AWS connect documentation change

Service: connect · 2026-03-13 · Documentation low

File: connect/latest/adminguide/how-to-create-campaigns-using-event-triggers.md

Summary

Added Predictive Insights integration for personalized recommendations in event-triggered campaigns

Security assessment

Changes focus on recommendation system integration and template personalization features. There is no mention of security controls, data protection, or access management in the added content.

Diff

diff --git a/connect/latest/adminguide/how-to-create-campaigns-using-event-triggers.md b/connect/latest/adminguide/how-to-create-campaigns-using-event-triggers.md
index baf5382ef..785f63924 100644
--- a//connect/latest/adminguide/how-to-create-campaigns-using-event-triggers.md
+++ b//connect/latest/adminguide/how-to-create-campaigns-using-event-triggers.md
@@ -25 +25,21 @@ Event sources are based on integrations in your Customer Profiles domain. detail
-  4. Verify your configurations and choose **Publish**.
+  4. (Optional) Configure the **Recommendations** section to integrate Predictive Insights with your event-triggered campaign. This enables you to deliver personalized template content through email and SMS channels.
+
+![Recommendations section in Amazon Connect console showing Recommender dropdown with frequently_paired_items selected, Calculated attribute for recommender with _last_interacted_item_id, Number of recommendations set to 3, and Recommendation attributes with Name, Price, Description, ImageLink, and Category selected.](/images/connect/latest/adminguide/images/how-to-create-campaigns-recommendations.png)
+
+Configure the following settings:
+
+     * **Recommender name** – Select the name of the recommender you want to use to generate recommendations for the profiles associated with the campaign. You can only use recommenders that are active to generate recommendations.
+
+Predictive Insights offers several types of recommendations. For more information, see [Step 3: Creating Predictive Insights](./predictive-insights-get-started.html#create-predictive-insights).
+
+     * **Calculated Attribute for recommender** – This setting is only required when using a _Similar items_ or _Frequently paired items_ recommender type. This context helps the recommendation engine understand which product to base suggestions on, enabling more relevant and targeted recommendations for your customers.
+
+For example, you might use a calculated attribute like `_last_interacted_item_id` that captures the purchased item ID.
+
+     * **Number of recommendations** – The maximum number of recommendations to generate for a profile. This can range between 1 to 3 recommendations.
+
+     * **Recommendation attributes** – Define which attributes of the recommendations response are used in your message template.
+
+For more information about Predictive Insights, see [Get started with Predictive Insights](./predictive-insights-get-started.html).
+
+  5. Verify your configurations and choose **Publish**.