This paper presents a method to forecast the sales path of an improved version of a high technology product defined in terms of its price path and multiattribute product specification. The approach is potentially useful to managers to answer what if questions on the effects of alternative price paths and product specifications of the upgrade on when and how many of their customers will upgrade. The proposed approach integrates an individual-level conjoint utility model with a hazard function specification. An illustrative application to the personal digital assistant (PDA) category confirms the predictive validity and potential usefulness of the proposed approach. Among the empirical findings are that higher upgrade costs and expectation of faster product improvement tend to delay buyers upgrading decisions.