사용 중인 플랜
추가 기능 코파일럿

The metrics and attributes described in this article are used to create the reports included in the Zendesk Copilot: Agent productivity prebuilt dashboard and the Generative AI Agent Tools prebuilt dashboard.

This article contains the following topics:
  • AI - Copilot suggestions dataset
  • Generative AI agent tools dataset

AI - Copilot suggestions dataset

The AI - Copilot suggestions dataset lists the metrics and attributes you can use to create Explore reports to measure the engagement and performance of the following Copilot features: similar tickets, merging suggestions, quick answers, suggested replies, and suggested macros.

This dataset is used to create reports in the AI suggestions tab of the Zendesk Copilot: Agent productivity prebuilt dashboard.

This section contains the following topics:

  • AI - Copilot suggestion metrics
  • AI - Copilot suggestions attributes

AI - Copilot suggestion metrics

This section lists and defines the metrics available in the AI - Copilot dataset.

Metric Definition Formula
Suggestions The number of AI suggestions shown to agents.

AI suggestions include similar tickets, merging suggestions, quick answers, suggested first replies, and suggested macros. Auto assist suggestions aren't included.

[Suggestion ID]
Accepted suggestions The number of AI suggestions accepted by agents.
  • A similar ticket suggestion is considered accepted when a user opens or hovers over at least one of the suggested similar tickets.
  • A merging suggestion is considered accepted when the user merges at least one of the suggested tickets.
  • A quick answer is considered accepted when a user gives it a thumbs-up.
  • A suggested first reply is counted as accepted when a user presses Tab to accept the suggestion. Acceptance also includes suggestions which were accepted and edited by agents.
  • A suggested macro is considered accepted when a user applies a suggested macro.
IF [Suggestion status]="Accepted" THEN [Suggestion ID] ENDIF
Ignored suggestions The number of AI suggestions that were not accepted by agents. IF [Suggestion status]="Ignored" THEN [Suggestion ID] ENDIF
% Acceptance rate The number of AI suggestions that were accepted, divided by the total number of AI suggestions shown. COUNT(Accepted suggestions)/COUNT(Suggestions)
AI suggestions tickets The number of tickets with an AI suggestion shown at least once. [Suggestion ticket ID]
Accepted AI suggestions tickets The number of tickets where an AI suggestion was accepted at least once. IF [Suggestion status]="Accepted" THEN [Suggestion ticket ID] ENDIF
Tickets with all AI suggestions ignored The number of tickets where no AI suggestion was accepted. D_COUNT(AI suggestion tickets)-D_COUNT(Accepted AI suggestion tickets)
Tickets The total number of tickets. [Ticket ID]
Good satisfaction tickets The number of tickets with a good satisfaction rating. IF ([Ticket satisfaction rating]="Good") THEN [Ticket ID] ENDIF
Rated satisfaction tickets Tickets that were rated either bad or good by the requester. IF ([Ticket satisfaction rating]="Good" OR [Ticket satisfaction rating]="Bad") THEN [Ticket ID] ENDIF
% Satisfaction score The percentage of satisfaction surveys rated good. COUNT(Good satisfaction tickets)/COUNT(Rated satisfaction tickets)
Agent replies The number of public replies added to a ticket by an agent. (Agent replies)
Assignee stations The number of agents a ticket has been assigned to. (Assignee stations)
Group station The number of groups a ticket has been assigned to. (Group stations)
Reopens The number of times a ticket was reopened. (Reopens)
Agents The number of active agents and administrators in your Zendesk account. [Agent ID]
Agents used AI suggestions The number of agents who used AI suggestions at least once by accepting or ignoring it. [Suggestion agent ID]
First reply time (min) The duration in minutes between when the ticket was created and the first public agent reply on the ticket. [First reply time (min)]
Full resolution time (min) The duration in minutes from when the ticket was created to its latest resolution. [Full resolution time (min)]
First reply time (hrs) The duration in hours between when the ticket was created and the first public agent reply on the ticket. VALUE(First reply time (min))/60
Full resolution time (hrs) The duration in hours from when the ticket was created to its latest resolution. VALUE(Full resolution time (min))/60
First reply time (days) The duration in days between when the ticket was created and the first public agent reply on the ticket. VALUE(First reply time (min))/60/24
Full resolution time (days) The duration in days from when the ticket was created to its latest resolution. VALUE(Full resolution time (min))/60/24

AI - Copilot suggestions attributes

This section lists and defines the attributes available in the AI - Copilot dataset.

Attribute Definition
Suggestion ID The unique ID for the AI suggestion.
Suggestion status The status of the AI suggestion. Values can be Accepted or Ignored.
Suggestion type The type of the AI suggestion. Values can be suggested macros, suggested first replies, similar tickets, merging suggestions, or quick answers.
Suggestion agent ID The user ID of the agent who interacted with the AI suggestion.
Suggestion ticket ID The ID of the ticket on which AI suggestion was provided.
Ticket ID The ID number of the ticket.
Ticket status The current status of the ticket. Values can be New, Open, Pending, On-Hold, Solved or Closed.
Ticket custom status name The name of a custom ticket status. This attribute appears only if you've activated custom ticket statuses.
Ticket custom status category The category that a custom ticket status is mapped to. This attribute appears only if you've activated custom ticket statuses.
Ticket custom status state Returns true if a custom ticket status is active, or false if a custom ticket status is deactivated.
Ticket group Name of the group where the ticket is currently assigned.
Ticket assignee The name of the latest ticket assignee.
Ticket brand The brand associated with the ticket.
Ticket channel The channel that initiated creation of the ticket.
Ticket form The current ticket form used on the ticket.
Ticket organization The name of the organization associated with the ticket.
Ticket priority The ticket's priority.
Ticket problem ID The ID of the ticket defined as a problem ticket.
Ticket requester The name of the user who is asking for support through a ticket.
Ticket satisfaction rating The satisfaction rating left by the customer on the ticket. Values can be Good, Bad, Offered, Unoffered.
Ticket subject The subject of the ticket.
Ticket tags Any tags associated with the ticket.
Ticket type The type of ticket. Values can be Question, Incident, Problem, Task.
Sharing agreement inbound Affiliated instances of Zendesk Support and companies who share tickets with current instance of Zendesk Support.
Sharing agreement outbound Affiliated instances of Zendesk Support and companies tickets are shared with.
Agent name The name of the agent.
Agent role The role of the agent.
Agent ID The user ID of the agent.
Agent email The email address of the agent.
Agent locale The locale of the agent.
Agent status The user status of the agent.
Agent tags A list of tags associated with the agent.
Agent time zone The time zone of the agent.
Time - AI suggestion provided A collection of attributes that return the time an AI suggestion was provided to an agent.
Time - Ticket created A collection of attributes that return the time of the ticket creation.
Time - Ticket solved A collection of attributes that return the time of the ticket resolution.
Time - Ticket last updated A collection of attributes that return the time of the last ticket update.

Generative AI agent tools dataset

The Generative AI agent tools dataset lists the metrics and attributes you can use to create Explore reports based on agents’ usage of the following generative AI features: summarize, expand, and make more friendly and make more formal. This dataset is used to create reports included in the AI agent tools tab of the Zendesk Copilot: Agent productivity prebuilt dashboard.

This dataset is also used to create the reports included in the Generative AI Agent Tools prebuilt dashboard.
Attention: The Generative AI agent tools dashboard will be removed in September 2025 as part of the Dashboard Builder transition process.

This section contains the following topics:

  • Generative AI agent tools metrics
  • Generative AI agent tools attributes
Note: This dataset retains data only from the previous 1200 days.

Generative AI agent tools metrics

This section lists and defines the metrics available in the Generative AI agent tools dataset.

Metric Definition Explore formula
Tickets The total number of tickets. [Ticket ID]
Tickets - No AI The number of tickets that did not use any AI tools. IF ([ai_tool_usage_use_case] = NULL) THEN [Ticket ID] ENDIF
Tickets - AI used The number of tickets where an agent used AI tools. IF ([ai_tool_usage_use_case] != NULL) THEN [Ticket ID] ENDIF
First assignment to resolution time (min) The duration in minutes between the first agent assignment and the ticket resolution. DATE_DIFF([Ticket solved - Timestamp],[Ticket first assigned - Timestamp], "nb_of_minutes")
Full resolution time (min) The duration in minutes between ticket creation and its latest resolution.  
First reply time (min) The duration in minutes between ticket creation and the first public agent reply on the ticket.  
Requester wait time (min) The total combined time in minutes the ticket was in the New, Open, and On-hold statuses. It measures how long a requester was waiting for the agent to reply.  
First assignment to resolution time (hrs) The time in hours between the first agent assignment and the ticket resolution. DATE_DIFF([Ticket solved - Timestamp],[Ticket first assigned - Timestamp], "nb_of_hours")
Full resolution time (hrs) The duration in hours between ticket creation and its latest resolution. VALUE(Full resolution time (min))/60
First reply time (hrs) The duration in hours between ticket creation and the first public agent reply on the ticket. VALUE(First reply time (min))/60
Requester wait time (hrs) The total combined time in hours the ticket was in the New, Open, and On-hold statuses. It measures how long a requester was waiting for the agent to reply. VALUE(Requester wait time (min))/60
Full resolution time (days) The duration in days between ticket creation and its latest resolution. VALUE(Full resolution time (min))/60/24
First assignment to resolution time (days) The time in days between the first agent assignment and resolution of the ticket. DATE_DIFF([Ticket solved - Timestamp],[Ticket first assigned - Timestamp], "nb_of_days")
Requester wait time (days) The total combined time in days the ticket was in the New, Open, and On-hold statuses. It measures how long a requester was waiting for the agent to reply. VALUE(Requester wait time (min))/60/24
First reply time (days) The duration in days between ticket creation and the first public agent reply on the ticket. VALUE(First reply time (min))/60/24
Full resolution time - No AI Usage Tickets (days) The duration in days between ticket creation and its latest resolution for tickets that did not use AI tools. IF ([ai_tool_usage_use_case] = NULL) THEN VALUE(Full resolution time (min))/60/24 ENDIF
Full resolution time - AI Usage Tickets (days) The duration in days between ticket creation and its latest resolution for tickets that used AI tools. IF ([ai_tool_usage_use_case] != NULL) THEN VALUE(Full resolution time (min))/60/24 ENDIF
First reply time - Business hours (min) The duration in minutes between ticket creation and the first public agent reply on the ticket within business hours.  
Full resolution time - Business hours (min) The duration in minutes between ticket creation and its latest resolution within business hours.  
Requester wait time - Business hours (min) The total combined time in minutes the ticket was in the New, Open, and On-hold statuses within business hours. It measures how long a requester was waiting for the agent to reply within business hours.  
First reply time - Business hours (hrs) The duration in hours between ticket creation and the first public agent reply on the ticket within business hours. VALUE(First reply time - Business hours (min))/60
Full resolution time - Business hours (hrs) The duration in hours between ticket creation and its latest resolution within business hours. VALUE(Full resolution time - Business hours (min))/60
Requester wait time - Business hours (hrs) The total combined time in hours the ticket was in the New, Open, and On-hold statuses within business hours. VALUE(Requester wait time - Business hours (min))/60
Public comments The number of public comments posted on the ticket. IF ([ticket_updates_comment_present] = TRUE AND [ticket_updates_public_comment] = TRUE) THEN [ticket_updates_id] ENDIF
Public comments per ticket The number of public comments posted per ticket. COUNT([Public comments]) / COUNT([Ticket ID])
Agent comments The number of comments posted on the ticket by agents. IF ([ticket_updates_comment_present] = TRUE AND [updaters_role] != "End-user") THEN [ticket_updates_id] ENDIF
Agent comments per ticket The number of agent comments posted per ticket. COUNT(Agent comments]) / COUNT([Ticket ID])
Good satisfaction tickets The number of tickets that were rated as Good by the requester (end user). IF ([Ticket satisfaction rating]="Good") THEN [Ticket ID] ENDIF
Rated satisfaction tickets The number of tickets that were rated as Bad or Good by the requester (end user). IF ([Ticket satisfaction rating]="Good" OR [Ticket satisfaction rating]="Bad") THEN [Ticket ID] ENDIF
% Satisfaction score The percentage of the tickets that were rated as Good by the requester (end user) from the total amount of satisfaction-rated tickets. COUNT(Good satisfaction tickets)/COUNT(Rated satisfaction tickets)
Summarize ticket The number of times a ticket summary is generated by an agent. IF ([ai_tool_usage_use_case] = 4) THEN [ai_tool_usage_object_id] ENDIF
Expand text The number of times the expand feature is applied to a comment in the composer by an agent. IF ([ai_tool_usage_use_case] = 3) THEN [ai_tool_usage_object_id] ENDIF
Tone shift - Any The number of times the tone shift features (make more friendly or make more formal) are applied to a comment in the composer by an agent. IF ([ai_tool_usage_use_case] = 2 OR [ai_tool_usage_use_case] = 1 ) THEN [ai_tool_usage_object_id] ENDIF
Tone shift - Formal The number of times the tone shift feature (make more formal) is applied to a comment in the composer by an agent. IF ([ai_tool_usage_use_case] = 1) THEN [ai_tool_usage_object_id] ENDIF
Tone shift - Friendly The number of times the tone shift feature (make more friendly) is applied to a comment in the composer by an agent. IF ([ai_tool_usage_use_case] = 2) THEN [ai_tool_usage_object_id] ENDIF
AI tool usage event id The AI tool usage event ID. Used to count the number of times an agent AI tool was used. [ai_tool_usage_object_id]

Generative AI agent tools attributes

This section lists and defines the attributes available in the Generative AI agent tools dataset.

Attribute Definition
AI usage type The AI tool usage type. Possible values include No AI Usage, Make more formal, Make more friendly, Expand, and Summarize.
AI usage - Ticket ID

The ticket ID associated with the AI tool usage.

Note: If an agent uses any AI tools on a new ticket before submitting their first comment, no associated ticket ID is recorded because no ticket ID exists until the agent clicks Submit. However, the AI tool usage is still counted as part of the AI usage type metric.

Agent name The name of the agent who used the AI tool.
Ticket ID The ID of the ticket.
Ticket status The status of the ticket. Possible values include New, Open, Pending, On-hold, Solved, and Closed.
Ticket group The group to which the ticket belongs.
Ticket brand The brand of the ticket.
Ticket channel The channel a ticket was created from.
Ticket form The current ticket form used on the ticket.
Ticket priority The ticket's priority.
Ticket problem ID The IDs of the ticket defined as a problem ticket.
Ticket assignee The name of the assignee.
Ticket requester The name of the user who is asking for support through a ticket.
Ticket subject The subject line (title) of a ticket.
Ticket tags The tags associated with the ticket.
Ticket type The ticket type: Question, Incident, Problem, or Task.
Ticket organization name The name of the organization associated with the ticket.
Ticket satisfaction rating The satisfaction rating left by the customer on the ticket. Possible values include Good, Bad, Offered, and Unoffered.
Ticket brand The brand to which the ticket belongs.
Time - Ticket first assigned Includes a number of attributes that return the time when a ticket was first assigned to an agent in the ticket history in various time measurements.
Time - Ticket solved Includes a number of attributes that return the time the ticket was last solved in various time measurements.
Time - Ticket created Includes a number of attributes that return the time when the ticket was created in various time measurements.
Time - AI tool used Includes a number of attributes that return the time when the AI tool was used in various time measurements.

Zendesk 제공