Published at: 2026-09-17
Service operation analysis
Analyze session, agent and channel service data.
Scenario
Service operation analysis aggregates session, personnel and channel data to help administrators locate service efficiency and quality issues.
Before you begin
- Already have the data viewing permissions for customer service Data Statistics and related Session Record.
- View scope is controlled by the organization and Data Permission; operations involving Chat Histories, transferring, or ending sessions should be performed by personnel with appropriate permissions.
View service overview
After entering customer service Data Statistics, you can view an overview of service status by time period and channel, which can be used to evaluate the overall service load and response quality.
Focus on the following indicators:
- Total Sessions: The number of sessions within the selected time period and channel.
- Average First Response Duration: Initiated by Account The average wait time to get the first reply from customer service after a new session.
- Average Session Response Time: During the session, the average length of time for a customer message to receive a reply from customer service.
- Average session duration: The average time it takes for a session from access to end.
- Service Satisfaction: Customer average level of Service Evaluation.
- Transfer rate: The proportion of sessions that were transferred to the total number of sessions.
The results can be filtered by time and channel to identify service links with delayed responses, session backlogs or abnormal transfers.

View and manage customer service sessions
In Customer Service Session List, you can filter Session Record by agent, customer, channel, and Session Status and view the chat content in a specific session.
When you have the corresponding permissions, you can perform operations such as transferring and ending sessions that are waiting for allocation or require intervention. It should be processed in conjunction with the session context to avoid affecting ongoing customer service.

View support agent data
View work data from the support agents dimension to understand the handled session volume, response time and service quality of different personnel or teams.
Key points to compare:
- Session processing volume;
- First response and session response time;
- service satisfaction;
- Online and session handling status.
The analysis results can be used for shift scheduling, training, group configuration and service rule optimization.

View WeChat Account
When an enterprise has accessed the WeChat Official Account channel, it can view the WeChat Account information following the service number in the WeChat Account module. When binding multiple service accounts, you can switch the service accounts to view the corresponding customer range.
This data can be used to understand the customer coverage and service interaction of WeChat channels; the Visibility of specific customer information shall be subject to Data Permission.

Using data for service improvements
It is recommended to review service data on a regular basis:
- First, identify the channels or time periods that need attention through the service overview.
- Check the related Session Record again to determine whether the problem belongs to session handling load, allocation rules, knowledge support or personnel capabilities.
- Combine support agent data to develop adjustment measures, such as optimizing shift schedules, supplementing training, or adjusting session allocation rules.
- After adjustments, continue to observe changes in First Response Duration, satisfaction, and transfer rates.
Expected Result
- Select the target statistical period in Customer Service Workbench > Service Operation Analysis.
- View session volume, response time, and agent data.
- Confirm that the indicators are consistent with the current filter conditions and expected business scope.
FAQ and troubleshooting
- When the data is empty, check the statistical period, channel and support agent filtering conditions.
-
When the indicator is abnormal, confirm whether the session has ended and completed Data Statistics. ## Related topics
- Initialization and basic configuration
- support agent management
- Handle sessions