A therapeutic impasse is a stalemate between a client and a therapist that grinds progress in therapy to a halt. Left unresolved, therapeutic impasses typically lead to the client dropping out of treatment (Weiner, 1974). Although there is some agreement in the clinical literature about what factors contribute to therapeutic impasses, there has been minimal empirical research in this area. Thus, the purpose of this study was to examine the factors associated with unresolved impasses that result in the client quitting therapy. A secondary purpose of this study was to examine whether therapeutic impasses evolve differently for clients with different styles of attachment to their therapists. In-depth interviews were conducted with 11 former psychotherapy clients who dropped out of therapy due to unresolved impasses with their therapists. Consensual Qualitative Research (CQR; Hill, Thompson, & Williams, 1997) was used to analyze clients' responses. In addition, clients completed the Client Attachment to Therapist Scale (CATS; Mallinckrodt, Gantt, & Coble, 1995), which assessed clients' degree of secure, avoidant-fearful, and preoccupied-merger attachment to their therapists. Results suggest that impasses in therapy are highly emotional events for clients. Whereas most of the clients reported that the therapists said or did something that bothered them, clients also admitted that they had difficulty expressing their dissatisfaction to their therapists, were reluctant to explore certain issues in therapy, and had issues or personality styles that interfered with therapy. Indeed, clients had significant pathology and may have been particularly difficult cases for their therapists. Progress in therapy also was impeded by disagreement over the structure and focus of therapy. Few clients felt that their therapists were aware that an impasse existed. Whereas most clients had an avoidant-fearful attachment style, those who also were high in preoccupied-merger attachment seemed to have especially negative experiences with impasses. The results of this study suggest ways in which therapists might decrease or resolve impasses, including assessing client variables that are associated with impasses, inviting feedback from clients about the process of therapy, and educating clients about therapy and the potential for problems.