710011Website Visitor Interaction Prediction AnalysisClosedGRD Proof of ConceptInnovate UKCompanies with a web presence spend vast sums attracting potential customers to their site, but when customers show an interest by making contact, their communications are not maximized to generate business and increase customer satisfaction. In fact, research shows that many commercial web sites have a low 'relevant' response rate (as low as 60% in many cases), leading to lost opportunities and revenue, and reduced customer satisfaction. We believe that by providing companies with a solution that intelligently analyses their customer interactions they can save costs and generate new business. Arcadium created ContactUsPlus to allow companies to manage customer interactions through their website. ContactUsPlus is a toolbar that can be easily added to any website and provides a contact form, live chat, call back facility, email integration and, uniquely, a full history of all previous interactions for both customer and representative. ContactUsPlus has been installed on 4000+ websites in the first 3 months since launch, with exposure to 100,000 web site visitors per month. The new business opportunity that we have identified is to provide business intelligence to companies by analysing and learning from their interactions with their web site visitors. The proposed proof of concept is to design and develop a unique set of algorithms that will analyse and 'learn' from web site visitor interactions, providing businesses with intelligent 'conclusions' to be used in serving their customers. The learning process ensures that the accuracy of our solution will increase over time across all of our customer base. Below is a sample of cost savings that our customers could make by using the proposed solution: 1. Automatically routing interactions to the correct representative/department, based on 'learning' from previous interactions. This will remove the need (and associated time/cost) to manually route messages, or to asking the customer to specify - an activity that reduces likelihood of sending the message in the first place). 2. Automatically suggesting a suitable responses to customer queries, based on previously learned behaviour - this will save cost/time in searching or creating a new response. 3. Automatically identify the likelihood of a purchase based on message content - this will enable our customers to focus on those web site visitors that are most likely to make a purchase, based on 'learned' behaviour form past interactions. 4. Automatically identify customers suitable for an up-sell based on message content and learned behaviour - leading to increase in sales based on targeted up-sell/marketing. By innovating this 'intelligent' layer and adding it to ContactUsPlus we believe that we can offer a unique proposition to the market that will lead to genuine cost savings and increased revenue opportunities for our customers. We will in turn leverage this innovation to gain a genuine competitive advantage, and grow our business.