This is the first report showing that SASB can be used for the content analysis of dreams. The present study uses clinical examples originating from psycho-oncologic outpatient treatment in order to show the chances and limitations of SASB content analysis. Successful rating requires evaluating all three dimensions of SASB (focus, affiliation and interdependence). There has to be a sufficient quantity of interactions between referents, i.e., the actors. Non-animated objects can be referents in SASB content analysis provided they are important for dream action. To code descriptions and affective states, they must be paraphrased into actions. Dream interactions described in the SASB framework may express the dreamer‘s habitual, more or less conscious interaction patterns. Furthermore, dream interactions may reflect unconscious fears, wishes, conflicts, and problem solutions. Similar to role playing and free association, the formulation of cyclic maladaptive patterns may provide an important source of creativity to the dreamer. SASB assisted dream content analysis may critically accompany clinical hypotheses. In the context of a SASB directed dynamic psychotherapy, empirical results can stimulate the process of focussing. SASB can be used for coding dream texts as well as dream induced association chains. The present study also reveals certain limitations of SASB content analysis: Dreams predominantly expressing emotions without clear actions (including actions inward, introject focus III) cannot be coded except by infering the underlying interaction patterns. The same is true for highly descriptive dreams which may be extremely interesting at the symbolic level but which cannot be rated in the absence of interactions.