Mastering the Pega CDH Decisioning Loop: A CLI Tool for Deeper Understanding
- Feb 16
- 2 min read
If you've spent any time working with Pega Customer Decision Hub (CDH), you know the power of its always-on brain. The "decisioning loop"—Evaluate, Present, Capture, Learn, Repeat—is the engine that drives personalized customer experiences. But truly grasping how each stage impacts the next, especially when new to the platform, can be a challenge.
Explaining the "intricate dance" of eligibility rules, propensity models, and action prioritization is often easier shown than told. That's why we built something practical for our own team, and now we're opening it up to all interested: a command-line interface (CLI) training tool that lets you run the CDH decisioning loop, one transparent command at a time.
Why a CLI Tool?
For data engineers, systems architects, and developers diving into CDH projects, the terminal is often a more natural environment for understanding logic and process flows. While the Pega UI provides an excellent holistic view, sometimes you need to strip away the abstractions and see the raw mechanics at play.
This CLI tool allows you to:
Run NBA for a customer: See exactly what actions are recommended and, crucially, why.
Present and Capture Responses: Simulate customer interaction and record the outcome (e.g., accepted, rejected).
Retrain the Model: Watch how the propensity model updates based on new customer feedback.
Observe Score Shifts: Run NBA again to immediately see how the decisioning changes with the newly trained model.
Transparency in Every Decision
One of the core design principles of this tool is absolute transparency. As you execute commands, you'll get clear feedback on:
Which actions were filtered out by eligibility rules.
Which actions were suppressed by history or other constraints.
Precisely how remaining actions were scored and ranked.
This level of detail is useful for debugging, understanding complex strategies, and building confidence in the decisioning output. It bridges the gap between theoretical understanding and practical application.
Designed for Technical Users
You don't need to pore over Python code (though it's there if you're curious!). The learning happens directly at the command line. This tool is specifically aimed at technical professionals who are comfortable in a terminal environment and want to quickly build a robust mental model of how CDH operates under the hood.
Get Started
We've found this tool incredibly useful internally for onboarding and deep-diving into CDH mechanics. We're excited to share it with the broader community.
Ready to take the CDH decisioning loop for a spin?
