Purpose The seasonality of business is an important determinant of the hotel industry, particularly in holiday destinations. Seasonality opens a series of questions which require particular attention and application of appropriate strategies both on the hotel enterprise level, and on the level of a tourist destination. Bearing in mind the dominance of small hotels on the tourist market, large market flexibility and creativity of small entrepreneurship, and the ability of fast adjustment to the changing market conditions (Vukičević, 2000.), small entrepreneurship in hospitality is given a function of a trigger of transformation of a tourist destination focused exclusively on business in one (winter or summer) season in the destination with a whole year business. Doctoral dissertation is focused towards actualizing the significance of small entrepreneurship in reducing seasonality in Croatian tourism. This research has managed to extend previous knowledge about the nature and the intensity of connections of features of hotels and seasonality of business in coastal destinations (Ćorluka, 2015), on continental Croatia and Republic of Northern Macedonia, as a destination in which tourist demand does not show pronounced seasonality. Present lack of research which prove the connection between small entrepreneurship and reduction of seasonality on tourism, and still even greater lack of specific content which relates on the characteristics of small hotels on reducing seasonality in hospitality on European and national level, have pointed on the need of focusing the doctoral dissertation on the relation of characteristics of small entrepreneurship and seasonality of business. The assumption is that there is a relation between the characteristics of small entrepreneurship and seasonality of business. Purpose of research was to deepen the existing scientific and professional findings about the seasonality of business in Croatian hospitality and the influence of initiatives of small entrepreneurs in hospitality on reduction of seasonality and reach new findings about the connection of characteristics of hotels and seasonal business. Research goals were: 1) to explore the theoretical determinants of hospitality, and in particular small entrepreneurship in hospitality and its significance for the tourist destination; 2) to determine the aspects of seasonality in hotel business through specifics of hotel business and connection of hotel business along with a determined time interval depending on the natural and institutional causes of seasonality; 3) to explore the seasonality of entrepreneurs in European hospitality, and in particular the place and role of small entrepreneurship in development of European tourist destinations; 4) to analyse the connection of characteristics of Croatian hotels and seasonality of business of Croatian hotel industry; 5) to explore the necessity of stimulation of entrepreneurial initiatives for increasing competitiveness of Croatian tourist destinations and inclusion of continental part of Croatia into the tourist flows; 6) to point out on the application of necessary measures and implementation of appropriate model of managing small entrepreneurship with a purpose of creating and applying new and/or innovated strategies of reduction of seasonality, in order to completely satisfy the existing and attract new market segments. Following the purpose and goals of the research, the following scientific hypothesis was set: scientific and professional knowledge about the hotel industry, small business in the hotel industry and seasonality of business, and knowledge about the state of small business in Croatian and European hospitality, as well as the impact of small hotels are the basic determinants of entrepreneurial initiatives that give direction to the development of tourism in the destination and result in a reduction in the seasonality of business in the Croatian hotel industry. Proof of the basic scientific hypothesis was made possible through the development of five auxiliary hypotheses, and a built-in conceptual model that tested the positive relationship between the characteristics of small hotels and seasonal business. Methodology Appropriate scientific research methods were used in processing data relevant to the topic of the doctoral dissertation and presenting the results of research for the purpose of proving the set basic and auxiliary hypotheses: description method, compilation method, comparison method, induction and deduction method, analysis and synthesis method, and quantitative data were processed using mathematical and statistical methods. For the purposes of empirical research, descriptive statistics methods were used in collecting, grouping, and analysing the characteristics of hotel enterprises and hotels covered by the research and business performance expressed through the rate of occupancy of hotel capacities. The obtained data are tabulated and graphically presented, and the aim was to calculate various numerical indicators that express the characteristics of the observed phenomenon. Correlation and regression analysis were used in testing the set research model. The study included twelve characteristics of the hotel (independent variable): hotel location, hotel size, hotel capacity, hotel categorization, hotel business years, hotel by purpose, hotel by length of stay, hotel by location - destination, hotel regarding categorization, hotel with regard to the way of doing business, hotel with regard to the form of business, hotel with regard to the channels of placement of hotel accommodation. Selected characteristics have been used in previous research, which has enabled their assessment as determinants of business seasonality. It is assumed that these hotel characteristics are correlated with the rate of occupancy of accommodation facilities in seasonal intervals (dependent variable). The study adhered to the view of three categories of seasonality with respect to tourism trends: high, middle, and low seasons (Kastenholz and Almeida, 2008). Hierarchical cluster analysis was used to group the occupancy rate into individual seasonal intervals, as one of the methods of multivariate analysis, namely the Ward method, and the results are presented in a dendogram. Euclidean distance was used to determine similarity. Range method, standard deviation and variance were used as a measure of dispersion, with the determination of the arithmetic mean for both individual months and seasonal intervals using descriptive statistics. The research was based on panel data, respectively, the hotel data in three seasonal intervals. Therefore, the observation units are hotels (for the Republic of Croatia N = 237, and for Northern Macedonia N = 51), and the seasonal intervals are units of time (T = 3). This led to the conclusion of 711 observations for each hotel covered by the survey in the Republic of Croatia, and 153 observations for each hotel covered by the survey in Northern Macedonia. The correlation values expressed by the Pearson coefficient showed the extent to which changes in hotel feature values were related to changes in hotel occupancy values in low, medium and high season seasonal intervals, and its sign determined the direction of correlation. The t-test was used to test the linear correlation coefficient. However, since the SPSS program was used for the calculation, a two-way test of the significance of the linear correlation coefficient (Two-tailed) was used, which is automatically included when calculating the correlation of two variables. The research was further deepened by regression analysis, respectively the model of multiple linear regression, in order to determine the analytical expression of the relationship between the variables. The regression model was used as a basis for forecasting the improvement of hotel occupancy in certain seasonal intervals by changes on the side of hotel characteristics. The appropriateness of using a regression model is based on the Variance Inflator Factor (VIF) diagnostic test, which determines which variables will be included in the model. With the help of the VIF test from the regression model, independent variables for which the existence of multicollinearity problems was determined were omitted.The research was conducted on a random sample of 237 respondents in Croatian hotels and on a sample of 51 respondents in Northern Macedonia which served comparative purposes and developed a discussion on the connection and impact of hotel characteristics on seasonality in two different destinations regarding hotel seasonality. An online survey questionnaire was used, consisting of two groups of questions. One group of questions related to the characteristics of the hotel business and the other to the characteristics of the hotel. Data were processed using IBM SPSS. Findings The results of the research confirmed the high level of seasonality of the Croatian hotel industry business and pointed out the problems and challenges that small hotels face in the fight against unprofitability of business. The analysis of small business in the hotel industry in Europe and the world highlighted the greater negative implications of the effects of seasonality on small hotels compared to large ones. This especially refers to the occupancy of hotel capacities and the realization of profit per unit of accommodation. The research further showed that seasonality in hotel occupancy mainly depends on the market segment in which the hotel offers its services. Changes in the structure of hotel guest profiles are directly reflected in the monthly patterns that are monitored in the occupancy profile. Therefore, hotels are directing their efforts towards attracting that segment of guests that will bring them greater occupancy. Research has shown that small hotels, due to their flexibility, can easily adapt to new trends and provide a guest service that will make him a satisfied and loyal guest. Tourist destinations with a higher number of quality small entrepreneurs achieve greater competitiveness, utilization of accommodation capacities and reduction of seasonality. Small hotels in Croatia participate with a very high share (73.5% in 2019), but in the total hotel capacity they make up only about 22%. In order to better position themselves in the tourist market, small hotels strive for a personalized approach and the most authentic product to attract guests, but the hotel offer still lacks strong diversification and specialization of hotel products. The Lorenzo curve clearly showed the high level of seasonality of the Croatian hotel industry, and in the period from 2010 to 2019 there are no significant changes, although a noticeable decrease in seasonality is noticeable by a slight shift in the concentration of tourist nights in the upper Lorenz curve. The Gini coefficient showed a higher seasonality of the total realized tourist overnights in the Republic of Croatia, compared to those realized in hotels and similar facilities. Empirical research has proven the hypothesis: there is a positive link between the characteristics of small hotels and seasonal business. The research using correlation and regression analysis concludes that the characteristics of hotels defined in the model are related, but neither correlation nor regression analysis has determined the significance of all characteristics for seasonal business of small hotels, both in the sample in the Republic of Croatia and in the North. Macedonia. The significance of the Pearson coefficient in the sample for the Republic of Croatia was noticed in a number of connections between the variables of hotel characteristics and the 'low season' in relation to the 'middle season' and the 'high season'. All relationships between independent variables and 'low season' were statistically significant at either 1% or 5%, and when examining the association between independent variables and 'mid-season' and independent variables and 'high season', a statistically significant association was observed only for the hotel location variable. On the sample for the Republic of Northern Macedonia a significant correlation was determined between the variables' type of hotel in terms of number of accommodation units' (HBHS) and 'low season' and 'type of hotel in terms of number of beds' (HBK)' and 'low season' at the level of significance of 0.01, and variables HBHS and 'mid-season', and variables HBK and 'high season' at the level of significance of 0.05, as well as variables 'year of operation of the hotel' (GPH) and 'low season'. Based on the previously mentioned, it can be stated that there is a correlation of independent variables of hotel characteristics set in the research model with hotel occupancy in certain seasonal intervals, but no statistical significance of the relationship between all independent and dependent variables in the examined model was determined. Regression analysis showed the direction of correlation within the relationships of the examined regression models for the sample of hotels in the Republic of Croatia and in the Republic of Northern Macedonia. However, while the significance of connections within the model was determined in the sample, this was not the case with the sample in the Republic of Northern Macedonia. The regression model for the middle and high seasons had a low level of representativeness, and predictions within the model were not found to make a statistically unique contribution to hotel occupancy. Therefore, it is very important to explore other parameters, such as content in the destination, in order to find out about the factors that affect the seasonality of the tourist destination and the seasonality of the hotel business. Originality of the research The originality of this doctoral thesis is reflected in the creation of scientific and applied contributions to economic theory and practice. The scientific contribution can be expressed through the development of the theory of the relationship between the characteristics of small hotels and the seasonality of business, given the lack of knowledge in this area. The next contribution is in a systematic review and critical analysis of previous research on the impact of small hotel characteristics on occupancy during the year and building a research model to measure the relationship between individual characteristics of small hotels and occupancy in low, medium and high season. The variables used in the research based on previous research have been adjusted for the needs of the doctoral dissertation. Testing the research model gained new insights into the relationship between the characteristics of small hotels and occupancy in each season (low, medium, high) and the diversity of results related to geographical features of the destination and indicated the justification for linking hotel characteristics and seasonality. Objective knowledge about small business in the Croatian hotel industry and the impact of small hotel characteristics on business seasonality, obtained by the research, is the basis for proposing measures and management models in small hotels in order to reduce seasonality in the Croatian hotel industry. The results of experimental research provide a basis for future research.