The web browser has become a central workspace for knowledge workers, where they make use of cloud-based applications to access and store their information. While this solution helps reduce the difficulty of syncing information between our numerous devices, it reintroduces and proliferates faults of the desktop, particularly information fragmentation. Information fragmentation is an increasingly important issue, as cloud-based applications typically silo their data, resulting in a replication of storage and organization in the absence of a unifying structure. To probe whether knowledge workers encounter information fragmentation and in what manner, we created Contextinator, a tool that assists in coordinating data for web-based projects. Contextinator provides a method for providing the centralized, unifying structure that cloud based storage makes difficult. Our findings contribute insight into the need for, and appropriateness of, projects as a unifying structure for the web. Our results point to two types of projects that we term ‘preparatory’ and ‘opportunistic’ based on when and for what reason users create them. We discuss the design of our system, the results of our mixed-method evaluation, and our observations about information fragmentation on the web.