People prefer a certain gain over an uncertain one (e.g., Kahneman and Tversky 1979; Gneezy, List, and Wu 2006), and uncertainty threatens the perception of personal control (e.g., Simonsohn 2009; Sonsino 2008; Gaucher, Hafer, Kay, & Davidenko, 2010). Nevertheless, research has also showed that under certain circumstances people love uncertainties as “surprises” can add unexpected pleasures to our life (Goldsmith and Amir 2010; Shen, Fishbach, and Hsee 2015). In this research, we examine consumers’ preference for surprise in the subscription services. A typical subscription business offers a “surprise-me” option, where customers receive the package without knowing what items are included, in addition to (or instead of) customers making their own selections (referred to as the “self-selection” option hereafter). Despite the surging popularity of this business model, little is known whether consumers prefer making their own selections or are willingly let providers surprise them. Theoretically, it offers an opportunity to examine consumers’ willingness to relent their control in choice when there are risks involved. Managerially, answers to these questions can help service providers better design the surprise-me option for both operational efficiencies and customer satisfactions. Our key hypothesis is that a sense of personal control is an important condition for consumers’ preference of the surprise-me option in subscription services. Feeling of control serves as a safeguard and thus allows consumers to attend to the pleasures of uncertainty. This also means that in situations where consumers do not feel in control, their preference for the surprise-me option decreases (relative to the in-control condition). This is because, the compensatory control theory (Whitson & Galinsky, 2008; Kay et al., 2008; 2009) suggests that when consumers feel not in control, they tend to compensate by trying to take things into their own hands (e.g., self-select) to regain control rather than relenting control to others (e.g., to be surprised). However, when it is difficult to regain control through self-selection due to choice difficulty (e.g., lack of topical knowledge), they may again prefer surprise as a way to simply delegate the choice (Aggarwal and Mazumdar 2008; Steffel and Williams 2017). Thus, we expect feeling of control to play an important role in preference of a surprise-me option. We conducted three studies to test this prediction. Overall, the set of studies supported the expected important role of feeling of control. The surprise option is valued only when consumers do not lack the sense of control hence do not need to regain it through choice or the chance of regaining control through choice is low due to choice difficulty (expertise or product assortment characteristics). Our research provides important theoretical contributions.