This paper will discuss the problem with successful predicting of product performance (reliability, quality, durability, safety, recalls, profit, life cycle cost, and other interconnected technical and economic components of performance). The best component for analysing the performance situation during service life, including predicting, is recalls, because, first, recall accumulates the safety, reliability, durability, quality, profit, and total economic situation. And second, there is open official and objective information about the number of recalls from Government (National Highway Trafic Safety Administration and others), as well as companies-producers.Therefore, for analyzing the situation with the product performance, including predicting, this paper considers the situation with recalls. First, it will demonstrate how dangerous the current situation is with recalls, safety, reliability, and durability, especially in automotive, including in the USA for last thirty years. Then it will be demonstrated that recalls directly connect with profit.It will analyse the basic causes that leads to unsuccessful predicting. This paper proposes the basic way to eliminating the above situation. This way consists of appropriate methodology and obtaining initial information for predicting the successful specific product performance. This way also shows how one can solve the above problem through increasing the interconnected reliability, durability, safety, and life cycle cost.