BI SIG: Doing Customer Intelligence with R



  • Find out about R, an Open Source system for writing analytic applications



    Jim Porzak, Director of Analytics, Loyalty Matrix Loyalty Matrix


    Presentation Overview


    Doing Customer Intelligence with R

    Traditional Customer Intelligence tools have evolved from more general business intelligence platforms. Thus they typically inherit the high cost and closed nature of these offerings. At Loyalty Matrix our proprietary customer intelligence platform, MatrixOptimizer®, is based on Microsoft’s SQL Server 2000 including Analysis Services for OLAP. While OLAP does provide flexible analysis and presentation methods, it is somewhat limited in presentation flexibility and, of course, has no advanced statistical or data mining methods.

    The open source data analysis and graphics system R fills this need. Over the last four months, we have integrated a number of R methods within the MatrixOptimizer (MO) so that many charts are automatically produced for EDA (exploratory data analysis). Additionally, we have integrated marketing campaign effectiveness statistics and charting functionality into MO using R. Since the basis of MO is a standard data warehouse schema, ad hoc R analysis can focus on methods and interpretation; not on data access chores.

    After a brief introduction to R and MatrixOptimizer, we will present the customer intelligence methods we have implemented in R as real case studies. Also, we’ll cover how we do the R to SQL Server integration and offer some hints on how you can take advantage of R’s power and flexibility. See for background.


    About the Presenter

    Jim Porzak

    Jim Porzak is Director of Analytics at Loyalty Matrix in San Francisco where his team is part of the client services group. Jim has been using R for over two years. He presented a paper on customer intelligence at the first R users’ conference, useR! 2004, in Vienna this spring.

    Jim has over 35 years experience designing, creating, delivering and marketing data and computing solutions in the US and Europe. Prior to joining Loyalty Matrix, he had his own consulting and training group, InfoIS, which provided data base analysis solutions to hundreds of clients and trained thousands of students throughout the West.

    Other prior positions include VP of Sales and Marketing at Bytel Corp – the former programmer tools company in Berkeley. In Brussels, he was a founding partner and Director (SVP) of the Standard Software Division at Vector International, a microcomputer engineering company, which launched Microsoft and Digital Research in Europe.

    Jim earned a BA in Math and Physics from Wabash College and pursued Physics at Wayne State University before specializing in computer applications.


    Event Logistics


    Cubberley Community Center

    4000 Middlefield Road, Room H-1

    Palo Alto, CA


    6:30 - 7:00 p.m. Registration / Networking / Refreshments (please arrive before 7:00 p.m.)

    7:00 - 8:30 p.m. Presentation and Discussion



    $15 at the door for non-SDForum members

    No charge for SDForum members

    Please call 408.494.8378 for student memberships

    No registration required

    More on the Business Intelligence SIG....