Advances in technological support for augmented personal memories make possible new ways of enhancing the process of product recommendation. Instead of simply analyzing information about a user’s past behavior in order to generate recommendations, a recominder system can additionally supply various types of information from the user’s augmented memory that allows the user to take a more active role in the search for suitable products. We illustrate the paradigm of recomindation with reference to a prototype implementation of the system Specter in a CD shopping scenario and the results of a study with 20 subjects, who found most of the recomindation functionality to constitute a useful enhancement of their shopping experience.