AWS Security ChangesHomeSearch

AWS connect: Removed GenAI Evaluation and Contact observed metric definitions

Service: connect · 2026-11-05 · Documentation low

File: connect/latest/adminguide/metric-primitive-definitions.md

Summary

Deletes the metric primitive definitions for the GenAI Evaluation category (Evaluated churn contacts, CSAT score, sale, self-service success, partial self-service success) and the Contact observed category (Contact Segment Count, Contact Segment Time), along with their associated filter/grouping table rows (LIFETIME_VALUE, REPEAT_CALLER, GenAI Evaluation and Contact observed groupings), and changes the trailing link from 'Observed metrics' to 'Assign permissions'.

Security assessment

The change is purely editorial documentation restructuring: it removes metric primitive definitions and filter table rows for GenAI Evaluation and Contact observed metrics and repoints a navigation link. No authentication, authorization, encryption, credential, or vulnerability content is added or fixed, so there is no security implication.

Evidence

-## Evaluated churn contacts

Diff

diff --git a/connect/latest/adminguide/metric-primitive-definitions.md b/connect/latest/adminguide/metric-primitive-definitions.md
index 79395fb..6b9e716 100644
--- a//connect/latest/adminguide/metric-primitive-definitions.md
+++ b//connect/latest/adminguide/metric-primitive-definitions.md
@@ -7 +7 @@
-After contact work timeAgent Active TimeAgent Interaction TimeContact Hold TimeAgent Pause TimeContact DurationContact Queue TimeContacts AbandonedContacts CreatedContacts HandledContacts Hold AbandonsContacts Put On HoldContacts QueuedContacts Transferred OutContact Handle TimeContact HoldsContact Resolution TimeContact Flow DurationAgent Greeting TimeAgent Interruption TimeAgent InterruptionsTalk Time AgentTalk Time CustomerNon-Talk TimeTalk TimeConversation DurationContacts RoutedAgent Contacts MissedAgent Idle TimeAgent Contact TimeAgent Online TimeAgent Error Status TimeAgent Non-Productive TimeAgent Connecting TimeCurrent Contacts In QueueCurrent Contact Queue TimeCurrent Contacts ScheduledCurrent Slots AvailableCurrent Slots ActiveCurrent Agents OnlineEvaluated churn contactsEvaluated CSAT scoreEvaluated saleEvaluated self-service successEvaluated partial self-service successFlows StartedFlows CompletedFlow DurationBot Conversations CompletedBot Intents CompletedBot Conversation DurationBot Conversation TurnsContact Segment CountContact Segment TimeMetric level filters supported per metric primitive categoryGroupings supported per metric primitive categorySupported top-level metric filters per metric primitive categoryGuidelines for Metric Primitive creation and usage with out-of-the-box metrics
+After contact work timeAgent Active TimeAgent Interaction TimeContact Hold TimeAgent Pause TimeContact DurationContact Queue TimeContacts AbandonedContacts CreatedContacts HandledContacts Hold AbandonsContacts Put On HoldContacts QueuedContacts Transferred OutContact Handle TimeContact HoldsContact Resolution TimeContact Flow DurationAgent Greeting TimeAgent Interruption TimeAgent InterruptionsTalk Time AgentTalk Time CustomerNon-Talk TimeTalk TimeConversation DurationContacts RoutedAgent Contacts MissedAgent Idle TimeAgent Contact TimeAgent Online TimeAgent Error Status TimeAgent Non-Productive TimeAgent Connecting TimeCurrent Contacts In QueueCurrent Contact Queue TimeCurrent Contacts ScheduledCurrent Slots AvailableCurrent Slots ActiveCurrent Agents OnlineFlows StartedFlows CompletedFlow DurationBot Conversations CompletedBot Intents CompletedBot Conversation DurationBot Conversation TurnsMetric level filters supported per metric primitive categoryGroupings supported per metric primitive categorySupported top-level metric filters per metric primitive categoryGuidelines for Metric Primitive creation and usage with out-of-the-box metrics
@@ -475,50 +474,0 @@ This metric counts the number of agents who are currently online in the contact
-## Evaluated churn contacts
-
-Count of contacts evaluated as showing churn propensity. A score of 1 means the evaluation identified churn risk; a score of 0 means no churn risk was identified. For churn propensity, a score of 0 is the favorable outcome. For more information about creating evaluation forms, see [Create an evaluation form in Connect Customer](https://docs.aws.amazon.com/connect/latest/adminguide/create-evaluation-forms.html).
-
-**Metric Primitive Name:** `Evaluated churn contacts`
-
-**Metric Primitive Category:** GenAI Evaluation
-
-**Supported Statistics:** AVG and SUM
-
-## Evaluated CSAT score
-
-Customer satisfaction score produced by contact evaluation. When you create an evaluation form with business outcome questions, each evaluated contact receives a CSAT score on a 1 to 5 scale. For more information about creating evaluation forms, see [Create an evaluation form in Connect Customer](https://docs.aws.amazon.com/connect/latest/adminguide/create-evaluation-forms.html).
-
-**Metric Primitive Name:** `Evaluated CSAT score`
-
-**Metric Primitive Category:** GenAI Evaluation
-
-**Supported Statistics:** SUM, AVG, MIN, and MAX
-
-## Evaluated sale
-
-Count of contacts evaluated as resulting in a sale. A score of 1 means a successful sale was identified; a score of 0 means no sale was identified. For more information about creating evaluation forms, see [Create an evaluation form in Connect Customer](https://docs.aws.amazon.com/connect/latest/adminguide/create-evaluation-forms.html).
-
-**Metric Primitive Name:** `Evaluated sale`
-
-**Metric Primitive Category:** GenAI Evaluation
-
-**Supported Statistics:** AVG and SUM
-
-## Evaluated self-service success
-
-Count of self-service evaluations where the customer fully self-served. A score of 1 means the customer fully resolved their issue without agent assistance; a score of 0 means they did not. For more information about performance evaluations for automated interactions, see [Performance evaluations for automated interactions](https://docs.aws.amazon.com/connect/latest/adminguide/performance-evaluations-automated-interactions.html) and [Generative AI performance evaluations](https://docs.aws.amazon.com/connect/latest/adminguide/generative-ai-performance-evaluations.html). For more information about creating evaluation forms, see [Create an evaluation form in Connect Customer](https://docs.aws.amazon.com/connect/latest/adminguide/create-evaluation-forms.html).
-
-**Metric Primitive Name:** `Evaluated self-service success`
-
-**Metric Primitive Category:** GenAI Evaluation
-
-**Supported Statistics:** AVG and SUM
-
-## Evaluated partial self-service success
-
-Count of self-service evaluations where the customer partially self-served. A score of 1 means the customer partially resolved their issue before agent assistance; a score of 0 means they did not. For more information about performance evaluations for automated interactions, see [Performance evaluations for automated interactions](https://docs.aws.amazon.com/connect/latest/adminguide/performance-evaluations-automated-interactions.html). For more information about creating evaluation forms, see [Create an evaluation form in Connect Customer](https://docs.aws.amazon.com/connect/latest/adminguide/create-evaluation-forms.html).
-
-**Metric Primitive Name:** `Evaluated partial self-service success`
-
-**Metric Primitive Category:** GenAI Evaluation
-
-**Supported Statistics:** AVG and SUM
-
@@ -597,20 +546,0 @@ The turns in completed conversations for which the invoking resource (flow or fl
-## Contact Segment Count
-
-This metric counts contact segments of a given state. A segment is one state a contact passed through, such as queued, connected to an agent, or paused. To create a custom metric that counts specific segments, filter this primitive by the desired segment type. This primitive counts each segment that overlaps the requested interval, whether the segment is still open or has ended. For more information, see [Observed metrics](./observed-metrics-concepts.html).
-
-**Metric Primitive Name:** `Contact segment count`
-
-**Metric Primitive Category:** Contact observed
-
-**Supported Statistics:** COUNT
-
-## Contact Segment Time
-
-This metric measures the duration of contact segments in a given state. Filter by segment type to select the state to measure. Over an interval, each segment is clipped to the interval boundaries: a segment that starts before the interval or ends after it contributes only the portion of its duration that falls inside the interval. For more information, see [Observed metrics](./observed-metrics-concepts.html).
-
-**Metric Primitive Name:** `Contact segment time`
-
-**Metric Primitive Category:** Contact observed
-
-**Supported Statistics:** AVG
-
@@ -673,6 +602,0 @@ Channel |  How the contact reached your contact center. |  Valid values: Voice,
-GenAI Evaluation |  Channel |  How the contact reached your contact center. |  Valid values: Voice, Chat, Task, Email  
-Metric score |  The raw score value from the business outcome evaluation question. |  For CSAT: 1, 2, 3, 4, 5. For all others: 0, 1  
-Contact observed |  Segment type |  The contact state the segment represents. |  Valid values: INITIATED, QUEUED, CONNECTED_TO_AGENT, PAUSED.  
-Channel | How the contact reached your contact center. | Valid values: Voice, Chat, Task, Email  
-Initiation method | Indicates how the contact was initiated. | Example: INBOUND, OUTBOUND, TRANSFER, QUEUE_TRANSFER, CALLBACK, API.  
-Subtype | The subtype of the channel used for the contact. | For valid values, see the **connect:Subtype** table row in the [SegmentAttributes](https://docs.aws.amazon.com/connect/latest/adminguide/connect-attrib-list.html#attribs-segment-attributes) section.  
@@ -696,2 +619,0 @@ contact/segmentAttributes/connect:ValidationTestType | Contact source
-LIFETIME_VALUE | Lifetime value  
-REPEAT_CALLER | Repeat caller  
@@ -725,2 +646,0 @@ FLOWS_NEXT_RESOURCE_QUEUE_ID | Only available through API
-LIFETIME_VALUE | Lifetime value  
-REPEAT_CALLER | Repeat caller  
@@ -740,23 +659,0 @@ INVOKING_RESOURCE_TYPE | Only available through API
-LIFETIME_VALUE | Lifetime value  
-REPEAT_CALLER | Repeat caller  
-GenAI Evaluation | AGENT | Agent  
-AGENT_HIERARCHY_LEVEL_ONE| Agent hierarchy level one  
-AGENT_HIERARCHY_LEVEL_TWO| Agent hierarchy level two  
-AGENT_HIERARCHY_LEVEL_THREE| Agent hierarchy level three  
-AGENT_HIERARCHY_LEVEL_FOUR| Agent hierarchy level four  
-AGENT_HIERARCHY_LEVEL_FIVE| Agent hierarchy level five  
-CHANNEL| Channel  
-EVALUATION_FORM| Evaluation form  
-EVALUATION_QUESTION| Evaluation question  
-EVALUATION_SECTION| Evaluation section  
-EVALUATION_SOURCE| Evaluation source  
-EVALUATOR_ID| Evaluator  
-LIFETIME_VALUE| Lifetime value  
-QUEUE| Queue  
-REPEAT_CALLER| Repeat caller  
-ROUTING_PROFILE| Routing profile  
-Contact observed | AGENT | Agent  
-CHANNEL | Channel  
-QUEUE | Queue  
-ROUTING_PROFILE | Routing profile  
-contact/segmentAttributes/connect:Subtype | Subtype  
@@ -787,2 +683,0 @@ FEATURE | N/A | Identifies if conversational analytics is enabled on the flow.co
-LIFETIME_VALUE | Lifetime value |   
-REPEAT_CALLER | Repeat caller |   
@@ -837,2 +731,0 @@ FLOWS_NEXT_RESOURCE_QUEUE_ID | Only available through API | ARN of the next queu
-LIFETIME_VALUE | Lifetime value |   
-REPEAT_CALLER | Repeat caller |   
@@ -852,23 +744,0 @@ INVOKING_RESOURCE_PUBLISHED_TIMESTAMP | Only available through API | Timestamp w
-LIFETIME_VALUE | Lifetime value |   
-REPEAT_CALLER | Repeat caller |   
-GenAI Evaluation | AGENT | Agent | Valid input to this key is Agent ARNs  
-AGENT_HIERARCHY_LEVEL_ONE| Agent hierarchy level one| Valid input to this filter is an Agent Hierarchy Level ARN.  
-AGENT_HIERARCHY_LEVEL_TWO| Agent hierarchy level two  
-AGENT_HIERARCHY_LEVEL_THREE| Agent hierarchy level three  
-AGENT_HIERARCHY_LEVEL_FOUR| Agent hierarchy level four  
-AGENT_HIERARCHY_LEVEL_FIVE| Agent hierarchy level five  
-CHANNEL| Channel| Valid values: Voice, Chat, Task, Email  
-EVALUATION_FORM| Evaluation form| Valid input is Evaluation Form ARNs  
-EVALUATION_QUESTION| Evaluation question|   
-EVALUATION_SECTION| Evaluation section|   
-EVALUATION_SOURCE| Evaluation source|   
-EVALUATOR_ID| Evaluator|   
-LIFETIME_VALUE| Lifetime value|   
-QUEUE| Queue| Valid inputs to this key are Queue ARNs  
-REPEAT_CALLER| Repeat caller|   
-ROUTING_PROFILE| Routing profile| Valid input is Routing Profile ARNs.  
-Contact observed | AGENT | Agent | Valid input to this key is Agent ARNs.  
-CHANNEL | Channel | Valid values: Voice, Chat, Task, Email  
-QUEUE | Queue | Valid input to this key is Queue ARNs.  
-ROUTING_PROFILE | Routing profile | Valid input to this key is Routing Profile ARNs.  
-contact/segmentAttributes/connect:Subtype | Subtype | For valid values, see the **connect:Subtype** table row in the [SegmentAttributes](https://docs.aws.amazon.com/connect/latest/adminguide/connect-attrib-list.html#attribs-segment-attributes) section.  
@@ -986 +856 @@ Metric definitions
-Observed metrics
+Assign permissions