The ‘Project Risk Screening Matrix’ derives from a broader effort to assist US government agency staff in reviewing proposed stream management and restoration projects more efficiently and effectively. The River Restoration Analysis Tool (RiverRAT) developed through this effort provides a thorough, comprehensive and auditable approach to review and evaluation of proposed stream actions and projects (www.restorationreview.com). The matrix was initially developed as the first step in applying the RiverRAT, its purpose being to assist reviewers in assessing the risk to natural resources associated with a particular proposal and matching the intensity of their review to the severity of that risk. Hence, the primary application of the matrix to date has been to identify and screen out low risk projects that may be dealt with expeditiously, and so freeing the time and technical resources needed to allow deep reviews of higher risk projects. A second form of screening emerged from this primary function because the matrix proved adept at identifying the minimum level of site and project characterization required to support initial risk assessment. On this basis, proposals lacking adequate information can also be screened out, being referred back to the proponent with a request for additional information. More recently, new and novel versions of the matrix, featuring modification and refinement of one or both of the original axes, have emerged to widen and refine its application to linear infrastructure (e.g. pipelines, roads, and electrical transmission lines), instream structures (e.g. large wood placement and culvert removal), and pre-application, regulatory, decision-support tools.