Nearly all businesses face the challenges of the “Data Deluge” created by increasing deployment of ecommerce, web and mobile systems. All are amassing huge data sets but have few if any adequate ways to make any sense of these data assets. This data stream is of immediate commercial and competitive concern and will continue to grow exponentially. liveRES Limited propose a Proof of Concept project to approach and develop advanced theory to exploit such data assets. Where Latent Semantic Indexing [LSI] has been the standard for data complexity reduction, the project will explore and produce a practical working example uniquely developing on Random Indexing – already a hybrid successor of LSI. Critically, to date activity addressing RI has focused on tackling documents and not data. The natural initial candidates are data documents in XML, and the project will use actual restaurant data. The project, designated ‘Restaurant & Retail Optimisation through Bookings and EPoS data Reduction Techniques’ [RROBERT], seeks to model harnessing the signature vectors generated by Random Indexing. It is predicted that not only will this innovation in RI represent the events and facts within a transaction, but also represents the order and timing in which events occurred, thus giving a new dimension to the geometric comparison of single or groups of the transactions (or groups of transactions). The objective is to prove that this data can be subject to dimension reduction, allowing speed of operation but without destruction of data content as part of the indexing operation. The output will be translational research from theory to a model of restaurant data environment that create subjective comparisons between customers (their booking personas) and between bookings. The commercial application will be an intuitive software aid for the modern restaurateur, which will enhance online promotion and bookings. The technology can then be theoretically extended to other data groups.