Chunk #7 — Building and Supporting the REDCap Consortium: A Phased Approach — Phase 1: Developing REDCap at Vanderbilt University (1 site: ~ 2004–2005) — Challenges:
Building novel informatics tools required trust and investment from the local research community. Early stages of software development included extensive work sessions with users and rapid cycle development designed to translate stated concerns and needs into software modifications and improvements. Building an initial base of users at Vanderbilt was laborious and required establishing trust among the research community that the software would be supported over a long period of time. Early adopters were invaluable at helping improve the software through feature suggestions and by championing the REDCap system to peers. In developing the software platform, we focused on an agile delivery method where practical solutions to investigator needs were implemented with rapid turnaround rather than ‘perfect architecture.’ This approach allowed us to address the needs of our institutional investors while obtaining rapid user feedback on feature implementation. We adhered to two strict principles when meeting with research stakeholders: 1) never promise a new feature or timeline, as this unnecessarily overcommits the development team and potentially places researchers at risk if features/timelines cannot be delivered; and 2) always promise to build new features/functions in a way that ensures backward compatibility with all existing projects.
