|
SPSS Launches Predictive Web Analytics | May 22 2003 |
|
SPSS has this week announced the launch of Predictive Web Analytics, a solution allowing companies to transform their web data into deeper customer intelligence. Unlike traditional web metrics, Predictive Web Analytics aims to provides meaningful historical and predictive insights, so maximising customer value and increasing returns from web investments. Predictive Web Analytics integrates the company's new NetGenesis 6.0 Web analytics platform with the proven predictive modeling capabilities of its Clementine (R)7.0 data mining workbench. The combination of these technologies enables Predictive Web Analytics to move beyond simple web metrics -- such as page counts, unique visitors and referral information -- to deliver insights that can be used to directly affect customer relationships. Predictive Web Analytics provides four essential analytical capabilities: segmentation of visitors based on their behaviour, detection of content and product affinities, automatic identification of the most significant paths taken through a website, and prediction of visitors' propensity -- for example, to purchase, to view particular content, or to churn. By providing historical and predictive web metrics, Predictive Web Analytics identifies opportunities that increase sales and improve customer satisfaction, enabling companies to quantify the return on investment (ROI) of complex web initiatives. 'For years, organizations have used SPSS predictive analytics to capitalize on high volume customer data from a range of offline channels,' said Colin Shearer, vice president of customer analytics at SPSS Inc. 'We developed Predictive Web Analytics so that organizations could have that same level of insight into their online business. Predictive Web Analytics allows organizations to maximize the value of their web channel by creating a 360-degree customer view that drives profitable multi-channel relationships.' For further information visit www.spss.com Simple printable version of this article. Email this article to a friend.
|
|
