The increasing use of the Internet creates a need to manage traffic while preserving equal treatment of content. We estimate demand for residential broadband, using high-frequency data from subscribers facing a three-part tariff, and use the estimates to study the welfare implications of usage-based pricing, a commonly offered solution to network congestion. The three-part tariff makes data usage during the billing cycle a dynamic problem; thus, generating variation in the (shadow) price of usage during the month. We provide evidence that subscribers respond to this variation, and use their dynamic decisions to estimate a flexible distribution of willingness to pay for different plan characteristics. Using these estimates, we show that usage-based pricing eliminates low-value traffic and improves overall welfare. Usage-based pricing might decrease consumer surplus, depending on what alternative is considered. Furthermore, we show that the costs associated with investment in fiber-optic networks (an alternative proposed to deal with congestion) are likely recoverable in some markets.