Message conversation pivot

Overview

The message conversation hub mainly analyzes and aggregates conversation data. A conversation is a unit used to track and manage dialogues between customers and agents or the system (bot).

A conversation can exist in multiple statuses, which are used to indicate different stages or conditions of the conversation. Some common conversation statuses may include “In Progress,” “Assigned,” and “Completed.” Using these statuses helps you track and organize conversations while also giving your team a clear understanding of the current status.

And each status can be found in the “Conversation Status” article to learn more about the meaning of each status and the corresponding actions.

Based on the image example, you can see that the same customer may have multiple conversations, and each conversation represents an independent dialogue sequence. After the previous conversation is completed, the next message will be treated as a new conversation.

These conversations are displayed in reports according to different statuses, and the situation at the time is shown based on the report execution time. Please note that the situation presented in the report is based on the data at the time the report is run, so it may not reflect the latest conversation status in real time.

Field

Name	Description	Additional Notes
Status	The current status of the conversation, such as “In Progress,” “Assigned,” and “Completed.”
Created Date	The “Created Date” usually refers to the start date of the conversation, that is, the time of the first message initiated by the customer or agent.	Messages proactively sent by system audience segmentation. The time of these messages will be included in the range of the created date.
Inbound Date	The “Inbound Time” refers to the time when the conversation enters the assignment queue and starts looking for a human customer service agent.
Acceptance Date	The “Acceptance Date” refers to the date when the conversation’s status enters Assigned, and the agent picks it up and starts providing service.
Agent	Indicates the agent currently assigned to the conversation or the agent who started providing service.
Source Channel	The conversation’s source channel, such as LINE, Facebook Messenger, etc.

How do you interpret the report?

The way you interpret the report will vary depending on the bot flow or business logic settings you have designed. Here are some common interpretation examples:

1. Channel comparison: Compare the number of conversations and the number of services across different channels. Observe whether one channel has a particularly high number of conversations, which may indicate that the channel needs more attention and improvement, or that there are special situations that need to be handled.

2. Status analysis: Observe the distribution of each status, such as Assigned, In Conversation, Resolved, etc. If one status has a particularly high number, it may mean that you need to pay attention to the handling efficiency of conversations in that status or improve the bot flow.

3. Time analysis: For example, a noticeable difference between the number of inbound times and acceptance times is an important point to observe. If there is a significant gap between these two numbers, you may need to further understand the reason, such as whether there are enough human resources or whether service agents can respond in time.

In addition to the examples above, interpreting the report also requires analysis based on your business needs and specific context.

Frequently Asked Questions

Has an inbound date but the status is “Conversation Script” or “Bot”?

If the conversation’s status is “Conversation Script” or “Bot,” and there is also an inbound date record, this means the conversation once entered the assignment queue, but later triggered a return to bot service through keywords or other means.

For example, if a customer seeks human customer service through a script at 18:23 on 05-14, the inbound date will be recorded as 05-14. However, the customer may later enter a keyword such as “leave,” triggering the system to return the conversation to bot self-service.

In this case, the inbound date records the time when the customer initially entered the assignment queue, and a subsequent specific keyword trigger causes the conversation to return to bot service.

Is the conversation count equal to the service record count?

The number of conversations is not equal to the actual number of service records.

Sometimes customers may obtain the correct answer through self-service or other means during the bot flow and do not need further intervention from a human customer service agent. In this case, the conversation count will increase by 1, but the actual number of service records will still be 0. This means the customer successfully resolved the issue during the conversation and did not need further human intervention.