How to analyze Agent Data Report on Smart Agent Chat?
- Getting Started
- Bot Building
- Conversation Design
-
Developer Guides
Code Step Integration Static Step Integration Shopify Integration SETU Integration Exotel Integration CIBIL integration Freshdesk KMS Integration PayU Integration Zendesk Guide Integration Twilio Integration Razorpay Integration LeadSquared Integration USU(Unymira) Integration Helo(VivaConnect) Integration Salesforce KMS Integration Stripe Integration PayPal Integration CleverTap Integration Fynd Integration HubSpot Integration Magento Integration WooCommerce Integration Microsoft Dynamics 365 Integration
- Deployment
- Agent Setup
- Analytics & Reporting
- Notifications
- Commerce Plus
- Troubleshooting Guides
- Release Notes
Table of Contents
Table of Contents
Chat Data ReportAgent Online/Offline Time ReportAgent Login/Logout Hours ReportAgent Online Hours ReportTips to Manage a Team using ReportsAgent Reports can be downloaded on the Smart Agent Chat tool which gives you all the information related to agents and their time spent on the tool. You can analyze your agent's conversations based on various metrics on a granular level. These reports are detailed reports and they cover most of the aspects related to users' conversations with an agent.
Apart from the downloaded Agent Data Report, you also receive Daily Statistics Report over email at the end of every day. You can read more about it in detail here.
Now let us talk about, how to download these agent reports on the tool. You can click on "Email Detailed Report", choose the time frame as shown in the GIF below and the report would be sent to you over email.
You would receive 4 reports, namely -
-
Chat Data Report
Example filename - ‘chat_data_report.csv’ -
Agent Online/Offline Time Report
Example filename - 'agent_online_and_offline_time_report.csv' -
Agent Login/Logout Hours Report
Example filename -'agent_login_and_logout_time_report.csv'
-
Agent Online Hours Report
Example filename - 'agent_online_hours_report.csv'
Chat Data Report
Business Name is the business via name value which was set in the business manager section.
Example: lalpathlabchannel
, lalpathlabchannel_hi
It is the unique conversation identifier of the user
Example : 28331000_14
It is an agent's unique name who has handled conversations on the system.
Examples: gogo
, saurabhsingh
Note: `gogo` is an internal name for our system (bot)
System generated client-specific user identifier
Example: 58145996
A unique identifier is used for authenticating the user, this is an alphanumeric user identifier from our client's system. And this is unique to an end-user as passed by the client in the integration
Example: 92609cd2758448ae4fbb93b0ed22d40ef3b57e1b
Device platform of the user - Android App, IOS App, Android SDK, Facebook, Jio, Web SDK, IOS SDK , Skype, Alexa, Voice, Twilio, Kookoo, Phonecom, Webhook, Whatsapp, Jio Chat Connector, Haptik IVR, ISO, JIO IVR
This is the value indicating the engagement order for this user's engagement. Say, an end-user spoke to gogo (Haptik bot) and then agent A. The order for the former and later engagements would be 0
and 1
respectively
The values are Yes
if the conversation was reassigned otherwise the value will be No
These are the following reassignment reasons
-
Not Reassigned
This will be present when the chat was not reassigned. -
Reassigned due to Agent Offline/Logout
There are two scenarios whenthe value will be "Reassigned due to Agent Offline/Logout" -- The chat was reassigned when an agent logged out, and the agent has assigned chats to them.
- If the user who was marked as
Waiting for user
comes back, but the agent who handled the chat initially is now offline or logged out.
-
Reassigned by {username}
The above value will be present when the chat is manually reassigned by an agent/TL to another agent/TL. -
Could not be determined
This value will be present if the Chat Reassigned column is True, but there is no reassignment reason. -
Data does not exist
This value will be present when the historical data does not exist.
Team Name is the team's system identifier to which the conversation was assigned to.
Example: App Support Team
Note: For Agent Name = `gogo`, Team name is the business's default team which was set in business manager section. Here, the Team name always may or may not be the business's default team. It depends on which team the conversation goes to, for an Agent.
This is the first message timestamp for the given conversation. Note that this can be the same for 3 rows if the conversation had 3 engagements from gogo to multiple agents. As this is on a conversation level.
Example: 2020-06-25 07:01:05+05:30
It is the message sent by the user that is the first message from the end user.
13. Chat pinned time
The time when the chat was pinned to the particular agent. A chat getting pinned means that the chat appears on the agent's screen.
14. Chat handover time
The time at which the chat was assigned to an agent, either from the bot or from an agent to another agent.
It is the time taken in hh:mm:ss
format by this specific agent for its first response to the end-user including the pending time.
It is an average calculated using all response times by this specific agent. Time here is in hh:mm:ss
format
Example: 30.5
(seconds)
17. Agent First Response Time (without queue time)
It is the time taken in hh:mm:ss
format by this specific agent for its first response to the end-user excluding the pending time
18. Avg Agent Response Time (without queue time)
It is the average time in hh:mm:ss
format calculated using all response times by this specific agent excluding the pending time
19. Agent Resolution Time (without queue and wait time)
Total time conversation was pinned to this specific agent in hh:mm:ss
format. This excludes the time in pending and waiting for state for this user engagement
Weights correspond to the number of times an agent responds after a user message, using these weights, we calculate the average response time of an agent
The total time a conversation was in the queued state before getting assigned to a specific agent in hh:mm:ss
format. This is the sum of all Queue Times this conversation has had before it got closed.
The total time a conversation was in "waiting state" before getting assigned to a specific agent in hh:mm:ss
format. The conversation was in a waiting state, because either the agent was not online, or the agent might have put the user under the "waiting for user" state. It is the sum of all waiting times the chat was in before it got closed.
Disposition reason selected by the agent while closing the conversation
Disposition sub-reason under selected reason selected by the agent while closing the chat
Any closing comment add by the agent while closing the conversation
This metric indicates how this conversation was closed
-
Agent
: If this conversation was closed by an agent -
Bot
: If this conversation was closed when the user reached the end step of the bot -
Autocomplete
: When the conversation got autocompleted because of bot state inactivity, waiting for user inactivity, or no agents online and complete on offline was set for the given team -
API
: When a conversation was completed via an external API -
Reassigned
: If the conversation is reassigned to another agent
- if
agent
then values would be theagent_name
who completed the conversation - if
gogo
then values would be the lastnode_name
which got identified. - if
autocomplete
then one of the following case- Bot completed - If completed because of waiting for bot state inactivity
- Team Offline Flag - If completed because of no agents online and complete on offline was set for the given team.
- Waiting for User - If completed because of waiting for user inactivity
It is the chat link for given conversation of the user
Example: https://<BASE_URL>/athena/user-info/12345678/?conv_no=14&coll_id=28331000
Agents can add notes while in a conversation. All added notes will be added in this column with a pipe '|' separation
Agents can set follow-ups when closing a conversation. The follow-up time set up is shown here
Agents can set follow-ups when closing a conversation. The follow-up comment added is shown here
When the end-user shares their conversation feedback by giving a rating, this will be recorded and visible in this column
When the end-user shares their conversation feedback by giving a rating, the user can also add a text comment which will be recorded and visible in this column
When a conversation engagement is abandoned by the end-user that is no user message sent while the agent has this conversation, it's counted as abandoned by the user and we show a Yes
value in this column
Delayed time is set in the Team settings. For all instances, wherein the delay message is sent i.e. the delay time has passed when no agent is assigned to the end-user, we show a Yes
string value in this column
Total number of messages which were sent by a specific agent to the end-user
The total number of messages received from the end-user.
38. Chat closed by agent
It enables you to identify agents who have closed a chat. If a chat involves more than one agent, the value will be True for the one who has handled it in the last.
It's worth noting that if the same user returns and requests to interact with an agent again, the value will be True for the agent who closed it in the last.
Let's gain more clarity with a small example:
Consider this is a chat flow: Gogo (a bot) > agent1 > agent2 > Gogo > agent3 > agent4 > Gogo
In this case, the column will be TRUE for agent2 and agent4.
39. Wait count
This column of the agent report tells you how many times your chat support agent had to set the 'Waiting' status during the conversation. It happens when a user stops replying during the chat discussion but returns later to get a solution to the reported issue.
For example: If you see 3 in the Wait count column of the agent report, it means your agent had to wait three times for the user before closing the chat.
Agent Online/Offline Time Report
It is an agent's unique name who has handled conversations on the system.
The date on which the agent was online.
If there was no data found for an agent, the report will state No Online time available.
The timestamp at which the agent's status was changed to Online.
If there was no data found for an agent, the report will state Agent was online before the start date selected.
The date on which the agent status was changed to offline.
If there was no data found for an agent, the report will state No Offline date available.
The timestamp at which the agent's status was changed to Offline.
If there was no data found for an agent, the report will state Agent online as of "".
This column specifies who changed the status of the agent to Online.
It can either be the agent name or the TL name.
This column specifies who changed the status of the agent to Offline.
It can either be the agent name or the TL name or the system (the system will mark the agent as offline after 2.5 minutes of inactivity by the agent).
The timezone you have downloaded the report in.
Agent Login/Logout Hours Report
It is an agent's unique name who has handled conversations on the system.
The login date of the agent.
If there was no data found for an agent, the report will state No Login date available.
The timestamp at which the agent has logged in.
If there was no data found for an agent, the report will state Agent was online before.
The date on which the user logged out.
If there was no data found for an agent, the report will state No Logout date available.
The timestamp at which the user logged out of the platform.
If there was no data found for an agent, the report will state Agent online as of "".
The timezone you have downloaded the report in.
Agent Online Hours Report
A new report for measuring the online hours of agents.
1. Agent Name
It is an agent's unique name who has handled conversations on the system.
2. Start Date
The time period starts from which the report is generated
3. Start Time
The day of the hour on the start day from which the report is generated
4. End Date
The time period end for which the report is generated
5. End Time
The day of the hour on the end day till which the report is generated
6. Idle Time Session
This displays the time when agents were idle and not receiving any chats
7. Time Zone
The timezone you have downloaded the report in.
Tips to Manage a Team using Reports
✅ Track No. of Delayed Chats and optimize agent strength per Team
✅ Track No. of Live Chats received at a given time
✅ View Response Times for a given time frame to monitor agent performance
✅ View Resolution Times on a Team/Agent level
✅ View Individual Chats to get quality feedback
✅ Analyze end of day Team-wise analytics values here.