A couple of years have passed since the so-called dot-com and researchers are now conducting post-mortem reviews of the valuation debacle. At the time, there was wide-spread agreement within the analyst community that the value of a web site was inherently related to the number of potential customers who came to the site for information and eventually purchased a product or executed a service. As a follow-up to this mode of thinking, we consider the relationship between stock prices and web metrics in addition to traditional accounting information for a sample of 15 top Internet companies. Specifically, we develop various regression models with the following four variables: unique visitors, revenues, gross margin and sales & marketing expenses. Our results support the hypothesis that web metrics did as equally good a job at explaining Internet stock prices as traditional accounting measures did. It seems that as stock price valuations for these companies increased, analysts required new metrics to justify their astronomical valuations and veered away from generally accepted accounting principles.