The article considers the use of technological templates for calculating the predictive labor intensity of machining. The processing of parts of the “sleeve” type on CNC lathes is considered. The process of determining the predictive labor intensity of structural and technological elements (features) of a part is implemented as an applied CAD module. At the same time, the contour of previously recognized features in the form of an electronic 2D model and a table of processed surface parameters are fed to the module input. The use of technological algorithms becomes possible due to the preliminary automatic feature recognition, which is implemented as a module of the considered CAD. At the recognition stage, various types of features are classified (flat end, open area, semi-open area, closed area, and thread). The result of recognition is the contour of individual features, consisting of individual segments of straight lines and arcs of circles and specified in the form of standard G-codes. The article describes the principle of developing new mathematical models of technological templates for feature data processing. Technological templates include automatic selection of cutting tools from the user base, selection of a toolpath strategy and selection of processing modes. This takes into account the location of the contours relative to the axis of the part (internal and external), and for closed areas, their geometric type is taken into account (undercut, groove, groove for thread exit, end undercut, and etc.). Such decision-making algorithm detailing is necessary because the choice of tool and toolpath depends on it. In order to take into account the part properties, the coefficient of labor intensity of materials is introduced. Further, technological templates include algorithms that use feature contours described in the form of G-codes to generate a control program (NC) for a CNC machine. To determine the labor intensity, a standard NC-simulation module is used. The results of model adaptation for the current production type are obtained.