“Customer perspective on dynamic marketing capability in international hotels of Indonesia during Covid-19: Confirmatory factor analysis”

ARTICLE INFO Evo Sampetua Hariandja (2021). Customer perspective on dynamic marketing capability in international hotels of Indonesia during Covid-19: Confirmatory factor analysis. Innovative Marketing , 17(3), 74-87. doi:10.21511/im.17(3).2021.06 DOI http://dx.doi.org/10.21511/im.17(3).2021.06 RELEASED ON Monday, 16 August 2021 RECEIVED ON Tuesday, 01 June 2021 ACCEPTED ON Wednesday, 04 August 2021


INTRODUCTION
The Coronavirus disease has an extreme effect on the world supply chain, domestically on financial market volatility and buyer request stuns, and negative impacts in key segments such as travel and tourism. The Coronavirus outbreak's consequences will undoubtedly be felt in the tourism value chain. Small and medium-sized businesses are expected to face significant difficulties (Sugihamretha, 2020). With this pandemic, the emergence of restrictions such as locks, closure of national borders, and rules of guarding distance imposed by many countries ultimately lowered the world economy, including the tourism industry, which includes the hotel industry (Japutra & Situmorang, 2021). According to Indonesia Statistical Agency (BPS, 2020), with the Covid-19 pandemic issue, cumulatively (January-February 2020), the number of foreigners visiting or traveling to Indonesia was down to 2.16 million arrivals or decreased by 11.80 % in comparison to the number of foreigners visiting in the same period in 2019, which was 2.45 million visits. The tourism industry is expected to lead the recov-

LITERATURE REVIEW AND HYPOTHESES DEVELOPMENT
1.1. Dynamic marketing capability Teece et al. (1997) defined dynamic capability as a company skill to implement, form, and respond to change in position to react to rapidly changing circumstances. Morgan et al. (2009) and Eisenhardt and Martin (2000) stated that the potential of a company to gain new and creative competitive advantages is referred to as dynamic capability. The term DMC refers to the specialized, structural, cross-functional, and contextual factors that enable the acquisition, combination, and transformation of marketing resources into value-added offerings for the target market. Additionally, dynamic marketing capabilities have increased the company performance in foreign markets, increasing the company level of international growth and the ability to maintain a competitive edge (Ledesma-Chaves et al., 2020).
In the marketing sector, DMC plays a significant part in the development of innovative products and services, pricing, advertising, shipment, sales, and partnership marketing (Walugembe et al., 2017). According to Song et al. (2005), Dutta et al. (1999), Day (1994), and Hariandja (2016a), marketing capabilities are defined as a way to coordinate organizations by leveraging the assets and capabilities of the organization to understand the client needs and make various types of goods that are differentiated from competitors and gain brand prevalence. Companies use DMC to reconfigure marketing tools in response to changing market conditions. In keeping the company's offering innovative, companies should consider developing a DMC as this approach takes advantage of the speed and market success of corporate innovation (Mitręga, 2020). In an unpredictable market situation, DMC is characterized as an organization's specific objective of creating, publishing, and integrating market awareness management processes to fulfill the customer value proposition (Hoque et al., 2021). According to Kamboj et al. (2015) and Konwar et al. (2017), in responding to changes in market conditions, DMC is required in certain marketing activities that refer to the effectiveness of company routines that support the company capabilities.

Market sensing
Market sensing is indeed the ability to recognize and respond to market changes, such as digital development market preferences and desires, creativity, and valuation proposition. Market sensing is also the wide development of marketing strategy from an organizational-to-customer needs and wants, as well as the implementation of this intelligence through the institution's organizational functions . Market sensing is capable of demonstrating its value in enhancing the success of hotels that pursue operational excellence (Santra, 2018). Market sensing enables companies to gain a greater understanding of market demands and how to counteract potential competitors' behavior (Buccieri et al., 2020). Furthermore, market sensing is important because corporations frequently miss a full knowledge of their own capabilities, and sensing should be more about customers' wants and needs, but also initiatives undertaken by competitors, according to Schreyogg and Kliesch-Eberl (2007) and Dias (2013).
According to Kok et al. (2003), Day (1994), and Hariandja et al. (2014), market sensing capability refers to a company's capacity to learn from clients, partners, and rivals in turn to sense, manage, and use data, but also to respond continually to changes and conditions in new and current markets. Rakthin et al. (2016) and Alshanty and Emeagwali (2019) stated that market sensing capability ensures that the quality of goods or services delivered is higher, post-sales services are provided to retain current customers, and the firm can accomplish additional marketing and financial feats. Firms need sensing capabilities, according to Day (1994) and Teece (2007), to identify opportunities and risks in their market environment. One type of sensing capability is market sensing capability. It includes the ability to gather and extract market data from both external and internal sources, view it in context, and draw actionable conclusions that can reduce the uncertainty of the market development system and enhance opportunities for successful products and services to be marketed (Lin & Wang, 2015).
Market sensing capability is divided into several groups by Day (2002). First, sensing practices are related to collecting and disseminating data about consumer needs, desires, perceptions, market differentiation, partnership continuity, and competitors' abilities and goals. Second, one should analyze sensed details, which corresponds to comprehending the data collected. Third, there are assessing practices that are primarily concerned with input reviews and audits (Likoum et al., 2020).
There are many practices involved in performing market sensing in the tourism industry. Sensing involves a receptive attitude against market information and a wide implementation across tourist attractions. Sense-making implies surveilling the theoretical frameworks of people from various organizations to prevent mob mentality and facilitate the proper application of accumulated business intelligence, and eventually, regulations are implemented to evaluate the optimal market sensing process' efficiency through a reflective process (Murray et al., 2016).

Market learning
Learning and knowledge management were interpreted by Zollo and Winter (2002) as a way of promoting the development of dynamic capabilities. The adaptability of marketing capabilities enables the creation of learning and experimentation strategies, perhaps through the advancement of expertise or business expansion. Market learning leads a company to fresh and diverse information regarding markets and rivals that differs from the company's current expertise, awareness, and interactions. The primary benefits are increased industry awareness diversity and prospects for further new ideas (Kim & Atuahene-Gima, 2010). Schneider et al. (2002) argue that learning how to build multiple dynamic capabilities simultaneously is critical since combining multiple capabilities creates more shareholder value than focusing on a single capability. It is important to understand the opportunities for concurrent learning between dynamic capabilities because it encourages gaining a deeper awareness of the effect of high-value strategic activities on learning, intelligence, and knowledge transfer (Bingham et al., 2015).
Market learning gathers learning through trends in consumer behavior and competitive conditions, while marketing capability gathers the company opportunity to blend the advertising combination components to effectively meet the target market.
Companies that perceive the marketplace as vol-atile would begin to grow the competitive market in a dynamic situation, and firms would continue to pursue stronger learning, reviewing the existing methods to find new ways of satisfying its consumers, by attempting to draw on corporate observational learning and the competence perspective of strategic plan (Weerawardena et al., 2006

Market targeting and positioning
Market targeting, also known as positioning capability, contributes to the company's ability to identify alternative markets and then choose appropriate market targets that are targeted for optimal impact. Combining resources and skills with changing markets necessitates the consideration of marketing competencies (Hariandja et al., 2014). Demand heterogeneity and distinct economic potentialities may be essential in identifying the most important target markets, as well as subsequent positioning and product or service growth (Akdağ et al., 2017).
position, as well as surviving during times of crisis. Moreover, the dynamic development of innovative strategic positioning through new goods, services, and business concepts reinforce how this process was a dynamic perception of a strategy whereby an organization achieves successful sustainable excellence (Sriboonlue, 2016).
Market targeting gains include making new tourism offerings that specifically satisfy the requirements and desires of various traveler segments, developing more efficient marketing campaigns to access and fulfill the identified traveler segments, also optimizing the tactical application of promotion activities to the greatest interesting prospects in the tourism industry. While positioning is the process of creating a distinct identity for a specific location in the minds of prospective travelers that distinguishes it from acceptable alternatives (Pyo, 2015). A hotel marketing of the importance of privacy and comfort may serve as a source of differentiation and positioning for hotels attempting to attract the market (Khoo-Lattimore & Prayag, 2015).
To achieve service excellence, hotels should use a range of dynamic marketing capabilities, such as market sensing, learning, market targeting and positioning. Elbanna and Elsharnouby (2018) discovered how hotels in complex and dynamic environments must create complex DMCs, one of which includes identifying a desired targeting and positioning. With this capability, hotels may conduct classifying marketing activities in a more informative and data-driven manner (Elsharnouby & Elbanna, 2021). Based on the literature review, the study proposes the following hypotheses: H1: There is a significant influence from market sensing on DMC.
H2: There is a significant influence from market learning on DMC.
H3: There is a significant influence from market targeting and positioning on DMC.

METHODS
This is a quantitative study with a descriptive research methodology and PLS-SEM data analysis.
This paper employs the Smart-PLS version 3.3.2 software to develop a measurement model based on validity and reliability testing, as well as to assess the structural model by obtaining substantial results on the relationship between variables after the data are collected from 530 respondents through distributed questionnaires. As part of the inferential statistical analysis, the measurement model is presented as an outer model, while the structural model is presented as an inner model. The characteristics of the respondent sample and general data descriptions were examined using statistical analysis of frequencies.
Additionally, this analysis employs SPSS 26 software to determine common method variance.
The questionnaire included examines three facets of DMC: market sensing, market learning, and market targeting and positioning. The statement in the questionnaire refers to a recent global situation, with the background set in 3-5-star international hotels in Indonesia. In this study, an interval measurement scale was used, namely by using the Likert scale: 1 means "strongly disagree", while 5 means "strongly agree." This study examines market sensing using four scales developed by Day (1994)

RESULTS
According to data collected from 530 respondents who are Indonesian citizens, female respondents counted 332 and represented 62.6%, while male respondents counted 198 and represented 37.4%. The questionnaire for this study was filled out by the majority of female respondents from the gender group. Following that, the majority of the 272 respondents who participated in filling out the questionnaire were between the ages of 20 and 24. However, the age distribution of the data collection is thought to be in line with the desired characteristics, which contributes to the response, including young consumers who had stayed at 3-5-star international hotels in Indonesia. According to the statistics from the questionnaire findings, the majority of respondents who have stayed at 3-5-star interna-tional hotels in Indonesia in this study came from Tangerang, with 187 respondents. These Gen Z are often regarded as the next generation of travelers, and they wield surprising power in family travel decisions (Wiastuti et al., 2020). This is demonstrated by 80.9% of those who filled out a questionnaire with a career as a college student, 80.6% with an income less than Rp10,000,000. The majority of respondents in this study who had stayed at 3-5-star international hotels in Indonesia 1-2 times per year were 354 respondents. The vast majority of respondents (289 in total) stayed at international 5-star hotels, according to the survey findings. Furthermore, the findings of the survey show that the vast majority of respondents (404 in total) traveled with their families. Table 1 displays the results of the respondents' survey.
The validity test is a stage that is used to determine how well the questionnaire will present the whole hypotheses that were collected and how well the use of correct concepts and consistent indicators can be relied on (Sekaran & Bougie, 2016, p. 220).
According to the test of convergent validity via the loading factor value as shown in Table 2 and Figure A1 outer model, 10 indicators from 12 indicators measuring latent variables are valid. Table  A1 displays the construct reliability and validity for three dimensions: market learning, market sensing, and market targeting and positioning. Table A2 outer loadings displays the variable market sensing with 3 indicators shows a valid loading factor value with a value greater than 0.7. The variable market learning with 3 indicators shows a value greater than the rule of thumb, namely 0.7 so it is said to be valid. Likewise, the variable market targeting and positioning with 4 indicators have a value greater than 0.7 and are declared valid.
Average Variance Extracted values are used to determine the pre-convergent test's validity. AVE has a value greater than 0.50 for the rules of thumb (Ghozali, 2021, p. 71). The three variables consisting of market sensing, market learning, and market targeting and positioning are also declared valid in the convergent validity test through AVE.
Cronbach's alpha and composite reliability have a purpose to show that the indicators' accuracy and consistency of these findings. Moreover, accuracy can be used to calculate the construct. With the rule of thumb, reliability must be greater than 0.7 (Ghozali, 2021, p. 71). According to the data in Table 2, all three variables are considered reliable because the value provided by data processing exceeds the rule of thumb > 0.7. Cronbach's alpha value of more than 0.90 indicates brilliant internal consistency, a value more than 0.80 indicates strong internal consistency, a value more than 0.70 specifies acceptable internal consistency, a value more than 0.60 indicates questionable internal consistency, a value more than 0.50 indicates poor internal consistency, and a value more than 0.50 indicates poor internal consistency, according to George and Mallery (2003). As a result, it is specified that all variables can be used in hypothesis testing.   Table 3 displays the outer VIF value that defined the indicator variables with a VIF value with a threshold of < 5 and interpreted no bias and multicollinearity issues. The greater the degree of collinearity, the higher the VIF, and VIF values above five are a definite indicator of high collinearity among the indicators, and thus an issue (Hair et al., 2019, p. 777). Besides, common method variance (CMV) was described by Buckley et al. (1990) as the amount of spurious covariance shared among variables as a result of a common data collection method. Harman's single-factor test is the most well-known technique for analyzing CMV in a single-method study design (Podsakoff et al., 2012). If the percent of variation exceeds 50%, there is a bias. Table A3 displays the results of the Common Method Variance calculation using factor analysis in the SPSS application, indicating that the variable is acceptable since the percent variance is less than 50%. The goodness of fit (GOF) measure puts a focus on the expected value's similarity to the observed value in the variable. Additionally, goodness of fit can be used to create an index of a model's suitability when the model is verified to adequately explain empirical evidence. According to Ghozali and Latan (2015), the GOF value is determined in equation (1) Therefore, the calculation is: With the GOF value obtained of 0.763, empirical data are suitable and fit the model as depicted in equation (2) above. The GOF ranging from 0 to 1, with 0.10, 0.25, and 0.36 denoting low, medium, and high global validity of the path model, respectively (Hussain et al., 2018).
The fitting and cross-validation functions are presented in predictive relevance, which is a technique for predicting observed variables and estimating from construct parameters. When Q 2 > 0, predictive relevance is calculated in the model; otherwise, when Q 2 < 0, the model is declared to have no predictive relevance. In this analysis, predictive relevance is determined using a blindfolding technique in which the value of the dependent variable yields a value greater than the rule of thumb Q 2 > 0, indicating predictive relevance. If the predictive relevance value is greater than zero, the model is predictive; a value of 0.02 shows the relevance of a weak prediction; 0.15 shows a moderate relevance; and 0.35 shows a strong relevance (Ghozali & Latan, 2015). The value of Q 2 described that the construct endogenous, which is DMC is 0.476, and through this value obtained, the model has proved to be in the highest category. Table  4 displays the findings of the data analysis for predictive relevance. According to the type of variable studied, hypothesis testing uses statistical calculations of different levels of significance. The t-test was used to demonstrate substantial mean differences in variables with interval or ratio scales (Sekaran & Bougie, 2016, p. 285). Hypothesis testing with the significance calculated in this study employs a confidence level of 95%, implying that the significance level is 0.05 on a one-tailed basis. The result, hypothesis testing uses a T-statistic of more than 1.96 as a rule of thumb to determine the significance of a hypothesis, as well as a P-value below 0.05. Furthermore, Table 5 indicates the effects of hypothesis testing in this analysis, indicating that the hypotheses are supported. According to H1, the result of testing this hypothesis indicates that it is supported as shown in Table 5, the T-statistic result is larger than 1.96, which equals 30.574, and the P-value is less than 0.05, which equals 0.000. The outcome of testing H2 shows that it is supported as shown in Table  5, the T-statistic result is larger than 1.96, which equals 33.314, and the P-value is less than 0.05, which equals 0.000. The result of the test suggests that H3 is supported as shown in Table 5, the T-statistic result is larger than 1.96, which equals 40.968, and the P-value is less than 0.05, which equals 0.000.

CONCLUSION
The data analysis conducted with the Smart-PLS 3.2.9 software generates three supported hypotheses in assessing the DMC of international hotels in Indonesia from the perspective of customers. According to the first hypothesis, there is a connection between market sensing and DMC. This means that market sensing carried out by 3-5-star international hotels in Indonesia has an effect on DMC carried out by hotels from the perspective of the consumer. The second hypothesis proposes a relationship between market learning and DMC. This means that market learning implemented by 3-5-star international hotels in Indonesia has an effect on DMC implemented by hotels from the consumer's perspective. The third hypothesis implies a relation between market targeting and positioning toward DMC. This means that the market targeting and positioning implemented by 3-5-star international hotels in Indonesia has an effect on the DMC from the consumer perspective. As a result of the hotel industry's tightening competition, especially for 3-5-star international hotels, numerous new hotels have sprung up. The findings of this study are projected to help the hotel industry take the necessary steps to remain competitive by using marketing strategies such as dynamic marketing capability whose dimensions consist of market sensing, market learning, and market targeting and positioning.
In terms of DMC and using the SENS3 indicator as to the indicator with the highest average descriptive statistics from the market sensing dimension, hotels must provide accurate and consistent details about the hotel to allow customers to make an informed choice about where to stay, for example, video tour of the room. Additionally, since the LEARN1 indicator has the highest average descriptive statistics in the learning dimension, hotels must gather feedback from customers to expand their marketing strategy, for example, online surveys for guest satisfaction. Following that, with the TAPO4 indicator having the highest average descriptive statistics around the targeting and positioning dimensions, hotels must adapt culturally to the local culture in the region where the hotel is located to attract the market appropriately.