An Empirical Approach to Marketing Innovation in Small and Medium Retailers: An Application to the Spanish Sector

Although numerous studies explore innovation, most of them focus on technological innovations. It is currently difficult to focus exclusively on products and processes when discussing innovation because new ideas can transform any part of an organization’s value chain. Indeed, marketing innovation is essential for small and medium firms and its study is extremely important given the current lack of empirical evidence. Furthermore, the retail sector in Spain plays a significant role in the country’s economy due to its contribution to gross domestic product and employment and to its role in the growth of cities and regions. In this context, the present study analyzes the determinants of marketing innovation practices of small and medium retailers and the relationships among the four types of innovation (product, process, organizational and marketing) to improve our understanding of marketing innovations in the retail sector. This study uses secondary data on innovation activity for Spanish companies in the years 2010-2012. Logistic regression models are estimated, and the results show that the introduction of new marketing methods has a positive impact on the propensity to engage in all other types of innovation (product, process and organizational). Furthermore, all of these methods positively impact the propensity to introduce marketing innovations. This study concludes with insights that can help managers to direct innovation processes in their companies and assist public administrators in the promotion of marketing innovations among small and medium retailers.


Introduction
Spanish public policy has placed particular importance on supporting small and medium enterprises (SMEs) because these firms are the backbone of industry. Moreover, marketing innovation is essential for small firms to transform products into profit (Soltani, Azadi, Hosseini, Witlox, & Passel, 2015). The tendency to engage in marketing innovation depends on the activities of an enterprise, and differences between the manufacturing and service sectors are significant . Companies in the service sector are more likely than those in the manufacturing sector to innovate in the areas of promotion, placement and price. Within the service sector, studies of the Spanish retail sector are important due to the significant contributions retail businesses make to Spanish GDP and employment. Furthermore, the retail sector is key to the growth of cities and regions (Medrano, Olarte-Pascual, Pelegrín-Borondo, & Sierra-Murillo, 2016). Taking into account the context in which small and medium retailers currently operate, as well as all of the public business programs launched to encourage firm innovation, an improved understanding of this situation is needed to create effective public policies for each type of business.
Thus, the objective of this paper is to understand marketing innovation and its determinants among small and medium retailers based on data from the Technological Innovation Panel (PITEC). More precisely, this study investigates (1) the factors influencing the four types of innovation (product, process, organizational and marketing) and (2) the relationships between marketing innovation and the other types of innovation. This paper, which focuses on marketing innovation, is organized as follows: first, the concept of marketing innovation and its repercussions in the literature are discussed. Second, the sample and methodology employed are described. Finally, the results and conclusions of the study are presented.

Theoretical framework and hypotheses
The first studies of innovation focused on product innovation in the manufacturing context. Although the literature on innovation is abundant, most of it refers to product and process innovations, which are called technological innovations. Certain studies have attempted to relate technological and organizational innovation (e.g., Brown & Duguid, 1991;Brown, 2002;Djellal & Gallouj, 1999;Kishore & McLean, 2002;Tether et al., 2002). However, non-technological innovations -i.e., marketing and organizational innovations -have generally been neglected in the literature.
Marketing innovation is defined as the "implementation of a new marketing method involving significant changes in product design and/or packaging, product placement, product promoting or pricing" (OECD/

Relationships between marketing innovation and the other innovation types
According to Walker (2004), innovation activities are thought to influence each other and thus need to be implemented in conjunction with each other. The disputed relationship between technological innovation and marketing innovation centers on whether they are complements or substitutes; according to Kijek (2013), it is quite intuitive that the introduction of a new process or product calls for changes in marketing strategies. For instance, a new product line often requires changes in packaging and sales channels. Lewis and Wackowski (2006) provide an example from the tobacco industry, in which the introduction of flavored cigarettes was intensively supported by marketing innovations. However, other authors argue that marketing innovations do more than merely support technological innovations (Bhaskaran, 2006 andRammer et al., 2009). Following this reasoning, Schubert (2010) maintains that marketing innovations are, on average, complements -rather than substitutes -for technological innovations but notes that this relationship is sensitive to external and internal factors, such firm size and technological opportunities. However, this author argues that marketing innovations make product and process innovations more successful. Furthermore, Walker (2008) concludes that organizational, marketing and service (or product) innovations are interrelated in public organizations. In a study of manufacturing firms in Turkey, positive relationships were shown between organizational innovation and process innovation, organizational innovation and marketing innovation, process innovation and product innovation, and marketing innovation and product innovation (Gunday, Ulusoy, Kilic, & Alpkan, 2011). Additionally, Soltani et al. (2015) show that although there is no direct cause-and-effect relationship between organizational and production innovation, both of these innovation types are influenced by marketing innovation.
Considering the existing literature on the relationships between marketing innovation and the other types of innovation, the following hypotheses are formulated:

H1.
Marketing innovation in small and medium retailers is positively associated with product innovation.

H2.
Marketing innovation in small and medium retailers is positively associated with process innovation.

H3. Marketing innovation in small and medium retailers
is positively associated with organizational innovation.

Features of firms that adopt marketing innovations
In In Spain, enterprises that conduct product, process or organizational innovations are more likely to innovate in marketing ). The findings of Soltani et al. (2015) for small food industries show that when product and organizational innovations increase, marketing innovation also increases. Hence, the following hypotheses are formulated for small and medium retailers: H4. Small and medium retailers that conduct product innovations are more likely to innovate in marketing.

H5. Small and medium retailers that conduct process
innovations are more likely to innovate in marketing.

H6.
Small and medium retailers that conduct organizational innovations are more likely to innovate in marketing.

Survey
Data from PITEC are used to conduct this study. This database follows the methodological guidance pro-

Statistical methods
To determine whether there is a multicollinearity problem, this study considers the tolerance and variance inflation factor (VIF). A tolerance smaller than 0.20 and/or a VIF of 5 or higher indicate a multicollinearity problem (O'Brien, 2007).
Considering the two aims of the study, the analysis is divided into two parts. In the first part, the probability that firms engage in each kind of innovation (product, process, organization and marketing) is examined to verify the relationship between marketing innovation and the other types of innovation. Logistic regression is selected because it adapts to all types of independent variables (metric and non-metric) and does not require multivariate normality (Hair, Black, Babin, Anderson, & Tatham, 2006

209
An empirical approach to marketing innovation in small and medium retailers: an application to the Spanish sector

Results and discussion
The database provides a sample of 463 companies, with just over one-quarter (27%) reporting that they implemented marketing innovation activities (table 2).
However, the data show the importance of the retail sector because this sector does not exhibit the same average behavior as other sectors. In other sectors, marketing innovation is the least frequently implemented innovation of the four types of innovation. In contrast, our analysis of the retail sector shows that marketing

Retail
Other Sectors  When all independent variables are included in the analysis, table 3 shows that none has a tolerance less than 0.70 and there is no VIF higher than 2. In conclusion, multicollinearity is not an issue in this study.

Table 4. Logistic regression for types of innovation in small and medium retailers
Note: *Statistically significant at the 0.1 level, **statistically significant at the 0.05 level, ***statistically significant at the 0.01 level.
An empirical approach to marketing innovation in small and medium retailers: an application to the Spanish sector

Conclusions
In the modern world, it is undeniable that globalization, stronger competition, increasingly stringent and segmented demand, rapid technological evolution and increasingly shorter life cycles have led to substantial growth in the relevance of innovation as a key factor in achieving leverage in the market. These obvious changes in both society and consumer behavior render previously appropriate models unsuitable in current conditions, and the marketing discipline -like practical and scientific expertise -is not exempt from these changes. In the current scenario, small and medium retailers are losing customers every day; these stores eventually close, and over time, cities slowly lose their cultural and economic vibrancy .
In this context, the present study analyzes the determinants of the marketing innovation practices of small and medium retailers and the relationships among the four types of innovation (product, process, organizational and marketing) in order to better understand marketing innovation in the retail sector.
The first result shows that marketing innovation is the second-most frequently implemented innovation by small and medium retailers, whereas in other sectors, marketing innovation is the least frequently implemented innovation of the four types of innovation. This result indicates the importance of marketing innovation in the retail sector and is consistent with Bhaskaran (2006) and Rammer et al. (2009), who note that small and medium enterprises, especially those An empirical approach to marketing innovation in small and medium retailers: an application to the Spanish sector in low-and mid-tech industries, may be more willing to engage in marketing innovation, which can be less costly than technological innovation.
The second result shows that the introduction of new marketing methods among small and medium retailers has a positive impact on the propensity to conduct all other types of innovation (product, process and organizational). However, introducing other types of innovation does not have the same impact. All other innovation types positively impact the propensity to introduce marketing innovations, albeit not of all types. We also conclude that product, process and organizational innovations present different determinants and different inter-relations among the different types of innovations.
The last result shows that the introduction of new organizational methods has a positive impact on the propensity of small and medium retailers to conduct all types of marketing innovation (design, pricing, promotion and placement). However, the introduction of new products or processes only has a positive impact on the propensity to conduct design innovations.
Thus, we conclude that the determinants of marketing innovation differ among the four types of marketing innovation in the retail sector.
In conclusion, it is important to highlight that the