Increasing dialogue efficiency in case-based reasoning (CBR) must be balanced against the risk of commitment to a sub-optimal solution. Focusing on incremental query elicitation in recommender systems, we examine the limitations of naive strategies such as terminating the dialogue when the similarity of any case reaches a predefined threshold. We also identify necessary and sufficient conditions for recommendation dialogues to be terminated without loss of solution quality. Finally, we evaluate a number of attribute-selection strategies in terms of dialogue efficiency given the requirement that there must be no loss of solution quality.