Web Site Performance Analysis Success Assessment of Information Driven Web Site on User Traces

Web metrics help to identify improvement potentials for web sites. In contrast to transaction based sites, the success of web sites geared toward information delivery is harder to quantify without direct feedback out of a transaction. We propose a generic success measure for information driven web sites by observing the users in context of the web site semantics. Thus we identify target pages, analyze the web page content and evaluate effectiveness and efficiency of the user actions with respect to the web site’s objectives. The user’s perspective has to be incorporated for a comprehensive success measure. We propose to integrate search queries from referrer information carrying the user’s intentions. Out of an integrated web site meta model we derive formally a new success measure. This approach uses common data mining techniques and text mining algorithms like PLSA and shows its applicability in two case studies and an independent user enquiry.