Within the services marketing literature there is a plethora of research on service failure and recovery, in particular with regards to outcomes such as customer satisfaction with service recovery as well as alternate evaluations of service recovery performance. However, to date, there is very limited literature on reconciliation as an outcome, particularly in conjunction with other dyadic conflict-related behaviours such as retaliation and avoidance, and other factors which predict these conflict-related dyadic states. By presenting three studies, study A which is a published manuscript, and B (unpublished manuscript) which is mostly separate from study A, and study C (unpublished manuscript) which synthesises constructs from both studies A and B, this thesis attempts to answer in a positivistic manner, the following research questions: RQ1: What is the role of service failure severity in predicting reconciliation, retaliation and avoidance after service failure? RQ2: What is the agent likeability’s role in predicting outcomes such as reconciliation and retaliation after service failure? RQ3: What role does agent empathy have in predicting outcomes such as reconciliation, retaliation and avoidance after service failure? RQ4: What role does an agent apologising sincerely have in predicting outcomes such as reconciliation, retaliation and avoidance after service failure? RQ5: What role does service recovery justice have in predicting outcomes such as reconciliation and retaliation after service failure? RQ6: What kind of post-recovery behaviours does reconciliation entail? Study A utilised a quasi-experimental survey approach where participants were sourced from a university library. Both studies B and C used a survey that required participants to recall a service failure and answer all questions based on that recollection, and where data from both studies B and C were collected through an online panel. Study A for the first time introduces the concept of likeability of the service agent to the literature regarding service failure and recovery and related this to outcome variables such as customer satisfaction with service recovery as well as reconciliation. The results of study A indicated that likeability of the agent was important in predicting customer satisfaction with service recovery and also reconciliation. The implications for managers of this study are to hire from within the customer base or target market since individuals are likely to reconcile with others who are similar to them when misdeeds are committed, as is the case with service failure. Study A is presented as a published manuscript and is relatively short in length. The purpose of study B was to examine the mediating role of reconciliation between failure severity and customer retaliation/avoidance and the moderating roles of empathy and apology in this path model. The findings from study B indicated that both empathy and apology had moderating effects between service failure severity and the mediator variable of reconciliation and the dependent variable of retaliation; but both did not moderate avoidance. Also, reconciliation mediated the relationship between service failure severity toward retaliation and avoidance, while service failure severity also had direct relationships with the two dependents. The implications of this study include that service agents need to be screened for qualities such as empathy and ability to be apologetic which can help ensure better outcomes following service failures. Study B, since it is mostly distinct from study A, is presented as a short unpublished manuscript in comparison with study C. The purpose of study C was to combine and expand on the findings of both studies A and B. In this study, the constructs of likeability of an agent, service failure severity, empathy/apology, distributive justice and interactional justice as well as reconciliation and retaliation were further examined. It was determined that the likeability of an agent, similar to study A, positively predicted reconciliation, but that also, counter to expectations, it also positively predicted customer retaliation. Moreover, it was also found that empathy and apology had moderating roles between agent likeability and distributive justice, whereas service failure severity had a moderating role between agent likeability and interactional justice. Lastly, distributive justice played a mediating role between agent likeability, and reconciliation and interactional justice had a mediating role between agent likeability and retaliation. This research reinforces the implications of study A and B by demonstrating similar roles of likeability, failure severity, empathy and apology in the service recovery process, however, adds to studies A and B by highlighting that there is also a negative side to likeability where customers may react more negatively to service agent’s they deem more likeable. Study C is presented as a very long manuscript since it resembles, in terms of breadth, the overall topic of this thesis. Studies A and B act as shortened manuscripts since they do not resemble what this overall thesis is about in terms of them being standalone articles. Overall, this thesis underpins reconciliation as an important construct concerning critical incidences, and to a lesser extent retaliation and avoidance, particularly within the realm of service organisations, with agent likeability, empathy and apology having significant effects on such outcomes. All constructs involved, particularly in study C, denote a framework postulating prosocial behaviours and traits as pertinent to resolving conflict amid service failure and hence can act as a gauge for service organisations to assess their relationships with customers and clients and how they can respond to ensure business success.