Metered pricing plans for services enable companies to increase their profits. Yet measuring consumer preferences for different forms of metered pricing is difficult, because metered prices simultaneously influence three consumer decisions: to purchase the service, to choose a particular pricing plan, and to use a particular quantity. These decisions strongly influence the number of customers that use the service, their usage, and profit. This article develops and validates augmented conjoint analysis methods that capture the interplay among these three decisions and allow for predicting the effects that different metered pricing plans have on consumer behavior and company’s profit. The empirical study reveals that the optimal two-part pricing plan yields 36–49% higher profits than optimal pay-per-use or flat rate pricing plans. Consumers’ reactions to changes in metered pricing plans are very heterogeneous. The fixed fee of a two-part pricing plan strongly influences the number of subscribers but hardly influences their usage. In contrast, changes in marginal prices strongly affect consumers’ usage but not their subscription. Data collected through ranking- and choice-based conjoint analysis yield comparable willingness-to-pay estimates and substantially outperform contingent valuation. Market researchers should also use pricing plan formats instead of usage formats to elicit the preferences for two-part pricing plans.