Selecting the correct price for a product is a crucial decision often faced by marketers. Further complicating matters are packaging issues, as price is directly related to the product attributes and the positioning of the product offering in relation to the competitors' products. The traditional method  of solving the pricing and packing issues is the parametric assumption-full conjoint analysis (also known as trade-off analysis). The purpose of this article is to present a nonparametric assumption-free machine learning (model-free) alternative to the conjoint paradigm that eliminates some of the thorny practical matters of implementing the high-wrought conjoint analysis: incorporating nonlinearities and nonadditivities in the final conjoint model solution. Two cases studies are discussed comparing and contrasting the traditional conjoint and the model-free approaches for determining the optimal price and packaging based upon dozens of product attribute combinations.