Skip to main content

Table of Contents

Testing the Action Scoring component

This section provides an example of how to test a sub-strategy with the NBA Strategy Framework. A similar technique can be used for strategies other than Action Scoring.

  1. In the Pega Customer Decision Hub portal, click Intelligence Strategies .

  2. Open the NBA Strategy Framework strategy.

  3. From the NBA Strategy Framework strategy, right-click the Action Scoring component, and then click Open strategy to open the Action Level Propensity strategy.

    The Action Level Propensity strategy will test expanding actions with active and eligible channels, in addition to running the Predict Action Propensity prediction to determine the action level propensity.
  4. Open the Test Run panel by clicking the slider button on the upper right side of the strategy canvas.

  5. In the Settings section, select Data transform, and then select a persona transform.

    To make any changes to the persona for the test, follow the steps used when setting up the person for the trigger strategy tests.
  6. In the For external inputs use strategy field, enter ExternalInputNBAFrameworkTest.

  7. Select Specify a single component within the strategy.

  8. Select the component that immediately precedes the component you are testing.

    For the Action Level Propensity strategy, since you are testing the Action Scoring component, select the Constraints component, as in the following figure:
  9. Click Run.

  10. Click on individual components within the strategy to check where actions are dropping out or where the strategy results are unexpected.

Did you find this content helpful?

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.

Pega Community has detected you are using a browser which may prevent you from experiencing the site as intended. To improve your experience, please update your browser.

Close Deprecation Notice
Contact us