[Beta]{class="badge informative"}
Use ÃÛ¶¹ÊÓƵ Experience Platform data for Decisioning aep-data
Journey Optimizer allows you to leverage data from ÃÛ¶¹ÊÓƵ Experience Platform for Decisioning. This allows you to extend the definition of your decision attributes to additional data in datasets for bulk updates that change periodically without having to manually update the attributes one at a time. For example, availability, wait times, etc.
Beta restrictions and guidelines guidelines
Before you begin, take note of the following restrictions and guidelines:
- A decision policy can reference up to 3 datasets total, across all its decision rules and ranking formulas combined. For example, if your rules use 2 datasets, your formulas can only use 1 additional dataset.
- A decision rule can use 3 datasets.
- A ranking formula can use 3 datasets.
- When a decision policy is evaluated, the system will perform up to 1000 dataset queries (lookups) in total. Each dataset mapping used by a decision item counts as one query. Example: If a decision item uses 2 datasets, evaluating that offer counts as 2 queries toward the 1000-query limit.
Enable a dataset for data lookup enable
To use data from an ÃÛ¶¹ÊÓƵ Experience Platform dataset for decisioning, you must first enable it for lookup via an API call. For detailed instructions, refer to this section: Leverage ÃÛ¶¹ÊÓƵ Experience Platform datasets in Journey Optimizer.
Leverage ÃÛ¶¹ÊÓƵ Experience Platform data leverage-aep-data
Once a dataset is enabled for lookup, you can use its attributes to enrich your decision logic with external data. This is especially useful for attributes that frequently change, such as product availability, or real-time pricing.
Attributes from ÃÛ¶¹ÊÓƵ Experience Platform datasets can be used in two parts of decision logic:
- Decision rules: Define whether a decision item is eligible to be shown.
- Ranking formulas: Prioritize decision items based on external data.
The next sections explain how to use ÃÛ¶¹ÊÓƵ Experience Platform data in both contexts.
Decision rules rules
Using ÃÛ¶¹ÊÓƵ Experience Platform data in decision rules allows you to define eligibility criteria based on dynamic, external attributes, ensuring decision items are only shown when relevant.
For example, let’s say an online retailer wants to promote product recommendations based on local store inventory. A product should only be eligible for recommendation if it is in stock at the nearest location. A dataset containing daily inventory updates is uploaded to ÃÛ¶¹ÊÓƵ Experience Platform. The rule logic checks if the inventory_count for a given product is greater than 0 for the customer’s preferred store. If so, the decision item is eligible.
To use ÃÛ¶¹ÊÓƵ Experience Platform data into decision rules, follow these steps:
-
Go to Strategy setup / Decision rules menu and select Create rule with dataset.
-
Click Create mapping to define how the ÃÛ¶¹ÊÓƵ Experience Platform dataset joins with data in Journey Optimizer.
- Select the dataset with the attributes you need.
- Choose a joining key (e.g., product ID or store ID) that exists in both the decision item attributes and the dataset.
note note NOTE You can create up to 3 mappings per rule. -
Click Continue. You can now access the dataset attributes in the Dataset Lookup menu and use them in your rule conditions. Learn how to create a decision rule
Ranking formulas ranking-formulas
Ranking formulas define the priority of decision items. By using ÃÛ¶¹ÊÓƵ Experience Platform dataset attributes, you can dynamically adjust the ranking logic to reflect real-world conditions.
For example, let’s say an airline uses a ranking formula to prioritize upgrade offers. If a customer has a high loyalty tier and current seat availability is low (based on a dataset updated hourly), they are given higher priority. The dataset includes fields like flight_number, available_seats, and loyalty_score.
To use ÃÛ¶¹ÊÓƵ Experience Platform data into ranking formulas, follow these steps:
-
Create or edit a ranking formula. In the Dataset lookup section, click Create mapping.
-
Define the dataset mapping:
- Select the appropriate dataset (e.g., seat availability by flight).
- Choose a joining key (e.g., flight number or customer ID) that exists in both the decision item attributes and the dataset.
note note NOTE You can create up to 3 mappings per ranking formula. -
Use the dataset fields to build your ranking formula as usual. Learn how to create a ranking formula
