Fifty to seventy percent of information system projects fail. Most of the failures are not the victims of flawed technology, but rather organizational and people related issues. When Vanderbilt University Medical Center began an intensive electronic health record (EHR) effort, a process was carefully designed to select the clinical areas where new tools could be developed and pilot tested. The Success Factor Profile was created to guide the selection of sites most likely to have innovation success. This paper describes both the tools and the processes used to select clinical sites for new computer tools development and pilot implementation. Early results demonstrated that the tools provided structure for the decision making process, permitting side-by-side comparison of "apples and oranges." Selecting the site most likely to succeed with computer application innovation and early implementation has broad applicability in healthcare informatics. Failure to succeed with early system users is not only costly, but also discourages users and developers alike, and may damage the reputation of the tools and systems across the institution.