The approach of tacit marketing knowledge

Knowledge can be divided into two types according to its capacity to be structured and codified, so that explicit knowledge is considered scientific and highly structured, whereas tacit knowledge is subjective, based on experience, intuition and judgment (^{Nonaka and Takeuchi, 1995}; ^{Gummesson, 2017a}). The interaction of both types of knowledge shapes organizational knowledge (^{Easterby-Smith and Prieto, 2008}) and pragmatic wisdom (^{Gummesson, 2017b}).

Various studies have found that a great deal of marketing knowledge is tacit, incarnated in people and, therefore, difficult to transfer (^{Bennett, 1999}; ^{Simonin, 1999}; ^{Archibugi and Pietrobelli, 2003}). Tacit marketing knowledge is significant because it acts as a social process by which employees with knowledge in the discipline acquire and transfer their learning to others (^{Ambrosini and Bowman, 2001}). This knowledge is usually rooted contextually, organizationally and socially in the organizations (^{Inkpen, 2008}), but adding to this the characteristics of the hotel sector (^{Hjalager, 2002}), it becomes relevant to promote transfer mechanisms in order to take advantage of it in various market aspects, for example, innovation and development of new products (^{Huang, Chang, and Henderson, 2008}), trust and collaboration (^{Szulanski, 2000}), relationship marketing (^{Gummesson, 2017a}) and strengthening of a competitive advantage.

The acquisition of knowledge may have multiple backgrounds (^{Liao and Barnes, 2015}), nevertheless, this research studies the ability to absorb knowledge regarding the discipline, an open culture of collaboration, trust between co-workers and communication media richness as variables of relevant impact on the acquisition of marketing knowledge and the effect of this on the transfer of it.

Marketing knowledge absorption capacity and acquisition of tacit marketing knowledge

Absorption capacity refers to the ability to absorb knowledge, with the expectation of acquiring new knowledge so that it can be shared after (^{Grant, 1996}). This capacity is likened to the set of processes and organizational routines that allow identifying, assimilating, transforming and exploiting new environmental knowledge (^{Cohen and Levinthal, 1989}; ^{Zahra and George, 2002}); therefore, it requires consent for the exchange of ideas and to allow companies to be more involved in the acquisition of that knowledge (^{Yli-Renko, Autio and Sapienza, 2001}).

^{Rodríguez Antón, Oliva, and Laguna (2003}) mention that the current hotel situation is one of stagnant demand, continuous change, and need for constant learning; however, it may be difficult to assess the effect of absorption capacity (^{Berchicci, 2013}), specifically in the evaluation of its effects on tacit knowledge, given the prevailing scarcity of empirical studies (^{Jafari and Mozhden, 2013}). Absorption capacity combines existing knowledge with new knowledge (^{Lane, Koka and Pathak, 2006}) in such a way that it impacts the development of the knowledge and potential of a firm to innovate and improve its competitive performance (^{Zahra and George, 2002}).

The acquisition of knowledge cannot be understood without absorption capacity, given that new knowledge can be potentially assimilated, transformed and exploited (^{Lane et al., 2006}; ^{Johnson, 2017}). There are high employee turnover rates in the field of hotel chains, which affects the retention of existing information and knowledge in the organization. Staff contact with clients is a key source of information on the markets and the business itself. Therefore, there is cause for knowledge acquisition and management to be professionally implemented in the hotel sector (^{Gjelsvik, 2002}; ^{Yang and Wan, 2004}). For this reason, it is possible to assume that:

H1: In small and medium-sized hotels in Playa del Carmen, the marketing knowledge absorption capacity has a positive effect on the acquisition of tacit marketing knowledge.

Open culture of co-worker collaboration and its effect on the acquisition of tacit marketing knowledge.

Within the institutions there is a need for a culture of collaboration that will motivate the members of the organization to look for new ways of implementing processes, where the working hours and interaction among its members will contribute to the acquisition and sharing of knowledge (^{Kucharska and Kowalczyk, 2016}).

The culture of collaboration is one of the most important conditions that can lead to the success of a business. Most employees have a curiosity and intellectual disposition that can facilitate the acquisition of knowledge and its dissemination (^{Davenport and Prusak, 2001}; ^{Kucharska and Kowalczyk, 2016}). In the generation of intra-organizational knowledge, intensive communication is required, as well as a culture that accepts new ideas and is prepared to support the exploration of its processes and activities (^{Ruggles, 1998}).

Some research (^{Zapata, 2004}) has found that organizational culture is the main facilitator of knowledge, and that more open and change-oriented cultures are better able to acquire and transfer knowledge internally. Research in different fields has found that the culture of the organization has a positive effect on knowledge acquisition (^{Liao, Chang, Hu and Yueh, 2012}). It has been found that a firm with an open and flexible culture is more likely to engage with learning actions, changing routines, processes or the value chain in favor of an appropriate strategic direction. Therefore, based on the above, it is proposed that:

H2. In small and medium-sized hotels in Playa del Carmen, an open culture of collaboration among co-workers positively influences the acquisition of tacit knowledge.

Mutual trust between co-workers and the acquisition of tacit marketing knowledge.

Mutual trust refers to the belief or expectation that, in order to carry out certain activities, support can be sought from another partner with the certainty that it will correspond to the necessary responsibility (^{Young and Wilkinson, 1989}), and that the other person will not take advantage of the vulnerability actions of the requester (^{Swan, Trawick, Rink and Roberts, 1988}). Trust is an important factor in building social relationships and exists when one party believes in the trustworthiness and integrity of the other (^{Morgan and Hunt, 1994}; ^{Hunt, Arnett and Madhavaram, 2006}) and rests on the expectations and predictions of collaboration between the parties involved (^{Maurer, 2010}).

Trust acts in relationships and facilitates both the access and dissemination of knowledge (^{Arrow, 1974}; ^{Nahapiet and Ghoshal, 1998}). In any kind of social exchange, it is important to include inter-institutional and intra-organizational relationships (^{Massey and Dawes, 2007}). It also plays a central role in knowledge acquisition and the sharing processes (^{Dhanaraj, Lyles, Steensma and Tihanyi, 2004}; ^{Maurer, 2010}).

Previous research has found that trust generates benefits for the organization through interaction and collaboration (^{Dirks and Ferrin, 2001}); furthermore, it has been shown that trust increases knowledge acquisition among members of a firm or among allies, leading to new business opportunities (^{Tsai and Ghohal, 1998}; ^{Maurer, 2010}). In the area of marketing, ^{Arnett and Wittmann (2014)} tested a model of tacit knowledge sharing between sales and marketing and concluded that when confidence in the co-worker is greater, tacit knowledge sharing tends to increase.

By its very nature, tacit knowledge requires frequent personal interactions with other members of the organization for the acquisition and transfer of tacit knowledge. Thus, relationships characterized by high levels of trust are more conducive to encourage acquisition (^{Maurer, 2010}) and the transfer of tacit knowledge (^{Lin, 2007}). From the above, it is possible to assume that:

H3. Confidence in co-workers has a positive effect on the acquisition of tacit marketing knowledge.

Communication media richness and the acquisition of tacit marketing knowledge.

Communication processes can become drivers or barriers to the development of the willingness to share knowledge (^{Wiewiora, Trigunarsyah, Murphy and Coffey, 2013}), but it also functions as a relevant mechanism for the acquisition of knowledge from internal and external sources (^{Boiral, 2002}), as well as the assimilation of practices in small and medium- sized enterprises (^{Johnson, 2017}).

Several authors (^{Mirc, 2013}; ^{Wiewiora et al., 2013}) agree that the integration of teams and regular meetings are essential for the knowledge acquisition and transfer process, and that the application of rotations, cross-cutting job integration teams, and regular communication are considered to facilitate the flow of knowledge between the members of an organization (^{Mirc, 2013}).

Although there are several communication mechanisms, informal intra-organizational communication also offers opportunities for the acquisition and transfer of tacit knowledge in the organization (^{Dyer and Hatch, 2006}); interaction and communities of practice among the members of the firm can become drivers of this process because they provide the opportunity to share knowledge, even at the initiative of the collaborators themselves (^{Zapata, 2004}). It is therefore proposed that:

H4. In small and medium-sized hotels in Playa del Carmen, communication media richness positively influences the acquisition of tacit marketing knowledge among employees.

Acquisition and transfer of tacit marketing knowledge

The process of acquiring and transferring knowledge undoubtedly involves several key people, one of whom is the holder of the knowledge and another is the recipient of the same. The result of knowledge acquisition depends on the efforts of both parties. Therefore, to motivate acquisition, the knowledge content received by a recipient and the source must perceive the knowledge as valuable (^{Desouza, Awazu and Wan, 2006}).

The acquisition of tacit knowledge requires continuous and intense contact between employees (^{Kale, Singh and Perlmutter, 2000}), and when knowledge is transferred from one person to another, it is interpreted by the recipient based on what they know and their experience. This may have a different meaning and value in the new environment. In marketing, tacit knowledge has been found to be a valuable resource which, when shared, increases opportunities for the development of new strategies in the different areas of marketing. The blockade and organizational barriers to their development can give only moderate results (^{Amiryany, Ard/ Pieter and Cloodt, 2012}; ^{Harvey, 2012}). Therefore, it is proposed that:

H5. The acquisition of tacit marketing knowledge has a positive impact on the transfer of tacit marketing knowledge.

Figure 1 shows the conceptual model of tacit knowledge transfer that this research proposes. This model represents the relational links between constructions that represent Marketing Knowledge Absorption Capacity (CACM for its acronym in Spanish), Open Collaborative Culture (CAC for its acronym in Spanish), Trust in Co-workers (CCT for its acronym in Spanish) and Communication Media Richness (RMC for its acronym in Spanish), the Acquisition of Tacit Marketing Knowledge (ACTM for its acronym in Spanish), as well as the Transfer of Tacit Marketing Knowledge (TCTM for its acronym in Spanish). It also indicates the hypothetical relationship between the variables.

Figure 1 Conceptual model of the precedents of the acquisition of tacit marketing knowledge and its impact on the transfer of knowledge. 

Measurement

The instrument used comprises 29 items. The measurement scales are interval scales, these being 5 and 7-point Likert scales (^{Hair et al., 2010}). Cronbach’s alpha, for the entirety of the measured variables, had a value of 0.819.

It was found that the standard deviation in all items is greater than 1 in the dimensional analysis; on the other hand, Cronbach’s alpha evaluated internal consistency, where the first construct “marketing-knowledge absorption capacity” yielded an alpha of 0.932. ^{Garson (2010)} comments that in social sciences the alpha cut should be 0.80 or higher for a set of elements to be considered for a scale, but it is possible to use 0.70 as acceptable for confirmatory research (Cronbach, 1951; ^{Thiétart, 2001}). In the second construct “open culture to collaboration”, the alpha was of 0.865; in the third construct “trust in co-workers”, the alpha was of 0.919; in the fourth construct “cultural distance”, the alpha was of 0.779, with this consistency being the lowest of all and as such the decision was made to eliminate it from the evaluation of the model. The lowest value was found in the fifth construct “communication media richness” with 0.816; in the sixth construct “acquisition of tacit marketing knowledge”, the alpha was of 0.943; and in the seventh and final construct “transfer of tacit marketing knowledge”, the alpha was of 0.874.

Measurement model

For the statistical analysis of the data, the procedure followed was the Structural Equation Modeling (SEM) technique to examine complex models that present a large number of constructs, indicators and relationships (^{Barclay et al., 1995}). Because Partial Least Squares (PLS) allows working with small samples and has less strict assumptions regarding data distribution (^{Chin and Newsted, 1999}). In the Partial Least Squares Structural Equation Modeling (PLS-SEM), the model is described by two components (^{Tenenhaus et al., 2005}): 1) the measurement model, which relates the manifest variables to the latent variable; and 2) the structural model, which shows the relationship between the latent variables. Structural models are validated in two stages (^{Henseler, Ringle and Sinkovicks, 2009}): 1) review of the measurement model through various procedures; and 2) validation of the structural model.

Based on the rule proposed by ^{Hulland (1999)} on keeping items with loads of 0.7 or more, those indicators from reflective measurement models that did not reach the acceptable level of reliability (Annex 2) were eliminated. Convergence validity is evaluated with the Average Variance Extracted (AVE) value, which must be greater than 0.5 according to the Fornell- Larcker criterion (^{Seidel and Back, 2009}). In the final model (model 2), the reflective constructs present a minimum value above the limit (0.59). Convergence validity is also demonstrated when the items carry a high load on their associated factors (load greater than 0.5) and these must be greater in the assigned construct than in any other. The cross-loads of the reflective indicators show a higher load in absolute value in the construct to which they have been assigned (Annex 3), this with respect to any other reflective construct (^{Seidel and Back, 2009}).

Dillon-Goldsteins Rho coefficient (compound reliability index) is used to evaluate internal consistency (^{Fornell and Larcker, 1981}). In the model, the composite reliability index for reflective constructs was greater than 0.88 (Annex 4), exceeding the minimum acceptable value of 0.70 (^{Seidel and Back, 2009}; ^{Hair, Anderson, Tatham and Black, 1998}). With regard to discriminant validity, ^{Fornell and Larcker (1981)} suggest that an AVE score of 0.5 indicates an acceptable level of discriminant validity. The average variance extracted for reflective constructs varies between .59 and .87 (Annex 2). Said validity for reflective measurement models is also demonstrated when the root of the AVE average of each construct is greater than the correlation with any other latent variable (^{Seidel and Back, 2009}; ^{Delić and Lenz, 2008}). In Annex 4, the matrix comparing the AVE root in the diagonal of the tables with the top triangle of the matrix containing correlations between constructs is presented. The AVE root is larger than the correlation for each respective column and row of reflective constructs, suggesting discriminant validity (^{Seidel and Back, 2009}; ^{Duarte and Raposo, 2010}) and acceptable convergent (^{Chin, 1998}; ^{Duarte and Raposo, 2010}). ^{Seidel and Back (2009)} agree that the discriminant validity is met by comparing the cross-loads of the indicators assigned to the reflective construct against the rest of them. In the model, no problems of discriminant validity are revealed (Annex 5), given that all indicators show higher loads in their respective constructs with respect to other reflective constructs (^{Duarte and Raposo, 2010}).

Structural model

The model explains 26.80% of the TCTM variance (R²=26.80), directly through ACTM and indirectly through CCT and RMC. The bootstrapping technique with 200 sub-samples was used to estimate the significance of the path coefficients in the model and to statistically compare the estimators. The results of the proposed model (Figure 2) support 60% of our hypotheses (Table 1), with a 95% level of confidence.

The results in Figure 2 indicated a direct, positive and significant effect of ACTM on CTM (β₅ = 0.5177). Furthermore, the indirect effect of CCT on TCTM (β_3* β₅) resulting from applying the bootstrap technique (^{Kenny, 2015}) was significant at 0.1848 (95% CI: .1089-.2617). Similarly, RMC indirectly supported TCTM, with an effect of .1932 (95% CI: .0975-.2829).

The proposed hypothesis regarding RMC and ACTM was significant and in the expected direction. So that RMC directly and positively affects ACTM (H4: β₃ = 0.3769). In relation to CCT, the model indicates the strength of the direct and significant relationship with ACTM (β₃ -0.3622), supporting hypothesis H3.

Figure 2 Final TCTM model with significant path coefficients.