Skip to Content

Hands-On Tutorial SAP Predictive Analytics, Automated Mode: Classification

This hands-on guide explains how a Business Analyst without deep statistical know-how can create a predictive model to identify customer contacts that are most likely to respond positively to a Marketing campaign.

SAP Predictive Analytics, Automated Mode is creating the predictive model that scores customers on their interest for a certain product. This analysis is based on the outcome of a previous Marketing campaign, which is stored locally in a flat file.

Since the data used in this guide is publicly available, the reader can follow the steps hands-on and carry out the same analysis.

Please note that this guide is giving a high-level introductory overview and only shows a small fraction of the available functionality.

Download from GitHub (when the pages open, click on “Raw” to save locally)

Happy Predicting!


You must be Logged on to comment or reply to a post.
  • Hi Andreas,

    Market campaign analysis classification tutorial is good and in detail.

    When we apply the model to sample data set generating the  values for “proba_rr_y” .  As per the tutorial , If we sorted the output on “proba_rr_y” and give 10% to the marketing team for the campaign.

    Here how can link the “proba_rr_y” values to input . Because in the output we have only KxIndex, y, rr_y, proba_rr_y and bar_rr_y columns.

    Can I link “proba_rr_y” values with input data based on KxIndex. Is it work as row_id for the source data?

    Could you please explain how can i find my source record which has more proba_rr_y value.

    Thanks & Regards,


    • Hi Andreas,

      I got it . You explained in the additional settings . Rowid column we created in the source that column will be used to recognize the source record.

      Thanks & Regards,