In addition to customer journeys providing a good mental model for managing content, they influence the next-best-action decisions made for your customers. You can configure the parametrized predictors in the adaptive models that affect the next-best-action decisions. You can also indicate which actions should be filtered out when the customer is in a specific journey stage and control when particular action gets presented as the top action.
Adding parametrized predictors to adaptive models
The first approach to influencing a next best action is adding parametrized predictors to adaptive models, which will pass the current journey and stage as well as the previous stage as a parameter for next-best-action strategy framework. For new implementations of Pega Customer Decision Hub, adaptive models used by the next-best-action strategy framework already have the parametrized predictors added and mapped. If you are upgrading your Pega Customer Decision Hub from a version older than 8.6, you must manually add the parameters to the adaptive models, as well as the ModelImpl strategy that calls the models. For more information, see Adding predictors to existing adaptive models.Add the following predictors to your adaptive models:
- Journey – represents the journey that the customer is currently in.
- JourneyStage – represents the stage that the customer is currently in.
- LastJourneyStage – represents the last stage of the journey that the customer has been in. This can be the current stage.
- DaysinCurrentStage – represents how many days have elapsed since the customer first entered the current stage.
- PriorStageInJourney – represents the stage prior to the one that the customer is currently in. If there were no stages before the current stage, the property is blank.
By default, actions that are associated with customer journey stages can still compete in Arbitration, even if the customer is not in the current journey stage. When the customer is not in the current journey stage, the action is not tagged with the journey metadata. If you do not want an action to be eligible for arbitration unless the customer is in the journey stage that the action is associated with, filter out the action by configuring an action-specific engagement policy. For more information, see Filtering out actions not associated with the current customer journey stage.
You can influence the next best action by choosing when a specific action is presented as the top action. By using upweighting, you can have more control and more confidence in the decision, about which action or set of actions will be presented as the top action for the particular customer journey. For more information, see Upweighting customer journey actions.