Enterprise Resource Planning (ERP) projects are considered commercial-off-the-shelf (COTS) solutions where one substantial COTS product (e.g. Oracle, SAP) is tailored to provide automated business functions. This paper presents effort and schedule estimating models for predicting ERP implementation. The first two models use product size to predict software development effort. Product size is measured in terms of the number of report, interface, conversion, and extension (RICE) objects that are tailored or modified in the COTS product suite. The third model predicts the duration of software development phase as a function of RICE objects and full time equivalent staff. These statistical models are based on empirical data collected from 22 programs implemented within the federal government over the course of ten years beginning in 2001. Result shows that number of RICE objects is a good predictor of ERP software development effort and duration.