Skip to main content

Challenge

Testing engagement policy conditions using an audience simulation

4 Tasks

15 mins

Visible to all users
Beginner Pega Customer Decision Hub 8.6 English
This content has been updated. Click here to continue your progress on the latest version.

Scenario

U+ Bank is currently doing cross-sell on the web by showing various credit cards to its customers.

They want to run an audience simulation to check how many potential actions are filtered out by each component of the policy and discover if a particular criterion is too broad or too narrow for their requirements.

Use the following credentials to log in to the exercise system:

Role

User name

Password

Decisioning Analyst

CDHAnalyst

rules

Your assignment consists of the following tasks:

Task 1: Check the configurations

Check the Next Best Action Designer configurations that U+ Bank currently uses to present customers with relevant offers.

Task 2: Prepare data set for simulation run

Run the PrepareSampledCustomers_REF data flow to prepare the data set for the simulation run.

Note: The Sampled Customers is not available in a persisted store. To initialize the customer data, first run the PrepareSampledCustomers_REF data flow.
If you already initiated the customer data as part of the previous challenge, you need not perform Task2.

Task 3: Create and execute an engagement policy audience simulation

Create and execute an audience simulation. Use the Sampled Customers data set as the audience and only engagement policy as the simulation scope. Inspect the result of the executed simulation by answering the following questions.

  • How many customers pass through the eligibility, applicability, and suitability conditions at the group-level and the action-level?
  • How many unique customers qualify for each action?

Task 4: Create and execute an arbitration audience simulation

Create and execute an audience simulation. Use the Sampled Customers data set as the audience and engagement policy and arbitration as the simulation scope. Inspect the result of the executed simulation by answering the following question.

  • How many unique customers qualify for each action?

Challenge Walkthrough

You must initiate your own Pega instance to complete this Challenge.

Initialization may take up to 5 minutes so please be patient.

Detailed Tasks

1 Check the configurations

  1. Log in as Decisioning Analyst with username CDHAnalyst and password rules.
  2. In the navigation pane of Pega Customer Decision Hub, click Next-Best-Action > Designer.
    NBA Designer
  3. Navigate through the tabs to see the configurations made to present customers with relevant offers.
    NBA designer tabs

2 Prepare data set for simulation run

  1. In the navigation pane of Pega Customer Decision Hub, click Data > Data Flows.
    Data flows
  2. Search for and then open the PrepareSampledCustomers_REF data flow to prepare the data set used for simulations. This data is based on a Monte Carlo dataset, which is generated.
    Note: The Monte Carlo data set generates a mock data set. As a result, different simulation runs have different results.
  1. Click Actions > Run.
    Run a data flow
  2. Click Submit.
  3. Click Start.
    Note: Notice that the prospect data is populated once the test run is complete.
  1. Close the data flow Test run window.

3 Create and execute an engagement policy audience simulation

  1. In Next Best Action Designer, click the Engagement policy tab.
  2. In the Business structure section, under Sales, click CreditCards.
    Credit card group
  3. Click Actions > Audience Simulations.
    Audience simulation menu
  4. Click Create simulation to create a new audience simulation.
    Create an audience simulation
  5. In the Create simulation dialog box, configure the following information.
    1. Audience: SampledCustomers
    2. Next-Best-Action scope: Engagement policies only
    3. Name: AudienceSimulation_E
      Audience simulation creation window
  6. Click Run to execute the simulation.
    Question: How many customers pass through the eligibility, applicability, and suitability conditions at the group-level and the action-level?
  7. Inspect how the customers are filtered at Next Best Action Designer level for every engagement policy condition.
    NBA level audience simulation results
    Tip: 993 customers pass through the eligibility conditions at group and action levels. As the applicability condition does not filter out customers, 993 passes through the group and action levels. There are no group-level suitability conditions defined and hence 719 customers pass through the action-level suitability conditions.
    Question: How many unique customers qualify for each action?
    Per action audience simulation results_arbitration
    Tip: When a simulation is run with only engagement policy conditions as the scope, the number of customers who qualify for an action are not unique. All customers who qualify for an action are counted. As a result, the total of customers per action does not tally with the Total unique customers count shown in the Next-Best-Action Desinger.
    Note: The Monte Carlo data set generates a mock data set. As a result, different simulation runs have different results.
  8. In the Offers section, open a credit card action (for example, click the Premier Rewards card).
    Open an action
  9. Click the Engagement policy tab.
  10. Click Actions > Audience simulation.
    Action level audience simulation menu
  11. In the Simulation list, select AudienceSimulation_E (S-#####).
    Select a simulation run
  12. Inspect how the customers are filtered at action level for every engagement policy condition.
    Action level eligibility results
    Action level applicbility results
    Action level suitability results

4 Create and execute an arbitration audience simulation

  1. In Next Best Action Designer, click the Engagement policy tab.
  2. Click CreditCards.
  3. Click Actions > Audience Simulations.
  4. Click Create simulation to create a new audience simulation.
  5. In the Create simulation dialog box, configure the following information.
    1. Audience: SampledCustomers
    2. Next-Best-Action scope: Engagement policies and arbitration
    3. Name: AudienceSimulation_A
      Audience simulation with arbitration
  6. Click Run to execute the simulation.
  7. Inspect how the customers are filtered at the Next Best Action Designer level for every engagement policy condition.
    NBA level audience simulation results
    Question: How many unique customers qualify for each action?
    Per action audience simulation results_arbitration
    Tip: When an audience simulation is run with engagement policy and arbitration as the scope, the number of customers who qualify for an action are unique, as only the top offer is considered. As a result, the total number of unique qualifying customers at the next-best-action level adds up to the total number of unique qualifying customers per action.
  8. In the Offers section, open a credit card action (for example, click the Premier Rewards card).
    Open an action
  9. Click the Engagement policy tab.
  10. Click Actions > Audience simulation.
  11. In the Simulation list, select AudienceSimulation_A (S-#####).
    Open an audience simulation run
     
  12. Inspect how the customers are filtered at action level for every engagement policy condition.
    Action level eligibility results
    Action level applicbility results
    Action level suitability results
    Note: The Monte Carlo data set generates a mock data set. Therefore, different simulation runs will have different results.
  13. On the right, in the Show population that passed as list, select Percentages to see the filtration effect in percentages.
    Simulation results in


Available in the following mission:

If you are having problems with your training, please review the Pega Academy Support FAQs.

Did you find this content helpful?

Want to help us improve this content?

We'd prefer it if you saw us at our best.

Pega Academy 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