The issue as to whether investments in information technology (IT) contribute significantly to organizational productivity has been of major concern for many years, and various studies have lead to seemingly contradictory results. In this paper, we analyze the relationship between IT investments and firm level productivity using regression trees (RT). Use of this data mining technique represents a novel approach to identifying elements of this relationship as most previous studies have primarily used econometrics-based techniques. While the use of traditional techniques has provided valuable results, our RT-based analysis revealed additional findings that were not identified in the previous studies. For example unlike the econometric-based studies that identify a uniform impact of IT investments on productivity, our RT-based analysis suggests that IT has an impact on productivity only when Non-IT Labor expenses are within the interior interval. Also, even within this range, the impact of IT is not uniform.