People often struggle to live up to their ideals because of a tension between what they actually want and what they ideally want. Unfortunately, rather than helping people with this, the way recommendation algorithms on social media are programmed may make this struggle worse. However, this need not be the case. Here we show that tailoring recommendation algorithms to ideal (vs. actual) preferences would provide meaningful benefits to both users and companies. To examine this, we built algorithmic recommendation systems that generated real-time, personalized recommendations tailored to either a particular person’s actual or ideal preferences. Then, in a high-powered, pre-registered experiment (N = 6,488), we measured the effects that these algorithmic recommendation systems had on individuals’ perceptions of the recommendation, perceptions of the company, and reactions after reading the recommended content. Our results suggest that both users and companies would be better off if recommendation algorithms learned what each person was striving for and nudged individuals toward their own unique ideals.