Modern software assessments evolved out of audits. Quality audits for years have served to ensure that elements of the manufacturing process conformed to documented standards. By the late 1970s, though, there was a growing sense that examining the existence of individual processes was not enough, especially in the case of software systems. Phil Crosby, in a number of publications based on his experience at IT&T including Quality is Free , argued that old-fashioned quality audits, even in hardware settings, neither adequately measured the reliability of an organization’s development capabilities nor functioned as fulcrums for improvement. Instead Crosby proposed that quality assessments should shift their focus from individual processes to an organization’s maturity grid and evaluate the extent to which individual processes were successfully integrated into an organization’s institutionalized network of supporting structures. The result, he argued, would provide a sense of the real capability of a whole organization . His insight evolved into the notion of process improvement capability maturity levels. Following Crosby’s and later J.M. Juran’s work, in the 1970s and 1980s IBM began to experiment with its own capability maturity standards for predicting software development efforts, and for several years IBM personnel, including Watts Humphrey, Al Pietrasanta, Ron Radice, Maribeth Carpenter, and Harvey Hallman, stood at the forefront of the field of software assessment.