How do you use Big Data to drive Marketing Optimization?


  • Author: Lillian Pierson
  • Date: 03 Jul 2013
  • Copyright: Image appears courtesy of iStock Photo

The Evolution of [X+1]

[X+1], or [X+1] Origin Digital Marketing Hub, is a marketing data management platform for the storage, integration, analysis, and visualization of marketing data. [X+1] helps marketers optimize marketing efforts, allowing them to increase rates of customer acquisition, manage purchasing funnels, upsell, and better target their audiences. The [X+1] Enterprise Data Management Platform (DMP) provides powerful big data-driven analytics to support the efforts of marketers from major client’s like JP Morgan Chase and Verizon.

In the recent years [X+1]’s customer base has been increasing rapidly. The Origin DMP currently houses terabytes of data and additional terabytes of data are added to the platform each month. It is not just data volumes that are increasing rapidly, the quantity of highly-skilled marketers that are using [X+1] is increasing as well. With these increases, the demands and customer expectations of [X+1] have increased dramatically.

In the not too distant past, marketers began asking more pointed questions of their data and demanding deeper insights into the data patterns that surround their audience targeting efforts. Marketers wanted to be able to better measure the return on investment for their efforts. What marketers really needed and wanted was to be able to get a better picture of what was happening with customers in real-time, but to do this required the ability to utilize terabytes of marketing campaign data in real-time analytic models. It was at this turning point where [X+1] executives clearly saw that they would have to reconfigure the platform’s analytic solution in order to produce the type of big data analytics that their clients were demanding.

thumbnail image: How do you use Big Data to drive Marketing Optimization?

Revolution R Enterprise Provides Scalable Big Data Analytic Solutions

To facilitate this, back in 2011, [X+1] Chief Analytics Officer Leon Zemel made the decision to integrate [X+1]’s Origin Enterprise DMP with the Revolution Analytics’ Revolution R Enterprise predictive analytics platform to produce the [X+1] Predictive Optimization Engine (POE). Revolution Analytics is a provider of advanced analytics solutions. It’s built on the open source R statistics language, but it has significant performance enhancements that are able to drastically cut computational time when compared to standard open source R. The Revolution R Enterprise solution allowed [X+1] to scale out to support more hardware devices and increased the reliability of the growing [X+1] platform.

Utilizing the Revolution R Enterprise predictive analytics platform within [X+1]’s Origin Enterprise Data Management Platform has helped [X+1] achieve unprecedented marketing optimization results. These results are due to the power of the predictive analytic models that are derived from Revolution Analytics’ RevoScaleR package for Big Data Analytics. Here is a brief summary of how the predictive optimization engine works. Marketers connect their relevant data sources and then enter other known factors about their model. Data sources often include internal sources like CRM systems, call center transcripts, and client website activity logs, along with external sources such as socio-demographic segmentation, behavioral data, search data, and web-user affinity information.

This mass of unstructured “big data” sits on the cloud-based [X+1] DMP. The DMP then feeds the Revolution R Enterprise where the data streams are mapped and reduced in Hadoop before being migrated to structured RDBM systems for use by data scientists at the analytics layer.

The Revolution R Enterprise solution allowed [X+1] to scale out to support more hardware devices and increased the reliability of the growing [X+1] platform.

Next, automated statistical models are produced, or, as in the case of custom models, data scientists work with the data at the analytics layer to produce a near real-time predictive custom model. These models are then hosted on a server that is designed to monitor the client-side interface (like a webpage browser or a mobile device) and collect small streams of real-time data that can be returned to the model and processed instantaneously by its predictive algorithms.

These real-time predictive analytic models have revolutionized the marketing optimization sector and have demonstrated impressive improvements in marketing performance. Revolution R Enterprise algorithms in [X+1] POE assist marketers with improved audience discovery, helping them to quantify and benchmark audience response rates and quality. In addition, through the [X+1] POE marketers are now able to see exactly where they stand with respect to their own current and anticipated (predicted) budget constraints. Lastly, a leading features of the [X+1] predictive optimization engine is the predictive model that it generates to quantify the best possible outcome given the marketing efforts that were implemented. Marketers can compare their actual results with the [X+1] POE-generated set of ideal results and even query specific suggestions about actions they can take to improve marketing efforts in areas of lagging performance.

To learn more about how to use open source R for statistical programming and analysis of large datasets, R-bloggers is a terrific resource. For production grade R statistical analytics, Revolution R Enterprise is an optimum solution. You can learn more about this solution at the Revolution Analytics website.

Related Topics

Related Publications

Related Content

Site Footer


This website is provided by John Wiley & Sons Limited, The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ (Company No: 00641132, VAT No: 376766987)

Published features on are checked for statistical accuracy by a panel from the European Network for Business and Industrial Statistics (ENBIS)   to whom Wiley and express their gratitude. This panel are: Ron Kenett, David Steinberg, Shirley Coleman, Irena Ograjenšek, Fabrizio Ruggeri, Rainer Göb, Philippe Castagliola, Xavier Tort-Martorell, Bart De Ketelaere, Antonio Pievatolo, Martina Vandebroek, Lance Mitchell, Gilbert Saporta, Helmut Waldl and Stelios Psarakis.