional theory of organization, this paper establishes the logic of “cognitive schemas-environmental scanning-business model innovativeness” based on viewpoints from both positioning and cognitive perspectives, and empirically tests the environmental scanning, which can fully present the mediating role that cognitive activity of “cognition-environment” plays in the relationship between cognitive schemas and business model innovativeness, rather than just focusing on the conceiving process of business model in entrepreneurs’ mind, which can more systematically show the business model design process.

6.3. Practical Inspiration

On the one hand, the reason why entrepreneurs design business models with different innovativeness is ultimately related to their knowledge. Although the high innovative business model could not be the most appropriate choice, entrepreneurs must hand over the power and consider absorbing outside industry talents into their founding teams to provide unique insights for the business model design process if they do not have relevant experience but want to establish competitive advantages with the help of highly innovative business models.

On the other hand, under the impact of Internet information technology, the traditional industry has become a very blurred border. Therefore, it is often difficult for entrepreneurs to accurately describe their competitors. This requires that they should pay particular attention to the information outside the industry when they keep an eye on the environmental information to ensure that their business model designs can both meet the current needs and effectively respond to the era development. However, it is not easy to pay attention to and understand information outside the industry. Therefore, entrepreneurs should maintain their courage and perseverance to change. They need to actively change their intrinsic cognition to break the existing knowledge about the business model in their minds to design a highly innovative business model to promote better and faster development of new ventures through creatively combining the knowledge with external environment information.

6.4. Deficiencies and Prospects

Some shortcomings from this paper need to be improved in future research. First, the sample size is not large enough, which may reduce the representativeness of the conclusion. In future research, in addition to increasing the sample size through supplementary investigation, resources can also be concentrated on a few industries to conduct a precise investigation. Second, the data are mainly collected by commissioned professional research companies and the person in charge of the incubators, so there may be measurement errors with the data collected by the researchers themselves because of different understanding. In future research, we can enhance training and process control (e.g., follow-up visits), provide reference materials and increase the proportion of personal research by researchers. Further, the questionnaire should be further streamlined and include more entrepreneurs’ opinions, and the obscure expressions should also be changed. Third, the scale developed by Zott and Amit (2007) [24] is mainly used for measuring business model innovativeness. However, this scale is mainly used for analysis and coding work by experts and scholars themselves, which may increase the difficulty for entrepreneurs to understand the measurement items. Therefore, in the future, it is necessary to strengthen the theoretical analysis of business model innovation and then develop a new measurement scale of business model innovativeness. Fourthly, based on this paper, the logic of “cognitive schemas-environmental scan-business model innovativeness” should be deeply explored with qualitative analysis methods to more fully reveal the mechanism of cognitive schemas’ influence on business model innovativeness.