Predictive Routing A/B Testing Report

Predictive Routing A/B Testing Report

Use the Predictive Routing A/B Testing Report to compare results for predictive models and predictors based on time-sliced A/B testing. This report includes a First Contact Resolution Rate calculation, which allows you to quickly see how often customer concerns were resolved on the first attempt, and allows you to contrast interactions that were processed when Predictive Routing was switched ON compared to when it was OFF. The report also profiles response time, engage time, wrap time, and other relevant Key Performance Indicators (KPI).

This report requires that specific RAA options be enabled: enable-gpr and enable-gpr-fcr. For more information, see the Genesys CX Insights Deployment Guide.

To get a better idea of what this report looks like, view sample output from the report: Sample_Predictive Routing_ABTesting_Report.pdf

The following tables explain the prompts, attributes, and metrics used in this report:

Prompts in the Predictive Routing A/B Testing Report

Prompt	Description
Pre-set Date Filter	Choose a date from the list of preset options. This prompt overrides the Start Time and End Time values. Default: Year to Date
Start Date	Choose the day and time from which to begin collecting data into the report. This prompt has no effect if Pre-set Date Filter is set to anything except None.
End Date	Choose the day and time at which to stop collecting data into the report.
Media Type	Select one or more media types for which to gather data into the report.
Predictor	Select one or more predictors for which to gather data into the report.
Model	Select one or more models for which to gather data into the report.
Tenant	Select one or more tenants to include in the report.

Attributes in the Predictive Routing A/B Testing Report

Attribute	Description
Tenant	Enables the organization of data by tenant.
Media Type	Enables the organization of data by media type.
Day	Enables the organization of data by the day/date on which the interaction occurred.
Predictor Switch	Enables the organization of data by whether predictive routing is ON or OFF.
Predictor	Enables the organization of data by the identifier for the predictor that was used to request scoring for predictive routing.
Model	Enables the organization of data by the identifier for the model that was used to calculate agent scores for predictive routing.
Result	The result of Predictive Routing processing. If there is an error, this metric displays the error message (gpmMessage) as a value between 1 and 15. 1 - Ok 2 - Authentication to scoring engine failed 3 - Scoring request failed 4 - Agent list is empty 5 - URS overload, interaction skipped 6 - Predictor not found 7 - Failed to build scoring request 8 - SetIdealAgent or SetReadyCondition execution error 9 - Interaction log not found in global map 10 - Unknown error 11 - Channel is not supported 12 - Reserved for future use 13 - Call Abandoned 14 - Call Routing Failed 15 - Predictive Routing is turned off or not used for this interaction

Metrics in the Predictive Routing A/B Testing Report

Metric	Description
Offered	The total number of customer interactions that entered or began within the contact center during the reporting interval, and were offered to a resource, excluding interactions that were abandoned within the short-abandoned threshold.
Accepted	The total number that customer interactions and warm consultations that were accepted, answered, or pulled by an agent, voice-treatment port, IVR port, or nonagent-associated DN (such as contact center resources that can alert) within the reporting interval.
Avg Handle Time (Fmt)	The average amount of time (HH:MM:SS), within the reporting interval, that this agent spent handling interactions that the agent received. Computed as handle time divided by the sum of accepted interactions and received consultations.
First Contact Resolution Result	First Contact Resolution (FCR) measures whether issues were resolved during the first customer attempt. A value of NO indicates that the customer raised the same issue again within 7 days. The logic for calculating FCR is as follows: Interaction data is retrieved from the INTERACTION_FACT table for 7 and 35 days, leveraging the START_DATE_TIME_KEY column. The interactions are grouped based on the value of CUSTOMER_ID and SERVICE_TYPE user data for each interaction. If the number of interactions within the reporting interval for a given CUSTOMER_ID and SERVICE_TYPE combination is more than 1, then FCR=NO. The FCR result for each interaction is stored in a separate table (FCR_ID).
ASA (Fmt)	The average amount of time (HH:MM:SS), within the reporting interval, that it took agents to accept, answer, or pull customer interactions
Avg Wrap Time (Fmt)	The average amount of time (HH:MM:SS), within the reporting interval, that this agent spent on customer interactions while in ACW (Wrap) state.
Avg Engage Time (Fmt)	The average amount of time (HH:MM:SS), within the reporting interval, that this agent was engaged with customers on interactions.
Avg Hold Time (Fmt)	The average number of seconds, within the reporting interval, that customers spent on hold for interactions. This metric is attributed to the interval in which the interactions were accepted by a resource.
Transfer Rate	The percentage of interactions that were transferred. Calculated as the total number of transferred interactions divided by the total number of interactions.