Subjects completed a purchase decision under a computer-controlled laboratory simulation that enabled acquisition of both attribute-value and recommendation information from five sources. Analysis of subjects' acquisitions and recorded statements during the decision process offered support for the existence of unique decision strategies for situations involving availability of both kinds of information. Based on the manner in which recommendation information was used, these strategies were subdivided into (1) a recommendation-forms-evoked-set (RFES) for consideration, and (2) a recommendation-forms-standard (RFS) for comparison to other brand alternatives. Differences in strategy use between two product categories were explored.