Genesys Cloud's anomaly detection feature automatically identifies unusual queue behavior in real time and alerts supervisors proactively through a configurable monitoring system. The feature requires specific permissions across multiple signal categories, including access to the Anomaly Detection dashboard, settings editing, event viewing and searching, and export capabilities. Organizations can enable anomaly detection on a maximum of 50 queues by navigating to Menu > Analytics > Anomaly Detection and accessing the Settings configuration panel.
Successful implementation requires queues to contain substantial historical data before enablement to ensure accurate prediction and anomaly flagging; enabling detection on newer or data-sparse queues may yield limited dashboard results. Configuration occurs through two primary settings tabs: the General settings tab allows users to specify the time zone for anomaly detection display and processing (with up to 24-hour model adjustment periods time zone changes) and select the week's start day, while the Monitored entities tab enables queue selection for monitoring.
The queue management interface provides search and filtering capabilities to refine queue selection. Users can filter queues by monitoring status (enabled or non-enabled) and set minimum and maximum average daily interaction parameters to assess volume and identify queues meeting historical data requirements. The system displays queue information including monitoring status, queue name, average daily interactions, and data sufficiency indicators. A "sufficient data" filter identifies queues meeting historical data requirements, while queues lacking adequate data display warning indicators signifying insufficient data accuracy and remain unavailable for monitoring.
The system tracks selected queue counts against the 50-queue organizational limit and allows users to save monitored entity selections or cancel to discard changes after applying filter criteria. The anomaly detection functionality depends on adequate historical data accumulation to provide dependable monitoring results and reliable real-time alerting capabilities
Anomaly detection automatically spots unusual queue behavior in real time and alerts supervisors proactively. Enable the queues that require anomaly detection manually. Before you enable the feature on a queue, ensure that the queue contains substantial historical data. The availability of data helps sharpen prediction and then flag anomalies. While you can still enable anomaly detection on newer queues or queues with less data, the Anomaly Detection Dashboard results may not return any useful data.
Note: Ensure you review the best practices before you identify the queues for anomaly detection. You can enable a maximum of 50 queues per org.
Click Menu > Analytics > Anomaly Detection.
Click Settings to configure your anomaly detection preferences and select the queues that you want Genesys Cloud to monitor. The settings include General and Monitored entities.
Use the General settings tab to configure the following time-related preferences that Anomaly Detection uses:
Time zone — Select the time zone to use for anomaly detection and display.
Note: After changing the time zone, anomaly detection models can take up to 24 hours to adjust to the new time zone.
Start day of week — Select the day that represents the beginning of the week for anomaly detection. For example, Sunday.
Click Save to apply your changes, or Cancel to discard them.
Use the Monitored entities tab to select the entities to be monitored for anomalies. Currently, you can configure monitoring for queues. Only queues that you have permission to access are available for selection.
For each queue in the list, you can:
Search for a queue by name using the search box or use Filter to narrow the list of queues.
To enable or disable anomaly monitoring for a queue, select or clear the checkbox in the Monitor? column.
To display only queues that have enough historical interaction data for anomaly detection, enable Only show queues with sufficient data.
Note: You can enable a maximum of 50 queues per org.
View the Average interactions per day for each queue to gauge its interaction volume.
The page displays the number of selected queues against the maximum number that you can monitor, for example, 9 / 50.
The queue list includes the following information:
Columns
Description
Monitor?
Displays whether the queue is selected for anomaly monitoring.
Queue
The name of the queue.
Average interactions per day
The average number of interactions handled by the queue per day.
Queue Filter
To filter the queue list, click Filter. In the filter panel, you can set:
Monitoring? — Select Yes to display queues enabled for monitoring or No to display queues that are not enabled for monitoring.
Minimum average interactions per day — Enter the minimum average number of interactions per day that a queue must have to appear in the list.
Maximum average interactions per day — Enter the maximum average number of interactions per day that a queue can have to appear in the list.
Click Apply to apply the filter criteria, or Cancel to close the filter panel without making changes.
Note: Anomaly detection depends on sufficient historical interaction data for a queue. Queues without enough qualifying data are not available for monitoring or does not provide sufficient data for reliable anomaly detection. A warning icon next to a queue indicates that the queue does not have sufficient data for accurate anomaly detection.
After you select the queues that you want to monitor, click Save to apply your monitored entities selections, or Cancel to discard them.