Skip to main content

Predictive Analytics Director

Suggest edit Updated on September 20, 2021

Pega Predictive Analytics Director (PAD) offers business-focused tools to rapidly develop models that accurately predict customer behaviors, such as offer acceptance, churn or credit risk. These powerful predictive models discover the hidden trends and patterns in your data, enabling opportunities and risks to be reliably evaluated.

You can use Decision Strategy Manager (DSM) to put predictive models at the heart of every business process. Making every prediction actionable increases automation and accuracy while optimizing the result of each customer interaction.

Unlike most tools that require experienced statisticians or data miners to develop models, PAD enables business users, including marketing or risk analysts, to quickly create high-quality predictive models. Developing models is simple and easy: While you define the objectives and judge the results, PAD takes care of analyzing and understanding how hundreds of attributes are related.

With PAD, you can:

  • Drive predictive intelligence into every business decision with advanced analytics functions that cover the end-to-end model development process from data preparation to model building and evaluation.
  • Develop more models in hours – not days or weeks – that predict any form of behavior to deliver the best actions across the customer lifecycle, including sales, service, retention, cross-selling, up-selling and risk.
  • Enhance operational efficiency, increase consistency and eliminate service representative “guesswork” by integrating predictions for multiple outcomes into automated, rules-driven business processes.

Key Features

Rapidly develop accurate predictive models

  • Visual, easy-to-use wizards enable analytical business users create powerful and accurate models for predicting customer behavior from batch customer data.
  • Automation streamlines complex scoring model development including automatic univariate analysis, predictor pre-processing, non-linear modeling, model building and evaluation.
  • Scoring for positive/negative behavior, extended scoring that infers the behavior of unknown cases and spectrum models that predict continuous behavior provide the highest levels of predictive power and reliability.
  • Multivariate scoring models include regression models, decision tree models and genetic programming models, offering multiple options for best managing accuracy versus simplicity.
  • Seamless integration with Pega Adaptive Decision Manager (ADM), makes predictive models “self-learning”,as ADM uses real-time data to instantly adapt predictive decisions during customer interactions.
  • Adherence to standards minimizes integration effort with third-party applications.

Drive better business outcomes

  • Batch execution can be used to accurately predict risk, calculate expected losses and ensure compliance with regulations.
  • Models can take into account multiple factors across the customer lifecycle to personalize the Next-Best-Action, such as past service interactions, purchases, payment history, likelihood of defection and spend objectives.
  • Integration with case management automatically determines the most efficient queuing and routing based on agent past performance and manages the sequence and timing of tasks.
  • Centralized predictive models ensure that best practices are always executed during customer interactions.
  • Automatic monitoring and real-time reporting on decisions made and outcomes enable predictive models to be continuously refined and improved.
Did you find this content helpful? YesNo

0% found this useful

Have a question? Get answers now.

Visit the Collaboration Center to ask questions, engage in discussions, share ideas, and help others.

Ready to crush complexity?

Experience the benefits of Pega Community when you log in.

We'd prefer it if you saw us at our best. is not optimized for Internet Explorer. For the optimal experience, please use:

Close Deprecation Notice
Contact us