Genesys Cloud provides comprehensive analytics for predictive routing queue performance through a detailed queue view interface accessible via Menu > Analytics > Analytics Workspace. The platform enables users to set a default time zone before accessing queue-specific analytics, which are organized across multiple tabs: displays predictive routing impact for a selected period, Impact shows performance across multiple KPIs beyond the target metric, and Predictive Model identifies top features influencing routing decisions.
The analytics framework presents metrics specific to selected KPIs, with Average Handle Time (AHT) analysis including AHT Difference Percentage (comparing predictive versus non-predictive routing), AHT Difference (expressed in hours, minutes, and seconds), Time Saved (calculated by multiplying total predictive-routed interactions by AHT difference), and total Interactions count comparing predictive-routed versus other routing methods. For multi-media type queues, breakdowns by media type are provided alongside aggregate metrics, with zero-value interactions excluded from total calculations to maintain accurate aggregations.
Genesys Cloud implements a fallback routing mechanism that activates standard routing when predictive routing times out or when queue agent count falls to three or fewer. The Non-Predictive (Timeout/Insufficient Agents) metric tracks interactions where predictive routing was attempted but non-predictive methods were ultimately used. Visualization tools include an Interactions bar chart presenting weekly routing volume comparisons across selected date ranges, and an AHT Difference % scatter plot tracking Average Handle Time variations on a weekly basis relative to a zero baseline.
The Next Contact Avoidance (NCA) KPI measures repeat contact prevention effectiveness, displaying the percentage difference in avoided repeat contacts between predictive and non-predictive routing methods. The Benefit metric quantifies the actual number of repeated contacts avoided through predictive routing during the selected period. Performance analysis is visualized through an NCA Difference % scatter plot displaying weekly values relative to a zero baseline, where positive values indicate improved predictive routing performance and negative values indicate underperformance compared to alternative routing methods. Users can modify the time period using the date selector and navigate between views using the back icon, with hover functionality providing detailed data for specific dates
The queue detail view of an individual queue presents different aspects of the performance of the selected queue. The page contains the following tabs:
Overview – Opens the queue detail view summarizing the impact of predictive routing for the selected period. More information about the overview tab is available in the subsequent sections of this page.
Impact – Opens the queue impact view that enables you to view predictive routing performance for a collection of KPIs in addition to the target KPI. For more information, see Predictive routing queue impact view.
The details of a specific queue depend on the KPI that you choose.
Use the Average handle time KPI to display the following details.
Detail
Description
Notes (For reference, refer to the image)
Average handle time (AHT Difference %)
Indicates the percentage difference in the amount of handling time of interactions that routed using predictive routing compared with the amount of handling time of interactions that routed using non-predictive routing methods during the selected time period.
When a queue supports more than one media type, a breakdown of the AHT Difference % per media type appears in addition to the AHT Difference % value.
The AHT Difference % does not match the sum or average of the % of each media type.
AHT Difference
Indicates the difference in the AHT value between the interactions that routed using predictive routing and the interactions that routed without predictive routing, during the selected date range per interaction. This value appears in hours, minutes, and seconds.
The AHT Difference value is only displayed for a specific (single) media type queue.
If a queue includes more than one media type, a breakdown of the AHT difference per media type appears.
Time saved
Indicates the amount of handling time saved when using predictive routing during the selected time period.
Time saved is calculated by multiplying the total number of interactions, using predictive routing, by the AHT difference for the specified date range.
When a queue supports more than one media type, a breakdown of the data according to media type also appears below the Time saved table.
Interactions
Indicates the total number of interactions that routed with predictive routing, compared with the number handled by other routing methods.
For example, in the Single media queue image below, 1010 is the total number of interactions that routed with predictive routing, and 2535 is the total number of interactions handled without predictive routing.
When a queue supports more than one media type, a breakdown of the data according to media type also appears.
If the number of interactions of any media type is zero, the total number of interactions differs, as the zero value is not considered for the total.
Interactions bar chart
Displays predictive routing and non-predictive routing volumes for the selected date range in weekly intervals. If Genesys Cloud finds no qualifying agent before predictive routing times out, or if the number of agents in the queue is three or fewer, the interaction routes viastandard routing, which is the fallback routing method. For more information about the agent selection process, see the Agent selection process.
The first week shown in the chart starts from the first day of the selected date range. The last week is a full or partial week, depending on the selected date range.
The Non-Predictive (Timeout/Insufficient Agents) bar represents all the interactions where Genesys Cloud attempted predictive routing but instead used non-predictive routing methods.
Ongoing value (AHT Difference % Scatter plot)
This scatter plot depicts weekly AHT Difference % values above or below a zero line.
A dot on the zero line indicates that predictive routing did not make a difference in the AHT of predictive interactions compared with non-predictive interactions.
Positive values above the zero line indicate that interactions routed using predictive routing have, on average, a longer handle time than interactions routed without predictive routing.
Values below the zero line indicate that interactions routed using predictive routing have a shorter handle time than interactions routed without predictive routing.
To see data for that specific date, hover above one of the points on the graph.
The first week shown in the chart starts from the first day of the selected date range. The last week is a full or partial week, depending on the selected date range.
Use the Next contact avoidance KPI to display the following details.
Detail
Description
Notes (For reference, refer to the image)
Next Contact Avoidance (NCA Difference %)
Indicates the percentage difference in the number of repeat contact avoided interactions that routed using predictive routing compared with the number of repeat contact avoided interactions that routed using non-predictive routing methods during the selected time period.
When a queue supports more than one media type, a breakdown of the NCA Difference % per media type appears, in addition to the NCA Difference % value.
The NCA Difference % does not match the sum or average of the % of each media type.
Benefit
Indicates the number of repeated contact-avoided interactions when using predictive routing during the selected time period.
The benefit is calculated based on the number of repeated contacts avoided.
When a queue supports more than one media type, a breakdown of the data according to media type also appears below the Benefit table.
Interactions
Indicates the total number of interactions that routed with predictive routing compared with the number handled by other routing methods.
For example, in the Single media queue image below, 27147 is the total number of interactions that routed with predictive routing, and 208 is the total number of interactions handled without predictive routing.
When a queue supports more than one media type, a breakdown of the data according to media type also appears.
If the number of interactions of any media type is zero, the total number of interactions differs, as Genesys Cloud does not consider the zero value for the total.
Interactions bar chart
Displays predictive routing and non-predictive routing volumes for the selected date range in weekly intervals. If Genesys Cloud finds no qualifying agent before predictive routing times out, or if the number of agents on the queue is three or fewer, the interaction routes via standard routing, which is the fallback routing method. For more information about the agent selection process, see the Agent selection process.
The first week shown in the chart starts from the first day of the selected date range. The last week is a full or partial week, depending on the selected date range.
The Non-Predictive (Timeout/Insufficient Agents) bar represents all the interactions where Genesys Cloud attempted predictive routing but instead used non-predictive routing methods.
Ongoing value (NCA Difference % Scatter plot)
This scatter plot depicts weekly NCA Difference % values above or below a zero line.
A dot on the zero line indicates that predictive routing did not make a difference in the NCA of predictive interactions compared with non-predictive interactions.
Positive values above the zero line indicate that predictive routing performance improved.
Negative values below the zero line indicate that predictive routing is not providing better results than other routing methods.
To see data for that specific date, hover above one of the points on the graph.
The first week shown in the chart starts from the first day of the selected date range. The last week is a full or partial week, depending on the selected date range.
Notes:
To change the time period, click the date at the top of the view and select a different date.
To return to the Predictive routing list, click the following back icon next to the queue view name.