The growth in the information technology (IT) services market and the increasing tendency of firms to outsource some or all of their IT functions necessitate better mechanisms for selecting IT vendors. For most projects, there are a multitude of potential vendors that differ in quality and other aspects that are difficult to assess at the time of contracting. In addition, many projects have outcomes that are difficult to measure or verify by outside parties. As a result, mechanisms that require verifiability of outcome, such as incentive contracting, only provide limited benefits in vendor selection and in some cases are ineffective or counterproductive. This article presents an alternative mechanism for selecting high-quality vendors using a 2-stage contract. In the first stage, the client engages a vendor for a pilot project and observes the outcome. Using this observation, the client decides whether to continue the project to the second stage on prespecified terms or to terminate the project. By setting compensation for the pilot sufficiently low and establishing a threshold performance level for continuation, the client can offer a contract that is only attractive to high-quality vendors. Using game theoretic analysis we find that this contract performs better for the client than selection among seemingly equally qualified vendors. This mechanism is useful in settings where vendor quality is uncertain, and especially in situations in which a pilot project is undertaken for other reasons (e.g., demonstrating technical feasibility), where the benefits of this contracting mechanism can be realized at little incremental cost.

Download is the prior conference version. See journal for the published version.