The concept of precision agriculture is based on the ability to improve the management of production factors using site-specific information. The optimal configuration of management zones for more precise management of farm inputs is one of the most important components in precision farming. The objective of this study is to develop a management zone delineation procedure based on a spatial statistics approach and evaluate its economic impact for Texas cotton production. Using an optimization model that utilizes a yield response function estimated from field experiment data through spatial econometric methods, we evaluate the economics of the management zone delineation procedure. We found that applying variable N rates based on the management zones delineated would result in higher cotton yields and higher net returns, relative to a uniform rate application based on field information and a variable rate application based on landscape position. This is indicative of the potential economic value of using a spatial statistics approach to management zone delineation in cotton production.