Purpose: Embedded within a primarily quantitative study of the predictive validity of consumer versions of two alliance measures– the WAI (Horvath, 1981) and the HAq-II (Luborsky et al., 1996), the goal of this qualitative sub-study was to concretize what consumers found helpful, not helpful, and would find helpful in their relationship with their case manager. As reported at SSWR in 2004, consumer WAI and HA-q II scores were generally positive; positive alliances predicted some consumer outcomes, nine months post-entry into ICM. Positive alliance scores, however, don't tell us in concrete and authentic ways what consumers are looking for in the case management relationship. Other than anecdotally, consumer voices are largely missing from the research literature on the client/case manager alliance, a gap which this qualitative analysis begins to address. Methods: Data from 67 consumers, from 19 non-urban agencies in Eastern Pennsylvania were gathered, using an exploratory, one-group, repeated measures design . Quantitative and qualitative data were collected at baseline and after three, six and nine months post-entry into ICM. Consumers had DSM-IV diagnoses of schizophrenia or schizo-affective disorder, were White, with equal numbers of men and women, with an average of nine lifetime hospitalizations, and involvement in the mental health system for 11 years on average. Using NVivo, verbatim responses to the following questions, asked of consumers at 3, 6, and 9 months post-entry into ICM, were analyzed: During the last three months, tell me about an incident or event when you did (or didn't) think your case manager was helpful. What would you have liked your case manager to do? Results: A combination of affective and instrumental responses was elicited about what they found helpful about their case managers: most common were “being there for me”, “getting rides” and “helping with money and budgeting”. “Not being there for me” and “not following through” were dominant themes about what they didn't find helpful. They would have liked more “social outings”, including eating out, shopping, going to a movie, or attending a sporting event. These all fall within the scope of acceptable ICM services, but may be perceived as not feasible, if caseloads are high (Average caseloads in this study were 18). At the same time, their helpfulness to consumers may not be recognized or valued by some case managers. Implications for Research and Practice: Findings suggest that consumers can provide concrete and insightful information about the specifics of their relationships with their case managers. That so many expressed strong desire for more social interactions with case managers may reflect excessive free time, without adequate diversions, suggesting areas for further service development. These desires may also reflect the wish to develop more social connections and rejoin the social world, at whatever level is comfortable, which, according to Jacobson and Greenley (2001), is vital to the recovery process in mental health. Future research on the alliance in ICM should include qualitative as well as quantitative data gathering and should be extended longer than nine months.