Service organizations prescribe the appropriate emotional display expected from employees during interactions with customers in order to maintain a reliable service quality. The requirement to display certain emotions by frontline employees is called emotion work. The aim of this doctoral thesis is to develop a better understanding of this construct. To this end, three research studies were conducted with the participation of 60 Spanish hotels. The initial sample consisted of 459 front-line service employees (waiters and receptionists) distributed in 117 work-units and 1069 customers. The first study describes and validates the Frankfurt Emotion Work Scale as a measure of emotion work in the Spanish language. Results confirm that construct and predictive validity are satisfactory, with six factors describing different dimensions of emotion work: the requirement to display positive emotions, the requirement to display negative emotions, the requirement to display neutral emotions, the requirement of being sensitive to customer’s emotions, interaction control and emotional dissonance. The second study aims to clarify the effects of supportive climate on emotional exhaustion. Supportive climate was defined as the team shared perceptions regarding the existing levels of friendliness and cooperativeness among team members. The proposed model assessed direct and moderating effects of working in a supportive environment on emotional exhaustion. The results confirmed both hypothesized effects showing that supportive climate acts as a contextual factor that protects employees from suffering emotional exhaustion. On the one hand, supportive climate at the work unitlevel had a negative and significant relationship with emotional exhaustion. On the other hand, we found a moderator effect of a supportive climate on the well-established relationship between emotional dissonance (the core dimension of emotion work) and emotional exhaustion. Our results reveal that employees who work in units with a low supportive climate are more vulnerable to feeling emotional exhaustion after experiencing emotional dissonance compared to employees who work in a unit with a high supportive climate. Consequently, a supportive climate protects employees who experience emotional dissonance from suffering emotional exhaustion. Finally, the third study proposes a model where emotion work (as a work-unit variable) was related to customer outcomes (at the individual level). The main objective of this study was to analyze how some dimensions of emotion work (the requirement to display positive emotions, the requirement of being sensitive to customer’s emotions and emotional dissonance) may affect customer satisfaction and loyalty. Firstly, the results supported the idea that frontline employees pertaining to the same work unit develop shared perceptions of emotion work beyond individual differences. That is,emotion work can be conceptualized and measured as a work-unit variable. Second, in contrast to our expectations work-unit positive emotional display was not associated with customer outcomes. And finally, the results show that work-units with higher sensitivity requirements and lower levels of emotional dissonance have more satisfied and loyal customers. In general, this doctoral thesis introduces emotion work as an important job demand for frontline employees. Emotion work affects employee’s wellbeing (measured as emotional exhaustion) and working in supportive environments (high levels of supportive climate) can alleviate the emotional exhaustion experienced as a result of emotional dissonance. In addition, this doctoral thesis offers a new perspective of emotion work (as a contextual variable) and its power to predict customer outcomes(customer satisfaction and loyalty). Finally, this doctoral thesis integrates differentliteratures (marketing, wellbeing and emotion work) and considers recent multilevel methods and multiple informants (customers and employees).