Analytics Tutorial

Introduction

A focus on numbers can really boost your support performance. Our built-in Analytics allow you to take a birds-eye view on your messaging service, identify critical areas and zoom in. In this tutorial we’ll break down every KPI at your disposal in Analytics.

Enter Analytics from your Dashboard’s left navigation bar under: Unified Messaging Analytics

Filters and resolution settings

You can filter each KPI’s data by date range, Widget, operator, and group. This way you define which part of the collected data will be displayed.

With the resolution options **Hours**, **Days** and **Months** you can switch between more detail and more overview in how the data is displayed. If you choose Auto, the resolution is automatically set based on your selected date range. For example, a date range set to Today, would result in the data on an hourly basis for that day.

Also, choose between four different styles to visualize your data: line graph, bar graph, pie chart and table chart. To export your selection as a CSV file, hit **Download** just above a KPI’s chart.

**Resolution limitations:**
Since not every resolution makes sense for every date range, we automatically adjust those that don’t offer a useful visualization.

For example, if you select a date range of a year and a resolution of **Hours**, the result would not be very insightful on any common screen. Therefore, our system automatically switches to a resolution of **Months**.

The automatic adjustment follows this schema:

  • Date ranges larger than 7 days default to a resolution of **Days**
  • Date ranges larger than 62 days (ca. 2 months) default to a resolution of **Months**
  • Date ranges larger than 5 years default to a resolution of **Years**

**Notes:**

  • If you’re using bot operators (available from our Business plan), you’ll find their statistics in the designated **Chatbots** section. All filter options and data points are differentiated between human and bot operators.
  • A conversation is considered started only once a first message has been sent.

Conversations

Metrics on your overall chat volume.

Conversations (by initial assignee)
What does it measure?
The number of conversations started in a given time.
Unit
Conversations
When is it counted?
Whenever the first message is sent in a conversation. This will be roughly equivalent to its creation date in the Message Center. There might be discrepancies if, for example, the contact filled out the pre-conversation survey but then waited some time before sending the first message. The value is counted for both assigned and unassigned conversations.
Why is it useful?
Allows you to measure the overall chat traffic and peak times with a focus on individual operators and groups.
Example
  • A contact starts a conversation on your website via the Website Messenger by sending a first message.
  • This metric only tracks operators assigned at the time of the first contact message. It’s not updated when the conversation gets reassigned to another operator. So, conversations that are not assigned at their start (e.g., in offline mode) are not displayed when you filter by a specific operator or group.
  • A contact reaches out to you through Facebook Messenger. The conversation will be counted when they send the first message.

Conversations (by current assignee)
What does it measure?
The number of conversations started in a given time.
Unit
Conversations
When is it counted?
Whenever the first message is sent in a conversation. This will be roughly equivalent to its creation date in the Message Center. There might be discrepancies if, for example, the contact filled out the pre-conversation survey but then waited some time before sending the first message. The value is counted for both assigned and unassigned conversations. Please note that this metric always attributes conversations to the operator and group it is currently assigned to (including “unassigned”), overwriting previous assignees.
Why is it useful?
Allows you to measure the overall chat traffic and peak times with a focus on individual operators and groups.
Example
  • A contact starts a conversation on your website via the Website Messenger by sending a first message. The assigned operator forwards the conversation to a colleague who solves the request and ends the conversation. In your Analytics data, the conversation will be attributed to the second operator and not show up in the first operator’s conversation data.

Conversation session state
What does it measure?
The number of conversation sessions per state (live,offline, re-engage).
Unit
Conversation sessions
When is it counted?
Live: Conversation is started by operator or contact while both are available.

Offline: Conversation is started by contact while operator is unavailable.

Re-engage: Conversation is started by operator while contact is unavailable.
This is only possible if the contact’s email address is saved in their profile.
A conversation is considered started when a first message is sent in it.
Why is it useful?
Allows you to directly compare how often new conversations immediately turn into live interactions and how often either contact or operator is not available at the time.

Average conversation duration
What does it measure?
The average time span between the first message sent in a conversation and the moment its status is set to “Ended” for a certain number of live and asynchronous conversations.

Unanswered conversations are excluded.
Unit
Seconds
When is it counted?
Whenever a conversation with operator messages is set to “Ended”. Resumed conversations are counted anew. If a conversation was handled by human and bot operators, the duration is recorded for each operator type separately.
Why is it useful?
Whether you’re aiming for quick resolutions or ongoing conversations, with this metric you keep track of these goals.
Example A conversation starts and ends 20 minutes later. A second conversation takes 40 minutes, ends, is resumed later on and ends again after another 40 minutes. Considering those two conversations, your average conversation duration is 50 minutes.

Messages
What does it measure?
The total number of messages exchanged across all conversations in a given time. Includes text, voice and media messages.
Unit
Messages
When is it counted?
Whenever a contact or an operator sends a message.
Why is it useful?
Shows you the overall volume of messages you exchanged with contacts. If you divide it by the number of conversations, you’ll get your average number of messages per conversation. The graph shows the number of **messages** (total), **operator messages** and **contact messages**.

Ended conversations resumed
What does it measure?
The number of ended conversations that have been resumed by either contact or operator.
Unit
Conversations
When is it counted? Whenever an ended conversation is resumed by either contact or operator.
Why is it useful?
Shows you how many conversations are followed up on after they were considered finished. A low value usually means that most issues are solved conclusively, suggesting a high service quality. Please note that “Thank you” follow-ups can resume ended conversations, too.

Average sessions per conversation
What does it measure?
The average number of conversation sessions per conversation.
Unit
Conversation sessions
When is it counted? Whenever a conversation session ends, this event is counted and the metric is updated. Only one value is stored for each conversation. Note that conversations can be resumed and ended any number of times.
Why is it useful?
Shows you how effectively you solve your contacts’ issues. A low value usually means that most issues are solved conclusively, suggesting a high service quality.
Please note that conversations in which contact and operator only write each other when the other side is not online will skew this metric because the conversation is ended more often than usual.
Average conversation status duration
What does it measure?
The average time a conversation stays in any active status ('New', 'Open', 'Pending')
Unit
Seconds
When is it counted? Whenever a conversation session ends, the time it has spent in each of the aforementioned status is counted and the metric is updated. Only one value is stored for each conversation. The timestamp will always mark the last time the respective conversation was set to ended.
Why is it useful?
Shows you how long your conversations spend in a specific state, helping you identify potential issues with unresolved cases, for example, if conversations spend a lot of time in the **Open** status.
Average unassigned duration
What does it measure?
The average time a conversation spends unassigned.
Unit
Seconds
When is it counted?
Whenever a conversation is assigned to an operator or ended while no operator is assigned.
Why is it useful?
Shows you how long your contacts have to wait until an operator takes care of them.

Conversation sessions

This section focuses on metrics of individual conversation sessions, namely their length and your response times.

Average live session duration
What does it measure?
The average duration of your live conversation sessions.
Unit
Seconds
When is it counted?
Whenever a conversation session ends while both contact and operator are active at the same time.
Why is it useful?
Gives you the opportunity to ratio live and non-live conversation sessions. Based on this, you can, for example, adjust your service times to become more available for live chats.

Total live session duration
What does it measure?
The total duration of all your live conversations combined.
Unit
Seconds
When is it counted?
Whenever a conversation session ends while both contact and operator are active at the same time.
Why is it useful?
Gives you the opportunity to ratio live and non-live conversation sessions. Based on this, you can, for example, adjust your service times to become more available for live chats.

Live session count
What does it measure?
The number of your live conversation sessions.
Unit
Conversation sessions
When is it counted?
Whenever a conversation session ends while both contact and operator are active at the same time.
Why is it useful?
Gives you the opportunity to ratio live and non-live conversation sessions. Based on this, you can, for example, adjust your service times to become more available for live chats.

First response time
What does it measure?
The time between a contact’s first message and your operator’s first reply in a new conversation session.
Unit
Seconds
When is it counted?
When an operator sends a first reply to a live conversation session while both parties are online.
Why is it useful?
Your first response shows contacts that their request has been acknowledged. A low FRT is a typical indicator of high service quality.
Example A contact starts a conversation by sending a message while an operator is available. 10 seconds later the operator replies. The first response time is 10 seconds.

First response time (service times considered)
What does it measure?
The time between a contact’s first message and your operator’s first reply in a new conversation session.The data is normalized by the service times defined for the Widget.
Unit
Seconds
When is it counted?
When an operator sends a first reply to a live conversation session while both parties are online. This metric only tracks data if service times are enabled for the Widget.
Why is it useful?
Your first response shows contacts that their request has been acknowledged. A low FRT is a typical indicator of high service quality.
Example A contact starts a conversation by sending a message while an operator is available. 10 seconds later the operator replies. The first response time is 10 seconds. If the contact writes 5 seconds before the start of the defined service time and the operator replies after 10 seconds, the first response time is 5 seconds.

Response time live conversation sessions
What does it measure?
The time between a contact’s message and your operator’s reply in a newly started or resumed live conversation.
Unit
Seconds
When is it counted?
When an operator sends a reply to a contact’s message while both parties are online.
Why is it useful?
When operators are available, your contacts expect quick answers. This metric shows you if they get them.
Example Your operator is available and a contact sends a message in a conversation. 10 seconds later the operator replies. The first response time is 10 seconds.

Response time live conversation sessions (service times considered)
What does it measure?
The time between a contact’s message and your operator’s reply in a newly started or resumed live conversation.The data is normalized by the service times defined for the Widget.
Unit
Seconds
When is it counted?
When an operator sends a reply to a contact’s message while both parties are online. This metric only tracks data if service times are enabled for the Widget.
Why is it useful?
When operators are available, your contacts expect quick answers. This metric shows you if they get them.
Example Your operator is available and a contact sends a message in a conversation. 10 seconds later the operator replies. The first response time is 10 seconds. If the contact writes 5 seconds before the start of the defined service time and the operator replies after 10 seconds, the first response time is 5 seconds.

Live conversation sessions forwarded to operators
What does it measure?
Number of forward and reassign actions on live conversation sessions
Unit
Reassignments of conversations
When is it counted?
When an operator reassigns or forwards a live conversation session.
Why is it useful?
Shows where initial routing could be improved.
Example Your operator thinks a colleague is better suited to help out the contact due to specicific product knowlegde and forwards the live session.

Live conversation sessions forwarded to groups
What does it measure?
Number of forward and reassign actions on live conversation sessions
Unit
Reassignments of conversations
When is it counted?
When an operator reassigns or forwards a live conversation session
Why is it useful?
Shows where initial routing could be improved
Example Your operator thinks a colleague is better suited to help out the contact due to specicific product knowlegde and forwards the live session

One-sided sessions
What does it measure?
The number of conversation sessions in which only one side participated.
Unit
Conversation sessions
When is it counted?
Whenever a conversation session is ended and only one side participated (sent a message or took any other deliberate action).
Why is it useful?
Gives you an overview of how many conversation sessions are asynchronous.
Example A contact sends a message outside of your regular service times (1). Your operator replies the next day around noon (2) and the contact in turn replies to that in the evening (3). This would amount to 3 one-sided sessions.

First offline response time
What does it measure?
The time between a contact’s latest message and your operator’s first reply in conversations started while no operator was available.
Unit
Seconds
When is it counted?
When an operator sends the first reply in a conversation started by a contact while no operator was available. Please note: This KPI only considers conversations containing contact details when the Widget is configured to only receive conversations with contact details. Conversations received without contact details aren't counted.
Why is it useful?
When contacts leave messages while you’re not available, they expect you to reply as soon as possible. This metric helps you ensure that they receive timely first responses, even when you aren’t online to reply immediately.
Example A contact starts a new conversation and leaves a message at 10 p.m., which is outside your service hours. Your operator replies at 8 a.m. the next morning. The FRT for this asynchronous conversation is 10 hours.

First offline response time (service times considered)
What does it measure?
The time between a contact’s latest message and your operator’s first reply in conversations started while no operator was available. The data is normalized by the service times defined for the Widget.
Unit
Seconds
When is it counted?
When an operator sends the first reply in a conversation started by a contact while no operator was available. This metric only tracks data if service times are enabled for the Widget.
Why is it useful?
When contacts leave messages while you’re not available, they expect you to reply as soon as possible. This metric helps you ensure that they receive timely first responses, even when you aren’t online to reply immediately.
Example A contact starts a new conversation and leaves a message at 10 p.m., which is outside your service hours. Your operator replies at 8 a.m. the next morning. If the service time for this Widget is defined as 7 a.m. to 5 p.m., the first response time is 1 hour.

Offline response time
What does it measure?
The time between a contact’s latest new message and your operator’s reply in a conversation resumed while no operator was available. The conversation needs to have previous operator messages.
Unit
Seconds
When is it counted?
When an operator sends the first new reply in a conversation with previous operator messages that was resumed while no operator was available.
Why is it useful?
Helps you ensure that your contacts receive timely responses to additional questions, even when you aren’t online to reply immediately.
Example A contact resumes an existing conversation with previous contact and operator messages at 10 p.m., which is outside your service hours. Your operator replies at 10 a.m. the next morning. The response time for this asynchronous conversation is 12 hours.

Offline response time (service times considered)
What does it measure?
The time between a contact’s latest new message and your operator’s reply in a conversation resumed while no operator was available. The conversation needs to have previous operator messages. The data is normalized by the service times defined for the Widget.
Unit
Seconds
When is it counted?
When an operator sends the first new reply in a conversation with previous operator messages that was resumed while no operator was available. This metric only tracks data if service times are enabled for the Widget.
Why is it useful?
Helps you ensure that your contacts receive timely responses to additional questions, even when you aren’t online to reply immediately.
Example A contact resumes an existing conversation with previous contact and operator messages at 10 p.m., which is outside your service hours. Your operator replies at 8 a.m. the next morning. If the service time for this Widget is defined as 7 a.m. to 5 p.m., the response time for this asynchronous conversation is 1 hour.
Offline conversation sessions forwarded to operators
What does it measure?
Number of forward and reassign actions on offline conversation sessions
Unit
Reassignments of conversations
When is it counted?
When an operator reassigns or forwards an offline conversation to an operator.
Why is it useful?
Shows where initial routing could be improved
Example A supervisor assigns a lingering conversation to a different operator.

Offline conversation sessions forwarded to groups
What does it measure?
Number of forward and reassign actions on offline conversation sessions
Unit
Reassignments of conversations
When is it counted?
When an operator reassigns or forwards an offline conversation to an operator group.
Why is it useful?
Shows where initial routing could be improved
Example A supervisor assigns a lingering conversation to a different operator.

Number of conversations received from a chatbot
What does it measure?
The number of times human operators receive conversations from chatbots.
Unit
Reassignments of conversations
When is it counted?
Whenever a chatbot reassigns a conversation. Automatic reassignments at the end of a conversation session are not counted..
Why is it useful?
It shows you how many contact requests require additional human support and how often contacts are willing to continue when the bot was unable to solve their issue. Depending on your setup and expectations, a higher number can be a positive result (e.g. contacts are forwarded to sales representatives who can close the deal) or a negative one (the chatbot could not handle the request on its own).
Example A chatbot is unable to understand and/or answer a request and reassigns the conversation to a human operator.

Quality

This section focuses on metrics to give an indication of the quality of your service. An interesting section to compare different operators and groups with one another.

Satisfaction rating
What does it measure?
The average rating your operators receive from contacts.
Unit
Star rating (from 1 to 5)
When is it counted?
Whenever a contact rates a conversation. Multiple entries per conversation are possible. Ratings submitted while no operator is assigned will be omitted.
Why is it useful?
Gives you quantitative insights into your contacts’ satisfaction with your operators and messaging support.

Feedback count
What does it measure?
The number of times your contacts left feedback.
Unit
Feedback
When is it counted?
Whenever a contact leaves feedback. Multiple entries per conversation are possible. Feedback submitted while no operator is assigned will be omitted.
Why is it useful?
Customers tend to give feedback on particularly good or bad service experiences. So, this metric can give you insights on how common they are.

Unanswered live conversations
What does it measure?
The number of conversations that received no operator reply although contact and operator were simultaneously online at some point.
Unit
Conversations
When is it counted?
When a live conversation reaches the timeout set in the Widget’s settings without an operator’s reply. The count stands even if an operator adds a reply later on.
Why is it useful?
Allows you to see when your operators fail to answer contacts although they’re available. If this happens too often, it might suggest that your operators have too many conversations at once to answer all directly.
Example An operator receives a conversation while available but does not reply before the conversation timeout is reached.

Unanswered ended conversations
What does it measure?
The number of conversations that end without a single operator message.
Unit
Conversations
When is it counted?
When a conversation is set to Ended without having received an operator’s message. The count stands even if an operator replies later on or the contact resumes the conversation.
Why is it useful?
There are legitimate reasons for leaving conversations completely unanswered, for example trolls or contacts that are on the wrong website. However, if the numbers here are especially high, it might be worth checking up on your operators.
Example A contact starts a conversation and doesn’t receive an answer until the conversation is set to Ended.
A conversation can be counted as unanswered in both “Unanswered...” metrics. For example, if a live conversation remains unanswered until its timeout and an operator ends it later on without having replied.
Triggered inactivity prevention actions
What does it measure?
The number of various inactivity prevention actions triggered.
Unit
Inactivity prevention actions
When is it counted?
Whenever an inactivity prevention action is triggered or fails. At least one inactivity prevention action needs to be enabled on a Widget for this metric to receive data.
Why is it useful? Shows you when and how often your inactivity prevention kicks in, and how effective it is.
Example A contact sends a message and the assigned operator does not respond in time. As a result, the inactivity message is sent to the contact.
Offline reassignments revisiting contacts
What does it measure?
The number of times revisiting contacts resume a conversation while the assigned operator is unavailable. The metric is split into two values: one shows contacts who were reassigned to an available operator, the other shows when this failed because no operator was available.
Unit
Inactivity prevention actions
When is it counted?
Whenever a contact resumes a conversation while the assigned operator is not available.
Why is it useful?
Shows you how reliably revisiting contacts are connected to a replacement operator when their assigned one is unavailable.
Example A contact sends a message while the assigned operator is not available. The conversation is reassigned to another, available operator after the time defined in the inactivity prevention settings.
Available inactivity actions
The following inactivity actions are available as events. Their actual occurrence depends on the configuration of your individual Widgets.
New/Open conversation set to unassigned
A conversation remained in the “New” or “Open” status for too long (based on your Widget settings) without the assigned operator reacting to it. To not keep the contact waiting, the respective conversation was unassigned, indicating to all available operators that an action is required.
New/Open conversation reassigned A conversation remained in the “New” or “Open” status for too long (based on your Widget settings) without the assigned operator reacting to it. To not keep the contact waiting, the respective conversation was assigned to a different, available operator.
New/Open conversation not reassigned A conversation remained in the “New” or “Open” status for too long (based on your Widget settings) without the assigned operator reacting to it. Still, the conversation could not be reassigned as none of the other operators had any free chat slots. If this event occurs more frequently, we suggest to increase your overall availability by adding more operators or increasing the existing operators’ number of chat slots.
Inactivity message sent The assigned operator of a live conversation did not reply in time (based on your Widget settings). The respective contact was notified about this.
Live conversation reassigned The assigned operator did not reply in time (based on your Widget settings). To not keep the contact waiting, the respective conversation was assigned to a different, available operator. Please note that this event is also triggered if the event Inactivity message sent was triggered before.
Live conversation not reassigned The assigned operator did not reply in time (based on your Widget settings). Still, the conversation could not be reassigned as none of the other operators had any free chat slots.
Resumed conversation reassigned A contact resumed a conversation while the last assigned operator was not available.
Resumed conversation not reassigned A contact resumed a conversation while the last assigned operator was not available. Still, the conversation could not be reassigned as none of the other operators had any free chat slots.

Chatbots

This section gives you detailed insights into the performance of your bot operators. Most metrics recorded for human operators are also recorded for chatbot operators. Only a few metrics tracking the interaction between human and chatbot operators are recorded separately, such as “Number of chatbot conversations reassigned to human operators”. Note that some metrics related to conversations without operators are by default assigned to the human operator category, such as “Missed opportunities”. This may be important to interpret raw data pulled using our Analytics API.

Chatbot conversation overview
What does it measure?
An overview of several conversation-related chatbot metrics: “Conversations”, “Number of chatbot conversations reassigned to human operators”, “Number of conversations ended by a chatbot” and “Number of timed out chatbot conversations”.
Unit
Conversations
When is it counted?
Depending on respective metric.
Why is it useful?
Provides you with a quick overview of the amount and flow of your chatbot conversations.
Example See descriptions of respective metrics below.

Average conversation duration
What does it measure?
The average time span between the first message sent in a conversation and the moment its status is set to “Ended” for a certain number of live and asynchronous conversations.
Unit
Seconds
When is it counted?
Whenever a conversation with operator messages is set to “Ended.” Resumed conversations are counted anew. If a conversation was handled by human and bot operators, the duration is recorded for each operator type separately.
Why is it useful?
Whether you’re aiming for quick resolutions or ongoing conversations, with this metric you keep track of these goals.
Example A conversation starts and ends 20 minutes later. A second conversation takes 40 minutes, ends, is resumed later on and ends again after another 40 minutes. Considering those two conversations, your average conversation duration is 50 minutes.

Total conversation duration
What does it measure?
The total amount of time your chatbots handled ongoing conversations.

This metric’s visualization compares chatbot data with that of human operators.
Unit
Seconds
When is it counted?
Whenever a conversation with operator messages is set to “Ended.” Resumed conversations are counted anew. If a conversation was handled by human and bot operators, the duration is recorded for each operator type separately.
Why is it useful?
Shows you how much your chatbots relieve the workload of human operators.

To allow you to compare the performance of chatbot and human operators, we display a graph for each operator type in this metric’s visualization.
Example
A conversation starts and ends 20 minutes later. A second conversation takes 40 minutes, ends, is resumed later on and ends again after another 40 minutes. Considering those two conversations, your total conversation duration is 100 minutes.

Average conversation session count
What does it measure?
The average number of conversation sessions your chatbots handled. This metric is displayed in relation to the average number of all conversation sessions led by human operators.
Unit
Conversation sessions
When is it counted?
Whenever a new conversation is created
Why is it useful?
Shows you how often your chatbots step in to support your human operators.
Example Two contacts start a conversation session with a bot operator, another starts one with a human operator. Both values will be displayed in this metric’s chart.

Operator chat slot utilization
What does it measure?
An operator’s average chat slot utilization during a given time in percent.
Unit
Chat slot utilization in percent
When is it counted?
Whenever the number of occupied chat slots changes for an operator.

Why is it useful?
It shows you how busy or idle individual bot operators were during a certain time. A low slot utilization might call for reducing the number of operators while a high slot utilization might suggest raising it.

Conversations
What does it measure?
The number of conversations started in a given time.

This metric’s visualization compares chatbot data with that of human operators.
Unit
Conversations
When is it counted?
Whenever the first message is sent in a conversation. This will be roughly equivalent to its creation date in the Message Center. There might be discrepancies if, for example, the contact filled out the pre-conversation survey but then waited some time before sending the first message. The value is counted for both assigned and unassigned conversations.
Why is it useful?
Allows you to measure your overall chat traffic and identify your peak times.
Example
  • A contact starts a conversation on your website via the Website Messenger by sending a first message.
  • A contact reaches out to you through Facebook Messenger. The conversation will be counted when they send the first message.

To allow you to compare the performance of chatbot and human operators, we display a graph for each operator type in this metric’s visualization.

Post-conversation survey results
What does it measure?
The number of times and when contacts selected certain answers in your surveys after conversations with bot operators.
Unit
Survey answer
When is it counted?
When a contact answers a post-conversation survey.
Why is it useful?
Allows you to gather representative results of your surveys.

Feedback count
What does it measure?
The number of times your contacts left feedback during or after conversations with bot operators.
Unit
Feedback
When is it counted?
Whenever a contact leaves feedback. Multiple entries per conversation are possible. Feedback submitted while no operator is assigned will be omitted.
Why is it useful?
Customers tend to give feedback on particularly good or bad service experiences. So, this metric can give you insights on how common they are.

Satisfaction rating
What does it measure?
The average rating your operators receive from contacts.
Unit
Star rating (from 1 to 5)
When is it counted?
Whenever a contact rates a conversation. Multiple entries per conversation are possible. Ratings submitted while no operator is assigned will be omitted.
Why is it useful?
Gives you quantitative insights into your contacts’ satisfaction with your operators and messaging support.

Number of leads
What does it measure?
The number of times contacts submit their contact details for the first time.
Unit
Number of leads
When is it counted?
The first time a contact leaves contact information like name, email or mobile number.
Why is it useful?
Shows you how effective your chatbots are in generating leads.

Number of chatbot conversations reassigned to human operators
What does it measure?
The number of times a conversation is reassigned to a human operator.
Unit
Conversations
When is it counted?
When a chatbot reassigns a conversation to a human operator.
Why is it useful?
It shows you how many contact requests require additional human support and how often contacts are willing to continue when the bot was unable to solve their issue. Depending on your setup and expectations, a higher number can be a positive result (e.g. contacts are forwarded to sales representatives who can close the deal) or a negative one (the chatbot could not handle the request on its own).
Example A contact navigates through your chatbot questions until the bot has enough information and reassigns the conversation to a designated human operator.

Number of conversations ended by a chatbot
What does it measure?
The number of times a chatbot sets a conversation to “ended”.
Unit
Conversations
When is it counted?
Whenever a conversation with chatbot messages (and none by human operators) is set to “ended” by a chatbot.
Why is it useful?
It shows you how effective your chatbots are in successfully handling contact requests without any help from humans.

Number of timed out chatbot conversations
What does it measure?
The number of times conversations time out while contacts interact with chatbots.
Unit
Conversations
When is it counted?
Whenever conversations with chatbot operators reach the bot conversation timeout.
Why is it useful?
A high value may indicate that many contacts abandoned their conversations because your chatbot was unable to handle their requests. This can point you to necessary tweaks on your chatbot setup.
Example A contact starts a conversation on your website and your chatbot can neither answer the question nor reassign the conversation. The contact loses interest and leaves the website, the conversation runs into its defined timeout.

Messenger availability

The **Messenger availability** shows the status of your Widgets on a timeline so you can determine when exactly your Widget was available for your customers. When you hover over a segment in the timeline, a tooltip with detailed information will show up.

Widget status
What does it measure?
A Widget’s availability status during a given time. The three possible status are online, offline and chat slots full.
Unit
Widget availability status
When is it counted?
Available (green): At least one operator is available.

Unavailable (light red): No operator is available.

Chat slots full (red): At least one operator is online, but all operators are fully occupied and can’t receive live conversations.
Why is it useful?
It shows you how busy your team was and thus how easy it was for contacts to start a live conversation on your Widget during a certain time. Long time spans of full chat slots might call for adding staff to satisfy the demand.
Please note:
If you edit the Widget’s configuration while operators allocated to it are logged in to the Message Center, this can lead to distorted values in the day’s timeline. We recommend making Widget edits while no operators are connected to the Message Center.

Operator status

This section gives you detailed insights into the daily activities of your operators.

The **Operator status** shows the status of your operators in segments on a timeline in different colors for each status. When you hover over a segment in the timeline, a tooltip with detailed information will show up.

Operator status
What does it measure?
An operator’s availability status during a given time. The four possible status are online, offline, away, and full chat slots.
Unit
Operator availability status
When is it counted?
Online (green): The operator is online, the only status in which operators are available for incoming live conversations.

Offline (black): The operator is offline.

Away (orange): The operator’s status is set to away.

Chat slots full (red): The operator is online but fully occupied until one of the ongoing conversations is ended or times out.
Why is it useful?
It shows you how busy or available individual operators were and thus how easy it was for contacts to start a live conversation with them during a certain time. Long time spans of full chat might call for adding staff to satisfy the demand.

Chat slot utilization

The **Operator slot utilization** gives you insights into how busy your operators are with the amount of incoming chats. It shows you the percentage of your operators' occupied chat slots at a given time.

Operator chat slot utilization
What does it measure?
An operator’s average chat slot utilization during a given time in percent.
Unit
Chat slot utilization in percent
When is it counted?
Whenever the number of occupied chat slots changes for an operator.

Why is it useful?
It shows you how busy or idle individual operators were during a certain time. A low slot utilization might call for reducing the number of operators while a high slot utilization might suggest raising it.

Goals, topics, surveys

This section covers the goals you reached in your conversations, the topics your contacts reach out to you for and how they respond to your surveys. These insights allow you to organize your service more effectively.

Goals reached
What does it measure?
When and how often you reached certain goals during live conversations.
Unit
Goals
When is it counted?
Whenever a goal is reached during a conversation. A goal is registered as reached when a contact visits the designated URL. Goals are only tracked if reached by contacts while an operator is assigned and at least one message was sent in the conversation.
Why is it useful?
Allows you to track the goals you defined for a Widget. This gives you insights into your service’s effectiveness.
Example You set up a Widget goal that’s reached when contacts visit your pricing page. When your contact navigates or is pushed to the page, the goal is recorded as reached.

Conversation topics
What does it measure?
The number of times and when a certain topic or a combination of topics was assigned to a conversation. Please note that the conversation must have already been started by a first message. Changing the topic of a conversation at a later time will update this KPI. Any newly assigned topics will replace the previously assigned ones in your data.
Unit
Topic assignment
When is it counted?
When a conversation is tagged with one or several topics.
Why is it useful?
Shows you which topics your contacts are more and less interested in.

Pre-conversation survey results
What does it measure?
The number of times and when a contact selected a certain answer in your survey before a conversation.
Unit
Survey answer
When is it counted?
When a contact answers a pre-conversation survey. Note that the KPI is recorded when the contact completes the survey, but will only be displayed once the conversation appears in the Message Center, even if this happens at a later point in time.
Why is it useful?
Allows you to gather representative results of your surveys.

Post-conversation survey results
What does it measure?
The number of times and when a contact selected a certain answer in your survey after a conversation.
Unit
Survey answer
When is it counted?
When a contact answers a post-conversation survey.
Why is it useful?
Allows you to gather representative results of your surveys.

Missed opportunities

Missed opportunities
What does it measure?
The number of times a contact tries to start a new conversation while no operator is available.
Unit
Conversations
When is it counted?
Whenever a contact sends a message in a newly started conversation and the conversation reaches its timeout, meaning the operator didn’t reply. It’s also counted when no operator is available at all and the Widget’s offline message is displayed to the contact, regardless of whether they send a message
Why is it useful?
Shows you how often contacts want to speak with you while no operator is there to respond. Too many of these might suggest extending your service times, team or chat slots.
If missed opportunities turn into conversations because the contact sends an offline message, their previous count as missed remains.

Contacts

In this section you can see details about your contacts, for example where they are from or what Browser they are using.

Unique visits
What does it measure?
The number of unique visitors you had conversations with.
Unit
Unique visits
When is it counted?
When a unique visitor is identified via cookie and starts a conversation. As soon as a conversation reaches its timeout, a unique visit is counted. This can happen once every 30 minutes and is based on the UUID assigned to a visitor.
Why is it useful?
Shows you how likely unique and potential first-time visitors are to use your messaging service.

Browsers
What does it measure?
The number of times your contacts use a certain browser when they start or resume a conversation.
Unit
Browser type
When is it counted?
When a contact starts or resumes a conversation, the browser they’re on is counted. This metric is suppressed if the conversation never appears in the Message Center.
Why is it useful?
Shows you which desktop or mobile browsers your contacts are most commonly on when using your messaging service.

Countries
What does it measure?
The number of times contacts located in a certain country start or resume a conversation.
Unit
Country
When is it counted?
When a contact starts or resumes a conversation, the country they’re located in is counted once. This metric is suppressed if the conversation never appears in the Message Center. Note that the metric will be registered for the operator who is assigned at the end of the conversation session (or no operator, if the conversation is unassigned at this time).
Why is it useful?
It shows you the geographic locations your contacts come from, which can also inform you about their language preferences preferences.

Contact response time to offline replies from operators
What does it measure?
The time between an operator’s latest new message, sent while the contact is offline, and your contact’s reply to it.
Unit
Seconds
When is it counted?
When a contact replies to a conversation in which the operator has sent the last message while the contact was offline.
Why is it useful?
Shows you how long it takes for contacts to follow up on answers they received while offline. This allows you to estimate when your operators will have to reply again and plan accordingly.

Contact response time to offline replies from operators (service times considered)
What does it measure?
The time between an operator’s latest new message, sent while the contact is offline, and your contact’s reply to it. The data is normalized by the service times defined for the Widget.
Unit
Seconds
When is it counted?
When a contact replies to a conversation in which the operator has sent the last message while the contact was offline. This metric only tracks data if service times are enabled for the Widget.
Why is it useful?
Shows you how long it takes for contacts to follow up on answers they received while offline. This allows you to estimate when your operators will have to reply again and plan accordingly.

Analytics dashboards

Analytics dashboards allow you to compile your own custom view with up to ten of the KPIs we offer. Build a dashboard featuring the KPIs you’re most interested in or individual dashboards for specific purposes, like operator statistics or a single Widget’s data. Select one of your dashboards as your default and it will always be the first thing you see when you open **Analytics > Dashboards**.

Click **Edit dashboards** to open an overview from which you can add, edit, clone and delete Analytics dashboards. Note that you cannot delete a dashboard as long as it is marked as your default.

In this view, you can define a name for your new dashboard and add KPIs. To add a KPI, click the “Add KPI” button in the KPI list and select its “KPI type”. You can drag KPIs up or down to specify their position in the dashboard. If you leave a KPI's default name unchanged, it will be displayed to all operators in their respective profile language. So, if you have a multilingual team, it may be better not to customize the names. To adjust a KPI’s name, just double-click it:

Click the cog symbol to the right of a KPI to configure its settings in the filter view below:

All settings are directly applied to your dashboard view, giving you a great range of options to configure it for your individual reporting needs.

Example conversation flow

Common scenarios and the metrics they impact.
For all scenarios below, we assume a Widget staffed with a firewall chatbot (always online, receives all new conversations) and privacy notice enabled.
Action Affected metrics
A contact arrives at your website and starts a new conversation. The privacy notice is displayed. No events triggered.
Your contact accepts the privacy notice. The conversation is set to its default topic and then assigned to the firewall chatbot.
“Conversation topics” - since the default topic is always set before a conversation is routed to an operator its value is recorded without any operator association (neither bot nor human).
“Operator status” - whenever the assigned operator’s number of free chat slots changes, their status is checked and recorded, in this case for a chatbot. As with all events triggered in a conversation, the respective conversation ID is attached to the event and can be extracted via the Analytics API.
“Operator slot utilization” - whenever the assigned operator receives a new conversation, their number of free chat slots changes. The new utilization is recorded, in this case for a chatbot.
The chatbot sends a message with selection buttons to the contact who clicks one of them. No events triggered.
Meanwhile a human operator goes online.
“Operator status” - since the status change is a result of the human operator coming online, the value is recorded without a conversation ID.
“Operator slot utilization”
The chatbot is unable to answer the requests and reassigns the conversation to “any” other operator. The operator who just went online has available slots and now receives the conversation.
“Operator status” and “Operator slot utilization” - recorded because the reassignment of the conversation increases the chatbot’s number of free chat slots.
“Conversation session state” - recorded for the newly assigned operator.
“Operator status” and “Operator slot utilization” - recorded because receiving the conversation increases the human operator’s number of free chat slots.
“Number of chatbot conversations reassigned to human operators” - recorded for the chatbot performing the reassignment.
“Number of conversations received from a chatbot” - This metric is recorded for the operator receiving the conversation. Please note that the conversation may also be set to “unassigned” if no operator is available to receive it.
Contact and operator chat with each other. No events triggered.
The operator ends the conversation. The contact is presented with the feedback and rating screen.
“Average conversation status duration” - recorded for the time during which the conversation held the status “new”. For simpler data interpretation, this value is recorded for the entire conversation session and is attributed to the last operator assigned, in this case a human.
“Average conversation status duration” - recorded for the time during which the conversation held the status “open”.
“Average conversation duration” - the time the human operator spent chatting with the contact.
“Average conversation duration” - the time the chatbot operator spent chatting with the contact.
“Operator status” and “Operator slot utilization” - recorded because ending the conversation increases the human operator’s number of free chat slots.
The contact rates the conversation.
“Satisfaction rating” - recorded for the currently assigned operator.
The contact leaves the website and as a consequence, the conversation session is ended.
“Conversation session state” - recorded for the currently assigned operator. Since no new session was started, the last “Conversation session state” data point is updated.
“Messages” (contact) - the number of messages the contact sent to each operator. Since the human operator sent the first actual message, thereby starting the count of this metric, no events of this type are recorded for the chatbot.
“Messages” (operator) - the number of messages each operator sent to the contact. Because the chatbot did not write any messages, but only presented buttons, no events of this type are recorded for the bot.
“Messages” (sum of messages between certain operator and contact) - also only recorded for the human operator.
“First response time live conversation sessions” - because the first message-based interaction was not with the chatbot, this event is recorded for the human operator.
“First response time live conversation sessions (service times considered)”
“First response time live conversations”
“First response time live conversations (service times considered)”
“Conversations (by initial assignee)” - recorded for the chatbot since it was the operator first interacting with the contact.
“Conversations (by current assignee)”
“Average live session duration” and “Average conversation session count” - recorded for the first operator assigned to the conversation session, in this case the chatbot.
“Number of conversations ended” - recorded for the operator ending the conversation.
“Unique visits” and “Conversation session state upon start” - recorded for the currently assigned operator.