MAINFRAME > Business Strategy > BI and Analytics

SPSS Modeler 15 Offers Real-Time Scoring for Faster, More Effective Insights

Illustration by James Bennett

According to the 2011 IBM Global CMO Study, the No. 1 priority for chief marketing officers is to enhance customer loyalty. For CEOs, the top priority is to get closer to the customers. And for CTOs, improving analytical skills across the company will be the largest investment area over the next five years.

Predictive analytics is the key to gathering and analyzing customer insights to understand behavior and take action to prevent churn or market the most relevant product to the right customer. Ultimately, knowing your customers better and engaging them throughout the customer lifecycle will lead to better results.

In the Ballpark: Approximately 70 percent of all data transactions in the banking, insurance, retail, telecom­­- munications, utilities and government industries originate on the System z platform.

With an increasing demand to better understand and service customers and citizens, organizations are under added pressure to optimize all areas driving customer satisfaction and loyalty, which can ultimately maximize performance and increase revenue. Companies can optimize customers’ lifetime value by creating tight-knit and trusting relationships with them through analytics.

Consumers are more likely to become repeat buyers and recommend a company’s products to their friends and family once trust has been established. Companies earn that trust by consistently delivering the types of results and experiences customers expect and through exceeding their expectations by anticipating needs in advance.


Scoring Big

To improve the accuracy of real-time decision making, exploit the point of interaction and build customer trust, many companies are investing in real-time scoring of their transactional data. With access to the latest information about consumers’ attitudes and opinions, it’s possible to predict the most desirable outcome for the business.

Approximately 70 percent of all data transactions in the banking, insurance, retail, telecommunications, utilities and government industries originate on the System z* platform, and many organizations now have a distinct opportunity to change the game with real-time predictive scoring.

Accurate and useful real-time scoring results depend greatly on complete, relevant and current data. The faster someone can use the latest and most relevant data in the decision-making process, the higher the quality of the decision. Because most relevant data—financial information, customer lists, personnel records and manufacturing reports—originates on the System z platform, it makes sense to bring predictive analytics to the data residing there.

The superior performance, security and scalability of System z can help organizations make the most of their predictive analytics solution. They can process and analyze the staggering volumes of ever-changing and increasingly varied data and base decisions on the newest and most relevant information—as it’s received.

For example, with the capability to understand real-time inventory, demand can be assessed and anticipated to ensure the right products are in stock and personnel is available. Or, by scoring transactional data in real-time at the point of impact, clients can reduce fraud in such applications as credit card transactions and insurance claims, or they can improve the up-sell and cross-sell of other high-volume customer marketing and sales initiatives.


Introducing IBM SPSS Modeler 15

IBM SPSS* Modeler helps organizations use predictive analytics to gain insight into their business and make improvements. With SPSS Modeler 15 for Linux* on System z, they can:

  • Integrate transactional applications running on IBM DB2* for z/OS* to improve the accuracy, speed and performance of real-time scoring, while reducing cost and complexity
  • Create models using both structured and unstructured data
  • Effectively manage the entire data-mining cycle from one workbench
  • Fine-tune selected models based on statistical confidence levels
  • Work with a variety of file formats pulled from (and written to) various databases for data preparation
  • Automate the execution of models and deploy the results into existing systems and processes

The combination of DB2 for z/OS and SPSS Modeler increases the timeliness of scoring because it offers a more efficient, reliable process and makes it available to more users. The qualities of service of IBM System z are designed to offer improved accuracy, speed and performance in scoring data while reducing the overall cost and complexity. It scores new and relevant data directly in online transaction processing (OLTP) applications, scales to larger data volumes to improve the accuracy of data models and patterns created, and provides the performance needed to meet or exceed the service-level agreements (SLAs) of the OLTP applications and departments.

IBM SPSS Modeler 15 for Linux on System z is an integral part of the IBM Predictive Analytics on System z portfolio. This combination can change how organizations weave analytics into the fabric of their business to fuel the processes, decisions and actions that optimize outcomes.

IBM Analytical Decision Management is designed to optimize and automate high-velocity decisions and consistently generate results. Using IBM business analytics technologies, it provides predictive models, business rules, optimization and scoring that help organizations weave analytics into the fabric of their businesses.

SPSS Statistics software can help companies solve business and research problems with ad hoc analysis, hypothesis testing and predictive analytics. Organizations use it to understand data, analyze trends, forecast, plan, validate assumptions and drive accurate conclusions.

SPSS Collaboration and Deployment Services for Linux on System z provides a unified platform for IBM SPSS Modeler and IBM SPSS Decision Management, helping provide a secure, reliable and scalable foundation for the integration of analytic results into existing enterprise infrastructures.


Rebecca Wormleighton is an IBM product marketing manager for Cognos software, focusing on synergy’s between Cognos business-intelligence software and IBM products. She can be reached at

comments powered by Disqus



2017 Solutions Edition

A Comprehensive Online Buyer's Guide to Solutions, Services and Education.

All Together Now

A Centralized Business Analytics Environment Delivers Greater Value

Unlocking the Secrets of Success

An introduction to business analytics

IBM Systems Magazine Subscribe Box Read Now Link Subscribe Now Link iPad App Google Play Store
Mainframe News Sign Up Today! Past News Letters