I85PREP
⌘ K
Pega Marketing Consultant study materials
Pega Marketing Consultant study resources

Pega Marketing Consultant practice questions

One question at a time. Build your Pega Marketing Consultant understanding.

First 50 of 200 questions. Each one lists the choices, the correct answer, and an explanation. The checker at the top of this page saves progress in this browser when it is available.

Continue in the practice tool above to load more questions, 50 at a time.

Question 1Next-Best-Action and decision management design

In Pega Customer Decision Hub, what is the primary purpose of a Next-Best-Action Designer engagement policy?

  1. To define the business constraints and priorities that shape which actions are eligible for a customer Correct answer.
  2. To map customer attributes from the source system into the customer profile
  3. To configure the database connection used by the decision engine
  4. To schedule when the batch arbitration job runs each night

Explanation

Engagement policies capture business rules such as outbound suppression, contact limits, and prioritization that the decision engine applies when selecting actions. They are business-facing configuration, not infrastructure or data-mapping settings, and they govern eligibility and ranking rather than job scheduling.

Question 2Next-Best-Action and decision management design

A consultant needs the decision engine to filter out any action whose eligibility condition evaluates to false before arbitration runs. Which configuration point should hold that condition?

  1. The customer profile schema
  2. The arbitration formula weighting
  3. The action's treatment template
  4. The eligibility condition on the action or its group Correct answer.

Explanation

Eligibility conditions are evaluated during the decision process to determine whether an action or group can be considered at all. Failing eligibility removes the action before arbitration, so it never competes for a position.

Question 3Next-Best-Action and decision management design

Which outcome does the arbitration stage of a Next-Best-Action decision produce?

  1. A normalized customer profile snapshot
  2. A data flow that ingests interaction history
  3. A list of contact policy violations for audit
  4. An ordered set of eligible actions that fills the available placement slots Correct answer.

Explanation

Arbitration ranks eligible actions using the configured prioritization formula and assigns them to available placement slots. The result is the ranked, slot-assigned set returned to the channel.

Question 4Next-Best-Action and decision management design

What distinguishes an action from a treatment in Pega Customer Decision Hub?

  1. An action is the business offer being decided, while a treatment is the channel-specific content used to present it Correct answer.
  2. An action applies only to inbound channels, while a treatment applies only to outbound channels
  3. An action is the rendered email, while a treatment is the database record of the offer
  4. An action is a reporting artifact, while a treatment is an execution artifact

Explanation

An action represents the business proposition, such as an upsell offer, and a treatment supplies the channel-specific presentation of that proposition. One action can carry multiple treatments so the same offer renders correctly in email, web, or mobile.

Question 5Next-Best-Action and decision management design

In a Next-Best-Action Designer configuration, what does a customer journey define?

  1. The list of users authorized to edit strategies
  2. The nightly batch window for model recalculation
  3. The sequence of stages and steps a customer progresses through for a business objective Correct answer.
  4. The set of database tables that store interaction history

Explanation

Journeys model the stages and steps a customer moves through toward an objective, and actions are mapped to journey stages so the right offer appears at the right point. They organize actions by customer context rather than by infrastructure or permissions.

Question 6Next-Best-Action and decision management design

Which artifact in the decision framework evaluates each candidate action against a propensity model and a business value to produce a score?

  1. The treatment rule
  2. The segment definition
  3. The data flow
  4. The arbitration formula Correct answer.

Explanation

The arbitration formula combines propensity, business value, and weighting factors into the score used to rank actions. It is the ranking logic applied after eligibility filtering and before slot assignment.

Question 7Next-Best-Action and decision management design

A consultant must ensure a customer never receives more than two outbound offers in a seven-day period across all channels. Which capability enforces this?

  1. Placement capacity settings
  2. Propensity thresholds on each action
  3. Model transparency reports
  4. Contact policy rules in the engagement policy Correct answer.

Explanation

Contact policy rules define limits such as maximum contacts per period and are applied during eligibility to suppress actions that would exceed the limit. They operate across channels because they are evaluated in the shared decision process.

Question 8Next-Best-Action and decision management design

What is the role of a predictive propensity model in Next-Best-Action decisioning?

  1. It validates that treatments render correctly on mobile devices
  2. It stores the final ranked output for each customer
  3. It defines which channels are available for a placement
  4. It estimates the likelihood that a customer will accept a specific action Correct answer.

Explanation

A propensity model predicts the probability of a positive response for a customer-action pair, and that prediction feeds the arbitration score. It is an input to ranking rather than a storage or rendering component.

Question 9Next-Best-Action and decision management design

A bank wants its decision strategy to prefer a retention offer over a cross-sell offer whenever the customer's churn risk is high, regardless of predicted propensity. Which mechanism best expresses this business priority?

  1. A lower propensity threshold on the cross-sell action
  2. A new data flow that recalculates churn risk hourly
  3. A higher business value or weighting factor applied to the retention action in the arbitration formula Correct answer.
  4. An additional treatment on the cross-sell action

Explanation

Business priorities are expressed by weighting actions in the arbitration formula, so a strategic action can outrank others even when raw propensity differs. This keeps the business intent explicit and auditable in one place.

Question 10Next-Best-Action and decision management design

During testing, a consultant finds that an eligible action never appears in the web placement despite a high propensity score. Which check should be performed first?

  1. Verify that the batch job runs on the application server rather than the database server
  2. Verify the action is mapped to the web placement and that its group is not suppressed by engagement policy Correct answer.
  3. Verify that the customer profile schema includes every possible attribute
  4. Verify that the propensity model has been retrained within the last 24 hours

Explanation

An action appears in a placement only when it is associated with that placement and survives eligibility, suppression, and contact policy checks. Confirming placement mapping and policy suppression isolates the most common causes before investigating modeling.

Question 11Next-Best-Action and decision management design

Which statement best describes how a group-level eligibility condition differs from an action-level eligibility condition?

  1. A group condition is evaluated after arbitration, while an action condition is evaluated before
  2. A group condition gates every action in the group, while an action condition gates only that action Correct answer.
  3. A group condition applies only to outbound channels, while an action condition applies only to inbound channels
  4. A group condition can only reference customer attributes, while an action condition can only reference action properties

Explanation

Group-level eligibility acts as a gate for all member actions, so failing it removes the whole group from consideration. Action-level eligibility is narrower and removes only the individual action, which allows fine-grained control within a group.

Question 12Next-Best-Action and decision management design

A marketing team wants actions to be selected based on the customer's current journey stage. Where should this logic be configured?

  1. By creating a separate decision data flow per journey stage
  2. By setting the placement capacity to the number of journey stages
  3. By adding the journey stage to the treatment's HTML template
  4. By mapping actions to journey stages and steps so only matching actions are eligible Correct answer.

Explanation

Mapping actions to journey stages and steps lets the decision engine consider only actions relevant to the customer's current stage. This keeps journey logic declarative and visible in the Next-Best-Action Designer configuration.

Question 13Next-Best-Action and decision management design

Which pairing correctly matches a Next-Best-Action Designer component with its function?

  1. Placement: stores the predictive model
  2. Engagement policy: defines suppression and contact limits Correct answer.
  3. Treatment: ranks eligible actions by score
  4. Arbitration: ingests and transforms source data

Explanation

Engagement policy is the component that expresses suppression rules and contact limits applied during eligibility. The other pairings mismatch components with functions that belong elsewhere in the architecture.

Question 14Next-Best-Action and decision management design

A consultant needs to limit which actions a specific customer segment can receive without changing the global strategy. Which approach is most appropriate?

  1. Retrain the propensity model using only that segment's data
  2. Create a segment-specific engagement policy or eligibility rule scoped to that segment Correct answer.
  3. Reduce the placement capacity for all channels
  4. Remove treatments from actions not intended for the segment

Explanation

Scoping eligibility or policy rules to a segment restricts actions for that audience while leaving the global strategy intact. This preserves reuse and avoids side effects on other segments.

Question 15Next-Best-Action and decision management design

What is the effect of a suppression rule configured in the engagement policy?

  1. It lowers the propensity score of matching actions
  2. It removes matching actions from consideration before arbitration Correct answer.
  3. It logs matching actions for later manual review
  4. It adds extra placement slots for matching actions

Explanation

Suppression rules exclude actions from the candidate set during eligibility, so they never reach arbitration. This is how business rules such as do-not-contact or recent-complaint suppression are enforced.

Question 16Next-Best-Action and decision management design

A client wants the same offer to appear with different creative content on the web and in email. What is the recommended configuration?

  1. Define one action with two treatments, one for each channel Correct answer.
  2. Define one treatment and edit its content at runtime per channel
  3. Define two placements and assign the same treatment to both
  4. Define two separate actions with identical properties

Explanation

A single action can carry multiple channel-specific treatments, keeping the business offer unified while allowing tailored presentation. This avoids duplicating the offer and keeps reporting consolidated.

Question 17Next-Best-Action and decision management design

Which data requirement must be satisfied before a propensity model can produce reliable scores for an action?

  1. The placement must be configured for all channels
  2. The action must have at least three treatments defined
  3. The model must be trained on historical outcomes linked to that action and relevant customer attributes Correct answer.
  4. The engagement policy must contain no suppression rules

Explanation

Propensity models learn from historical outcome data associated with actions and customer attributes, so training data quality and relevance determine score reliability. Treatments, placements, and suppression rules affect presentation and eligibility but not model training.

Question 18Next-Best-Action and decision management design

A consultant observes that the highest-propensity action is consistently selected, crowding out strategically important offers. Which adjustment addresses this while preserving propensity modeling?

  1. Disable the propensity model for all actions
  2. Apply weighting factors or business value adjustments in the arbitration formula Correct answer.
  3. Remove eligibility conditions from strategic actions
  4. Increase the placement capacity until all actions appear

Explanation

Weighting factors and business values let strategic priorities influence ranking alongside propensity, achieving balance without discarding predictive insight. Disabling models or expanding capacity changes behavior in ways that do not reflect business intent.

Question 19Next-Best-Action and decision management design

How does the decision engine use the customer profile during Next-Best-Action processing?

  1. It schedules when models are retrained
  2. It supplies the customer attributes that eligibility conditions and models reference at decision time Correct answer.
  3. It stores the final ranked actions returned to the channel
  4. It defines the arbitration formula used for ranking

Explanation

The customer profile provides the attribute values that conditions, segments, and models evaluate when the decision runs. It is an input to the decision rather than a store for results or a definition of ranking logic.

Question 20Next-Best-Action and decision management design

A consultant must explain why an action was excluded from a customer's decision result. Which combination of factors should be examined?

  1. Only the treatment content associated with the action
  2. Only the customer's segment membership
  3. Eligibility conditions, suppression and contact policy rules, and placement mapping Correct answer.
  4. Only the propensity score of the action

Explanation

An action is excluded when it fails eligibility, is suppressed by policy or contact limits, or is not mapped to the placement being filled. Reviewing all three areas gives a complete picture of exclusion causes.

Question 21Next-Best-Action and decision management design

Which statement accurately describes the relationship between placements and arbitration output?

  1. Each placement defines available slots that arbitration fills with ranked actions Correct answer.
  2. Each placement defines a separate arbitration formula
  3. Each placement can only ever hold a single action
  4. Each placement bypasses eligibility to reduce latency

Explanation

Placements represent the available presentation slots in a channel, and arbitration assigns ranked actions to those slots. The number of slots therefore caps how many actions can be returned for that placement.

Question 22Next-Best-Action and decision management design

A consultant is asked to design a strategy where customers who recently complained receive no marketing actions for 30 days, but still receive service-related messages. Which design achieves this?

  1. A global suppression rule that blocks all actions for 30 days after any complaint
  2. A propensity threshold applied to all actions equally
  3. A placement capacity reduction for all channels
  4. A suppression rule scoped to marketing actions that checks the complaint date, leaving service actions unaffected Correct answer.

Explanation

Scoping the suppression rule to marketing actions and keying it to the complaint date blocks only the intended category while allowing service messages. A global rule would over-suppress and harm service communication.

Question 23Next-Best-Action and decision management design

Two actions in the same group have overlapping eligibility conditions, but business rules require that only one of them can ever be offered in a single decision. Which mechanism enforces mutual exclusivity?

  1. Assigning both actions to the same arbitration exclusivity group so only the top-ranked one is selected Correct answer.
  2. Adding a second treatment to each action
  3. Giving both actions identical propensity scores
  4. Placing both actions in different placements

Explanation

Exclusivity groups ensure that among competing actions only the highest-ranked one is awarded a slot, preventing conflicting offers in the same decision. Identical scores or separate placements do not guarantee mutual exclusivity.

Question 24Next-Best-Action and decision management design

A model's performance has degraded, and the business wants to continue using it while limiting exposure of low-confidence predictions. Which approach is most defensible?

  1. Remove the model and rank actions only by business value
  2. Apply a propensity threshold or adaptive model control that suppresses actions when predicted likelihood falls below an acceptable level Correct answer.
  3. Increase placement capacity so other actions compensate
  4. Delete historical outcome data to retrain from scratch immediately

Explanation

Thresholds and adaptive controls let the business cap exposure when model confidence is low while keeping the model in service. This balances risk management with continuity rather than discarding predictive capability outright.

Question 25Next-Best-Action and decision management design

An organization wants to run a controlled experiment comparing two arbitration weightings for the same action set. Which design supports valid comparison?

  1. Apply the new weighting to all customers and compare results to last year's performance
  2. Define two strategy variants with different weightings and randomly assign the customer population between them Correct answer.
  3. Let channel owners choose which weighting to apply per interaction
  4. Change the weighting and the eligibility rules at the same time for all customers

Explanation

Random assignment between variants isolates the effect of the weighting change from other factors and supports valid causal comparison. Comparing to prior periods or changing multiple variables confounds the results.

Question 26Next-Best-Action and decision management design

In Next-Best-Action Designer, which component defines the business hierarchies (such as Business Issue and Group) that organize the actions a customer can be considered for?

  1. The arbitration component
  2. The contact policy component
  3. The engagement policy component
  4. The taxonomy component Correct answer.

Explanation

The taxonomy is the structural component that organizes actions into Business Issues, Groups, and Treatments, giving the strategy its business hierarchy. Arbitration and engagement policy operate on the actions once the taxonomy structure exists, and contact policy governs contact limits rather than hierarchy.

Question 27Next-Best-Action and decision management design

A consultant is configuring an engagement policy. Which element determines whether an action is even considered for a specific customer based on customer data and context?

  1. The contact limit
  2. The arbitration weight
  3. The propensity threshold
  4. The applicability rule set Correct answer.

Explanation

Applicability (eligibility) rules in the engagement policy filter which actions are relevant to a given customer before arbitration. Contact limits constrain volume, arbitration weights influence ranking, and propensity thresholds are not a standalone filter at this stage.

Question 28Next-Best-Action and decision management design

What is the primary purpose of arbitration in Next-Best-Action Designer?

  1. To import customer data from external systems
  2. To rank eligible actions and select the best one for each placement Correct answer.
  3. To schedule when campaigns are executed
  4. To define which offers exist in the catalog

Explanation

Arbitration evaluates the eligible action set and chooses the highest-value action per placement using weights and scores. It does not create offers, import data, or schedule campaign runs.

Question 29Next-Best-Action and decision management design

In a decision strategy, which shape is used to branch logic based on whether a condition such as a customer property is true or false?

  1. The Map shape
  2. The Group By shape
  3. The If shape Correct answer.
  4. The Filter shape

Explanation

The If shape evaluates a condition and routes the strategy down a true or false path. Filter removes records matching a condition, Group By aggregates, and Map transforms values.

Question 30Next-Best-Action and decision management design

Which property category is used to store the numeric score produced by a predictive model so that arbitration can use it?

  1. The context property
  2. The parameter property
  3. The propensity property Correct answer.
  4. The symbolic property

Explanation

A propensity property holds the model output score that arbitration consumes as the likelihood component. Parameters are inputs, context properties describe the interaction, and symbolic properties hold text values.

Question 31Next-Best-Action and decision management design

A consultant wants to define the customer-facing channels (such as Web and Email) where actions can be presented. Which concept in Next-Best-Action Designer represents these presentation surfaces?

  1. Segments
  2. Issues
  3. Placements Correct answer.
  4. Treatments

Explanation

Placements represent the channel surfaces, such as a web banner or email slot, where arbitration output is presented. Treatments are the creative variants, segments group customers, and issues are taxonomy categories.

Question 32Next-Best-Action and decision management design

Which statement best describes the function of a treatment in Next-Best-Action Designer?

  1. It sets the maximum number of contacts per customer.
  2. It defines the eligibility rules for a group of actions.
  3. It stores the customer's interaction history.
  4. It provides the channel-specific creative content for an action. Correct answer.

Explanation

A treatment supplies the actual content and channel-specific presentation for an action once it is selected. Eligibility is handled by engagement policy, history is stored in interaction data, and contact limits are set in contact policy.

Question 33Next-Best-Action and decision management design

What does the term "action" represent in Next-Best-Action Designer?

  1. A single execution of a campaign schedule
  2. A business offer or message that can be presented to a customer Correct answer.
  3. A model that predicts churn
  4. A database table that stores customer records

Explanation

An action is the business offer or message that the decision engine can select and present, and it belongs to a group in the taxonomy. It is not a schedule, a table, or a model.

Question 34Next-Best-Action and decision management design

A consultant needs to ensure that a specific action is only offered to customers whose loyalty tier is Gold. Where should this condition be configured so it filters the action before arbitration?

  1. As a weight in the arbitration shape
  2. As a treatment-level content rule
  3. As a contact limit in the contact policy
  4. As an applicability rule on the action in the engagement policy Correct answer.

Explanation

Applicability rules in the engagement policy determine which customers an action is relevant for, filtering it before arbitration. Treatment content, contact limits, and weights do not filter by customer attribute.

Question 35Next-Best-Action and decision management design

A marketing team wants to prioritize actions that generate higher expected revenue. Which arbitration configuration directly supports this?

  1. Increasing the contact limit
  2. Setting the propensity weight to zero
  3. Assigning higher business value weights to revenue-generating actions Correct answer.
  4. Adding more treatments to each action

Explanation

Arbitration combines propensity and business value using weights, so raising the value weight for revenue actions increases their priority. Zeroing propensity would ignore likelihood, contact limits control volume, and more treatments do not change ranking.

Question 36Next-Best-Action and decision management design

A consultant must ensure that customers who opted out of marketing receive no promotional actions. Which mechanism is most appropriate?

  1. An additional treatment
  2. A higher arbitration weight
  3. A suppression rule in the contact policy Correct answer.
  4. A new placement

Explanation

Suppression rules in the contact policy prevent specified actions or groups from being presented to customers matching a condition such as an opt-out. Weights, treatments, and placements do not enforce opt-out suppression.

Question 37Next-Best-Action and decision management design

A consultant is building a decision strategy and needs to select the single best action for a placement from a set of scored candidates. Which shape accomplishes this?

  1. The Group By shape
  2. The Aggregate shape
  3. The Set Property shape
  4. The Prioritize shape Correct answer.

Explanation

The Prioritize shape orders candidates by a score and can keep the top result, selecting the best action for a placement. Group By and Aggregate summarize, and Set Property assigns values.

Question 38Next-Best-Action and decision management design

A client wants to test whether a new engagement policy improves acceptance rates without disrupting the live strategy for all customers. Which approach fits Next-Best-Action Designer?

  1. Disable arbitration entirely
  2. Permanently replace the existing engagement policy
  3. Remove all treatments from the actions
  4. Use a control or test group configuration to compare outcomes Correct answer.

Explanation

Configuring a test or control group lets a subset receive the alternate treatment while the rest continue normally, enabling a controlled comparison. Permanently replacing the policy, disabling arbitration, or removing treatments would disrupt the live strategy and prevent valid comparison.

Question 39Next-Best-Action and decision management design

A consultant observes that an action is eligible but never selected. Which factor is most likely influencing arbitration ranking?

  1. The placement is misconfigured
  2. The action's combined propensity and value weight is lower than competitors Correct answer.
  3. The action has no treatments assigned
  4. The customer segment is too large

Explanation

When an action is eligible, arbitration ranks it against competitors using weighted propensity and value, so a lower combined score causes it to lose. Missing treatments affect presentation, placement misconfiguration would block all actions, and segment size does not determine ranking.

Question 40Next-Best-Action and decision management design

Which data set is typically used to store the customer profile that the decision engine reads during Next-Best-Action processing?

  1. The interaction history data set
  2. The customer data set Correct answer.
  3. The model snapshot data set
  4. The campaign schedule data set

Explanation

The customer data set holds profile attributes that strategies reference during decisioning. Interaction history records past contacts, model snapshots store model metadata, and campaign schedules govern execution timing.

Question 41Next-Best-Action and decision management design

A consultant wants to record which action was presented and whether the customer accepted it, so that future models can learn from outcomes. Which concept captures this?

  1. The placement definition
  2. The interaction history Correct answer.
  3. The arbitration weight
  4. The taxonomy hierarchy

Explanation

Interaction history captures the outcome of each decision, including what was presented and the response, which feeds model learning. Taxonomy, weights, and placements are configuration elements rather than outcome records.

Question 42Next-Best-Action and decision management design

A consultant must ensure that a group of related actions shares common eligibility conditions while individual actions add their own conditions. What is the recommended configuration?

  1. Rely on arbitration weights to enforce eligibility
  2. Put all conditions on each individual action
  3. Configure group-level applicability rules and add action-level rules as needed Correct answer.
  4. Use only contact limits to control eligibility

Explanation

Group-level applicability rules apply to all actions in the group, and action-level rules add specifics, avoiding duplication. Putting all conditions on each action duplicates logic, while contact limits and weights do not enforce eligibility.

Question 43Next-Best-Action and decision management design

Which statement best describes how the decision engine uses the customer profile during Next-Best-Action processing?

  1. It disables contact policy when a profile is present
  2. It evaluates applicability rules and strategy logic against profile attributes to determine eligible actions Correct answer.
  3. It permanently stores the profile in the arbitration output
  4. It replaces the taxonomy hierarchy with profile values

Explanation

Profile attributes are referenced by applicability rules and strategy logic to determine which actions are relevant and how they are scored. The profile does not replace the taxonomy, is not stored in arbitration output, and does not disable contact policy.

Question 44Next-Best-Action and decision management design

A consultant is asked to explain why a specific action was excluded for a customer. Which combination should be examined?

  1. Only the treatment content
  2. Applicability rules, contact policy limits, and arbitration ranking Correct answer.
  3. Only the campaign schedule
  4. Only the placement configuration

Explanation

Exclusion can result from failing applicability rules, being blocked by contact policy, or losing in arbitration ranking. Treatment content, schedules, and placements alone do not explain exclusion of a specific action.

Question 45Next-Best-Action and decision management design

A consultant needs to design logic where customers who submitted a complaint in the last 30 days receive no marketing actions. Which configuration achieves this?

  1. A suppression rule in the contact policy keyed on the complaint date Correct answer.
  2. An additional treatment for complainers
  3. A new business issue in the taxonomy
  4. A higher arbitration weight for marketing actions

Explanation

A suppression rule keyed on the complaint date blocks marketing actions for the specified window. Weights would increase priority, treatments add content, and taxonomy changes do not enforce time-based suppression.

Question 46Next-Best-Action and decision management design

Two actions in the same group have overlapping applicability conditions, and business rules require that only one be selected. Which approach best enforces this?

  1. Increase the contact limit
  2. Assign identical weights to both actions
  3. Remove all treatments from one action
  4. Use arbitration to rank them and select the top one per placement Correct answer.

Explanation

When both actions are eligible, arbitration ranks them and selects the top action per placement, satisfying the single-selection requirement. Identical weights would not guarantee a single selection, removing treatments breaks presentation, and contact limits do not enforce mutual exclusivity.

Question 47Next-Best-Action and decision management design

A model's performance has degraded, and the business wants to keep using it while limiting exposure of low-confidence predictions. Which configuration is most appropriate?

  1. Disable interaction history capture
  2. Remove the model entirely from arbitration
  3. Increase the business value weight to maximum
  4. Apply a propensity threshold in the engagement policy to exclude low-confidence actions Correct answer.

Explanation

A propensity threshold filters out actions whose model scores fall below an acceptable confidence level, limiting exposure while keeping the model in use. Removing the model, maximizing value weight, or disabling history capture does not address low-confidence predictions.

Question 48Next-Best-Action and decision management design

A consultant must limit which actions a specific customer segment can receive without altering the global strategy for other segments. Which configuration is most appropriate?

  1. Modify the global arbitration weights for all customers
  2. Change the placement definitions globally
  3. Add treatments to every action
  4. Configure segment-specific applicability rules in the engagement policy Correct answer.

Explanation

Segment-specific applicability rules restrict actions for that segment while leaving the global strategy intact for others. Global weight changes, universal treatments, and global placement changes affect all customers.

Question 49Next-Best-Action and decision management design

A consultant wants to compare two arbitration weightings for the same action set in a controlled experiment. Which configuration best supports this?

  1. Assign a control group and a test group with different weight settings Correct answer.
  2. Remove arbitration for the test group
  3. Apply both weightings simultaneously to all customers
  4. Use identical weights for both groups

Explanation

A control group and test group with different weight settings isolate the effect of the weighting change on outcomes. Applying both simultaneously, removing arbitration, or using identical weights prevents a valid comparison.

Question 50Next-Best-Action and decision management design

A consultant defines a Next-Best-Action strategy in which the business wants an action to be offered only when the customer's current context (such as the page the customer is browsing or the balance of their account) satisfies a condition at the moment of the interaction. Which configuration element should the consultant use to express this real-time condition?

  1. A treatment rule that restricts which channel can be used
  2. A contact policy limit that caps the number of impressions per day
  3. A relevance condition defined on a customer property evaluated at run time Correct answer.
  4. A static segment rule that is refreshed nightly in the data set

Explanation

Relevance (applicability) conditions are evaluated by the decision engine at run time against the customer and interaction context, so an action becomes eligible only when the live condition is met. Segment membership and nightly refreshes reflect historical data and cannot react to the current context. Contact policies and treatment rules constrain volume and presentation, not the contextual eligibility of the action.