The internet presents people with an increasingly bewildering variety of choices. Online consumers have to rely on computerized search tools to find the most preferred option in a reasonable amount of time. Recommender systems address this problem by searching for options based on a model of the user's preferences. We consider example critiquing as a methodology for mixed-initiative recommender systems. In this technique, users volunteer their preferences as critiques on examples. It is thus important to stimulate their preference expression by selecting the proper examples, called suggestions. We describe the look-ahead principle for suggestions and describe several suggestion strategies based on it. We compare them in simulations and, for the first time, report a set of user studies which prove their effectiveness in increasing users' decision accuracy by up to 75%. Copyright 2006 ACM.