CHAPTER 5. Evaluation
5.2 Experiment Details for Proposition 1
國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
79
Proposition 2-C: For SMEs who has lesser knowledge in exploiting the benefits of interaction, the effectiveness of our proposed model will be higher.
After validating our proposed model, we can further test the usefulness of the information system we designed as the supporting artifact for SME users to use our proposed model. Hence, the last proposition within this research will be:
Proposition 3: The proposed information system can effectively help SMEs analyzing and managing interaction pattern to create service innovation and improve their service value.
The next section will describe the details and results of the experiments for the above mentioned propositions.
5.2 Experiment Details for Proposition 1
5.2.1 Experiment Challenges and Design Principles
The hypothesis 1 is “Does interaction serves as an important factor in business service innovation”. To make the experiment result to be easy to understand and convincing, we intend to find a proper comparison target which can benchmark with interaction to see the effectiveness and importance of interaction within businesses through by comparing the comparison target with interaction. The choosing of the benchmarking target will be very important because the experiment result will be persuasive and robust enough only if the target is a widely recognized business making theory; also it must be simple enough to be manipulated in an experiment to
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
80
easily see the difference. Moreover, the comparison target must show a great difference than interaction to show the difference clearly between two concepts.
After researching, a good candidate fitting the requirement of the comparison target is the resource based view of alliance by Das & Teng (2000). The resources aspect of alliance has been a widely accepted theory, and its usefulness is proved already; if we can benchmark our proposed model with the resourced based of alliance theory, the significance of the experiment will be sufficient thus convincing. Also, the main concept of resource base theory – company alliance to attain resource they required - is an easy-to-adopt theory because of its simplicity. Lastly, the resource based view of alliance is showing the feature of goods dominant logic by claiming that company choose partners based on the resource it lacks or is considered important (like an operand), thus is showing great difference from our research that encourages a company to find good partner candidates through interaction analyzing and manipulating, which is a more service dominant logic concept. Accordingly, we intend to use the resource based of alliance as the comparison of interaction in terms of business service innovation and value creating.
After selecting a proper comparison target, to model the differences between resource aspect and interaction aspect, a comparison basis which can show their difference must be discovered or developed. Considering the characteristics of these two aspects, we use the functions of alliance as our comparison basis, which refers to a list of alliance functions that might happen between two partners that were claimed in researches of Varadarajan & Cunningham’s (1995) and Todeva & Knoke (2005).Some examples of these functions are like franchising, co-marketing, and joint innovation. We believe that SMEs with different points of views will do the alliance partner selecting based on different reasons. For resource based partner selecting advocate companies, because they intend to find for partners based on the perceived
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
81
ability and resource which they do not have, we can assume they will choose the alliance function to execute with its partner based on partner’s ability and resource.
However, for those SMEs who take the interaction point of view, because interaction is the key of their alliance, so instead of choosing which function to execute based on partners ability and resource, these SMEs choose the functions to execute more based on their interactions. Concluding, in the experiments to verify the effectiveness of interaction within service innovation and service value creating, resource based company will choose alliance function based on partners’ perceived abilities and resource, while interaction based company will choose the alliance function based on their interaction condition. By using the alliance function, we can easily model two different concepts within our experiments, and verify whether interaction based stands an important in service innovation or not through comparing their alliance performance.
However, although we can develop proper experiment method, there are still more difficulties to tackle; one of the most critical problem is that it is hard to find company owners whom had experience in alliance for service innovation and are willingly to perform both theory of alliance to see the differences. Secondly, the time to see the differences between executing the two aspects might be very long and not controllable within our research. Finally, there might be other factors which are not considered in this research that can influence the results of our experiments, like government policies or economic conditions.
To overcome the research challenges in this research, we intend to use simulation techniques to testify the propositions. Simulation technique serves an excellent analyzing tool for problems that are impossible or extremely expensive to observe the changing in real world, but are possible to analyze if proper and able-to-validate model is formulated (Maria 1997). By using simulation, we can design models of
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
82
company’s service innovation performance in alliance to test whether resource aspect or interaction aspect will do better for companies without influenced by the time and other uncontrollable constraints. On the other hand, through simulation, we can avoid using only several types companies because of the aforementioned constraints in this research, which enables us to come out with a research contribution that is regarded as without loss of generalities. Concluding, simulation techniques enables us to observe the difference between resourced based theory and interaction based theory by simulating companies apply different approaches, and the service outcome and value it acquires if we are able to design a proper and convincing simulation model.
5.2.2 Experiment Design Details
During the simulation process, we will generate 100 SMEs and 10,000 customers in a tourism attraction region (to simulate the Pillow Mountain Leisure Agriculture Area). Each of the SMEs provides one type of service - eat, accommodation (live), or entertainment to customer; and each of the SMEs has their own ability in marketing to link to its customers. Each type of the service is further extended into three styles, for example like eat style 1, 2, 3. Here we assume that each customer needs all but different styles of service, and will not go to SMEs that don’t provide the exact services styles he wants. Given the assumption that each SME only provides one type of service, alliance is required to attract customers. The outcome value of the service innovation of the SME’s alliance will be calculated by the number of customers the SME can serve. Figure 5.1 shows an example. (SME A acquired Customer A (match) through obtaining ability live 2 and entertain 1, and improved marketing ability from 1 to 4 from the alliance with SME B and C; however, the alliance did not give the proper ability and marketing ability to acquire Customer B and C (not match) , so SME A is only able to acquire Customer A.)
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
83
Figure 5.1 Example of SME’s alliance and customer acquiring
Alliance Successful Rate:
In order to make the scenario within the simulation closer to real world, we create a factor - alliance successful rate, which represents the situation that alliancing or cooperating with other companies does not always bring benefit to business. The alliance successful rate is considered as the possibility that SMEs could obtain positive value outcome from their alliance, for example, when alliance successful rate in the entire environment is low, SMEs will have low chances of gaining benefits from their alliance; contrarily, if the rate is high in the context, then SMEs will possibly gain more new ability through their alliance.
Other than the first reason for using alliance successful rate within our experiments, it also serves as an indicator about whether SMEs in this region is good at building alliance. Reason for this is because one of this research’s objectives is to provide good support for SMEs who don’t know how to do service innovation from alliance, to model this feature, we intend to use low alliance successful rate to represent these SMEs.
Also, we intend to use this factor to demonstrate some differences between resource based aspect SMEs and interaction based aspect SMEs. In this research, we
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
84
consider that SMEs of the resource based aspect will put their efforts into enhancing this factor because they choose partners with higher accuracy (e.g., choosing partners on their perceived ability), and hence will put more emphasis on improving the effectiveness of this alliance - which shall lead to a higher chance of building up successful alliance; on the other hand, interaction based aspect SMEs will focus on other things (will be discussed later), which indicates a lower chance of a successful alliance building. Through the alliance successful rate, we can show different characteristics of SMEs ability in building alliance, and demonstrate them in our simulation process to see if any possible findings will occur.
Alliance Functions and Alliance Constraints:
Although the alliance successful rate could demonstrate some different features of the two aspects, it still did not mention about how to present the alliance function in our simulation process, which is necessary within this experiment to prove the importance of interaction in service innovation and service value creation within alliance. In order to model the resource aspect and interaction aspect of alliance, we select 5 functions of alliance according to Varadarajan & Cunningham’s (1995) and Todeva & Knoke (2005) researches that are considered to be fitting into SME’s situation and attaining the benefits for service innovation. These functions include joint innovation, co-marketing, co-servicing, co-service & marketing and franchising.
Each function will lead to a different outcome, like joint innovation will possibly lead to a new service gain, and co-marketing will enhance SME’s marketing ability.
In the simulation process, each alliance function will have its alliance constraints, which is designed and based on the function’s feature, and will serve as a factor to model different context situations. SMEs of the resource based aspect, by definition, will choose alliance function to perform based on the perceived ability of their
‧
SMEs of resource based aspect will have to follow the alliance constraints because of their alliance partner selecting strategy and thus limit their possibility of trying other alliance possibility. On the contrary, SMEs of interaction aspect do not follow the alliance constraints because they do not decide the alliance function by the perceived ability, instead by the interaction details. Through the alliance function constraints, we can separate the focus of the two different aspects and see their comparisons. Details of the alliance functions are listed in Table 5.1. ( Ex: If two SMEs: SME A and SME B both are in the service type “eat”, if they are following the resourced aspect alliance choosing strategy, the possible alliance function will exclude Co-Service & Marketing because the alliance was constrained. However, if they follow the interaction aspect, it is still possible for SME A and SME B to have the Co-Service & Marketing alliance function between them.)Table 5.1 Exemplar alliance constraints
Alliance functions Alliance constraints Result
Joint Innovation No Both SMEs acquire a new type of service or new style of service which they didn’t have before
Co-Marketing No SMEs acquire partners marketing ability
Co-Servicing Happens within same service types SMEs
SME A acquires SME B’s service type and style.
SME B acquires SME A’s service type and style.
Co-Service &
Marketing
Happens within different service types SMEs
SME A acquires SME B’s service type and style.
SME B acquires SME A’s marketing ability Franchising Happens within same
service types of SMEs
Franchising SME acquires partner’s marketing ability
The franchising partner acquires the franchiser SME’s service type and style.
Following the above mentioned simulation design details, we can design different settings in each simulation round to model the different service system context. In Table 5.2, setting 1 is the comparison basis – SMEs without choosing resources based aspect and interaction based aspect. Setting 2 stands for SMEs of resources based
‧
focusing on selecting partners base on their perceived abilities (thus following alliance function constraints), and focus on improving the alliance to have stronger alliance outcome ( thus having higher alliance successful rate). The last setting is setting 3 of the same level of successful rate with setting 1, in which the setting represents the SMEs with higher interaction tendency by not following the constraints of alliance (i.e., the situation with SMEs more focusing on finding the most proper value proposition provided by the partners, instead of their partner’s core ability).Table 5.2 Details of each setting
Settings Successful Rate Alliance Constraints Represents
Setting 1 Low SME will follow the
constraints
Comparison basis
Setting2 High SME will follow the
constraints
SMEs with resource based aspect
Setting3 Low SME will not follow the
constraints
SMEs with interaction based aspect
5.2.3 Experiment Results and Conclusions
Figure 5.2, 5.3, and 5.4 show the results of simulation. Figure 5.2, 5.3 stands for the different ratios of new ability gained and marketing ability gained through alliancing with adding one more new partner in different settings. The ability and the marketing ability are acquired through the alliance with other SMEs, and decided by the type of service the partner is servicing and the alliance function they are executing. Our simulation program simulates alliance building and ability gaining processes, accumulating the total ability gained and then calculating the ability gaining ratio.
Figure 5.4 shows the distribution of SMEs with different amount of customers, like the number of SMEs who can acquire more than 1000 customers, or the amount of 3000-customers- acquirable SMEs. The number of customers a SME could acquire is
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
87
computed by SME’s ability and marketing ability and the customer’s taste (Like Figure 5.1). For example, if SME A acquired the ability Eat 1, Live 2 and Entertain 1 by alliance, then the simulation program will count the number of customers whose requirement are as just as SME A’s service.
The results of setting 1~3 show that while SMEs are focusing on the more resource aspects, the ratio of their ability increased by per new partner is much higher than SMEs that are not focusing on resource based aspects. By these two figures, it is obvious that through concentrating on resource based aspects, SMEs are having better efficiency in ability gaining from alliance.
However, from Figure 5.4, we can see that setting 2 and 3 are showing some different characteristics from Figure 5.2, 5.3. In Figure 5.4, the settings 1, 2 indicates that a much higher number of entities is attaining inferior outcome value in terms of the number of customers obtained (the left hand part), however setting 3 shows a much better outcome value in terms of acquiring more customers by having more entities in the right hand part in Figure 5.4 than setting 1,2 .
Figure 5.2 New service obtained per new partner
0.60 0.65 0.70 0.75 0.80
Setting1 Setting2 Setting3 New service gained/ per new partner
‧
Figure 5.3 Marketing increased per new partner
Figure 5.4 Actual market size distributions of all settings
Combining the results of the three figures, while Figure 5.2, 5.3 indicates that SME’s has a higher performance in ability gain under a higher resource based tendency in alliance situation, Figure 5.4 shows some SMEs although gaining more ability per new partner in resource-focusing, in fact, are not doing better than higher interaction tendency SMEs. A possible explanation of this contradiction is that even though high resource based focused SMEs are more efficient in finding partners to increase their abilities, they might fail to meet to their customers’ expectations.
However, for SMEs with higher interaction tendency can find partners that are able to
0.00 Marketing ability gained/ per new partner
0
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
89
provide critical components to its service value proposition to the customers, hence results in a better consequence.
If applying these explanations to our propositions, we can see through Figure 5.4 that the comparison of setting 3 with setting 1, 2 is showing that interaction-centric SMEs are doing much better in alliance for service innovation than resource-centric SMEs. Accordingly, we can say that our proposition – interaction serves an important role in service innovation and service value creation is supported, because it has the same even better performance- acquire more customers, than the well-recognized resourced base alliance selection approach in this simulation. Hence, we could argue that interaction is an important factor in SMEs alliance building if they pursue better outcomes from the alliance, and are able to go on with our other experiments given this premise being justified.