The Agent Desk is the principle interface for your organization’s agents to engage with visitors using Unblu. It also provides features for Supervisors to monitor and organize agents and teams.
This guide aims to give you a good working knowledge of the Agent Desk’s features. Once you’ve read through this guide, you will be able to focus on engaging with visitors, instead of trying to work out how to do so.
|Depending on the configuration your organization chose to implement, the look and feel of the interface may differ from the pictures in this guide. Additionally, some functions may not be available if they have not been licensed by your organization. However, you should have no difficulties applying the explanations in this guide to your particular circumstances.|
To access the Agent Desk, agents must be authenticated. Unblu offers a variety of ways to authenticate users. One way is a traditional login form that asks for your username and password:
Your organization may have opted for a different authentication method. For further information on the various authentication methods available, please consult the authentication page.
The Agent Desk is made up of three areas:
The top bar contains a number of control elements. It also holds tabs of all of the conversations you are currently participating in.
The main content area takes up the lion’s share of the screen. It is where conversations, collaboration layers, and other content are displayed.
The top bar provides a number of useful features.
The hamburger icon can be used to hide the navigation menu on the left.
The plus icon can be used to initiate various types of conversation.
Next to the plus icon is a tabbed list of your open conversations. The active conversation panel is highlighted by an orange bar along the bottom edge of the tab. Any conversations with unread chat messages display a badge.
In the example above, the agent is participating in two conversations. The conversation with Luc is currently displayed in the main content area, and there are two unread messages in the conversation with Viktor.
With automatic request dispatching enabled, the top bar contains a toggle that allows agents to signal when they are available to accept incoming conversations.
Finally, your avatar displays your current online status and provides access to your Agent Desk settings.
In the picture above, the green badge on the avatar indicates that the agent is online.
As mentioned above, the plus icon in the top bar allows you to start various types of conversation:
Join, Join screen, and Join mobile all open a modal dialog showing a PIN that you can share with a visitor to launch a conversation in Unblu:
A PIN must be redeemed within a certain time period. The dialog includes a countdown showing how much time remains to redeem the PIN.
If the PIN isn’t redeemed it expires. The dialog below appears when this has occurred:
It’s always possible to generate a new PIN by pressing the Generate new PIN button. Doing so automatically invalidates the previous PIN.
Show and Show Screen both open a conversation panel with a new conversation. In both cases, you have to invite users to join you with the invite icon .
The type of conversation that’s created when a user enters the PIN depends on which option you selected in the plus icon menu:
Join launches a conversation with embedded co-browsing.
Join screen launches a conversation with screen-sharing on the user’s side. The user will be required to choose whether they want to share a particular application window or their entire screen with you.
If they choose not to share anything with you, the conversation continues as a live chat.
Join mobile launches a conversation with mobile co-browsing.
Show launches a conversation with universal co-browsing on the agent’s side. Just like other users, you can choose what to share in the co-browsing session.
Show screen launches a conversation with screen-sharing on the agent’s side.
|The choice you make in the plus icon menu isn’t final. You can switch between types of conversation after the initial selection. For example, you might select Join to see a visitor’s view of your organization’s website, but later need to launch document co-browsing to review a file together.|
The queue displays all conversations that you could join as the assigned agent. Like in the inbox, the items in the queue contain useful information about each conversation:
The visitor’s name
The last message the visitor entered
Where the visitor started the conversation
How long the visitor has been waiting
The visitor’s approximate location
The operating system and browser the visitor is using
Above the queue, you can see how many conversations are waiting in the queue, and the average time visitors' conversations have been waiting in the queue.
The picture below shows a queue with just one waiting conversation. The toast in the lower left-hand corner of the Agent Desk appears when a request is added to your queue.
The icon to the left of the visitor’s name indicates the type of conversation request.
Audio call request
Video call request
Screen sharing request
Conversation requests addressed to you personally are listed under the heading Assigned to me:
If you’re another agent’s deputy and they didn’t respond to a personal conversation request, the delegated conversation is listed under the heading Delegated to me:
The queue only provides filters for named areas and languages. However, you can select multiple named areas and languages by which to filter your queue.
The following picture shows the filter selection dialog. To confirm your selection click on Apply, then click on OK to apply the filter to the queue.
In addition to the information mentioned above, the queue provides two features that aren’t needed in the inbox: a Preview button and an Answer button.
The Preview button allows you to see what has happened so far in a conversation without actually joining it. The picture below shows what a conversation preview might look like. Note the blue banner beneath the top bar that states You are previewing this conversation.
The Answer button allows you to join the previewed conversation without having to return to the queue view. Alternatively, you can click on the Answer button on the conversation item in the queue.
The Forward menu gives you the possibility to forward the conversation to an agent, team or named area without joining the conversation.
The Close button at the bottom of the screen simply closes the preview.
|In the preview, the buttons for forwarding the conversation , inviting other participants to join the conversation , and ending the conversation are disabled.|
If another agent answers a conversation while you are previewing it, Unblu informs you in a modal dialog and disconnects you from the conversation:
For more information on the options available once you answer a conversation, see the section Conversation panel below.
Automatic request dispatching is a way for your organization to automate the conversation assignment process. Rather than choosing which conversation requests to answer, Unblu reserves conversation requests for you.
If your organization uses automatic request dispatching, the queue will look slightly different from the description above. Your workflow in the queue will also be slightly different.
This is what the queue looks like in the Agent Desk when automatic request dispatching is active:
To the left of your avatar in the conversation bar, you can see a toggle labeled Ready for next conversation. By activating the toggle, you inform the dispatch mechanism that it may reserve conversation requests for you.
In the queue, the conversation will sport a Reserved label. (This label is applied to all reserved conversation requests, not just requests reserved for you.)
There are no buttons to accept or decline the conversation from the queue. There is also no way of previewing the conversation.
If you are available for conversations, the dispatcher will assign you (and your coworkers) conversations in turn, up to your maximum number of active conversations. When it reserves a conversation for you, a toast appears in the lower left-hand corner of the Agent Desk. The toast includes buttons for you to accept or decline the conversation. It also displays a countdown showing how much time remains for you to do so.
If you click on the toast itself, Unblu opens a preview of the conversation.
If you decline a conversation request, or do not react to the reservation before the countdown displayed on the toast reaches zero, your status is set to unavailable. If the reservation timed out, a modal dialog is displayed that allows you to set your status to available again or to remain unavailable.
Once you’ve been authenticated, the Agent Desk opens in the inbox view.
The inbox provides access to all the conversations you are currently participating in. Depending on how Unblu is configured at your organization, the conversations may be grouped in two tabs, My Conversations and Secondary Conversations.
Each conversation listing provides several useful pieces of information about the conversation:
The names of the other participants in the conversation
The status of the conversation’s other participants, i.e. whether they are online or not
The last message of the conversation
The time the last message was transmitted
The number of messages you haven’t read yet, if there are any
In the picture above, the agent is the assigned agent in two conversations, one with Sven, one with Maria. The green badge on each visitor’s avatar shows that they are online. The badge on the My Conversations tab indicates that the Agent has three unread messages in total. There are two unread messages in the conversation with Maria and one in the conversation with Sven.
In the picture of the Secondary Conversations tab below, the Agent Chris Johnson has invited the Agent to join a conversation with the client Viktor Rabe. Chris is online (as indicated by the green badge), but Viktor isn’t. In the top bar of the Agent Desk, you can also see three open conversation views: one with Sven, one with Maria, and one with Viktor and Chris.
The Inbox menu item sports a badge which displays the sum of unread messages in the My Conversations and Secondary Conversations tabs.
To rejoin a conversation, simply click on its listing in the inbox.
For more information on the options available once you answer a conversation, see the section Conversation panel below.
On each of the inbox tabs, above the list of conversations, you see a number of filters. These allow you to limit the conversations you see to those that satisfy certain conditions. If, for example, you were looking for an ongoing conversation with a client who contacted you via WhatsApp, you could use the Channels filter to display only WhatsApp conversations. Alternatively, you could use the Visitors filter to display only the conversations with that particular client.
On the My Conversations tab, the Agents filter can be used to look for secondary agents. On the Secondary Conversations tab, the same filter allows you to search for the assigned agent of conversations that you are participating in as a secondary agent.
Finally, there are three predefined filters:
The Starred filter shows only conversations you have starred with the star icon to the left of the participants' avatars.
The Online filter will only show conversations where the context person is online.
The Unread filter only displays conversations with messages that you haven’t read yet.
The filters can be combined to limit the number of conversations displayed even further.
Scheduled conversations are conversations that you can set up to start at a certain time in the future.
Clicking on the Scheduled conversations entry in the navigation menu on the left opens the overview, which contains two tabs:
The Upcoming tab lists all of the scheduled conversations you organized with a scheduled start time in the future. This includes conversations that you started before their scheduled start time, if the start time has not yet been reached. In the picture below, this applies to the second conversation in the list.
Scheduled conversations with a start time in the past that haven’t started yet are also listed on the Upcoming tab. They are marked with an Overdue badge, like the first conversation in the picture above.
The Ongoing tab lists all of your scheduled conversations with a start time in the past that you have already started.
On each tab, conversations are displayed in chronological order.
On the Upcoming tab, the further in the future a conversation is scheduled to start, the further down the list of scheduled conversations it will appear.
On the Ongoing tab, the more more recently a conversation was scheduled to start, the closer to the top of the list of conversation it will appear.
Scheduled conversations that have ended are accessible from the conversation history, like any other conversation.
To create a scheduled conversation, proceed as follows:
Open the scheduled conversation overview and click on Schedule a conversation. If there is more than one template for scheduled conversations, a dropdown will open where you can choose a template. If there is only one template for scheduled conversations, clicking on the button takes you to the next step.
Selecting a template opens the New conversation fly-in page.
Here, you must enter the following information:
A topic for the conversation
The date and time the conversation is scheduled for
You can change of all these elements after creating the conversation.
Click on Save to complete the creation process. This closes the fly-in page and displays a modal dialog where you can open the conversation without starting it.
The modal dialog includes a public link that you can send to multiple visitors.
To create personal invitations for individual participants or edit the details of the conversations, click on Open. This opens the conversation in the main area of the Agent Desk.
Alternatively, close the dialog window. Doing so will return you to the Upcoming tab of the scheduled conversation overview.
If you want to edit any of the information you entered when setting up the conversation, click on the information icon . This opens the Conversation details fly-in page.
Next, click on the Edit button at the bottom of the fly-in page. You can now change the topic and the scheduled start date and time of the conversation, provided your participant type is authorized to do so. By default, this is only the case for the conversation’s assigned agent.
You create invitations for scheduled conversations the same way way as for other types of conversation:
Email invitations generate an email with a link that opens the scheduled conversation in the Visitor Desk.
Private link invitations generate a link that you can send to the invitee by email, add to a shared calendar appointment, or provide them some other way. Again, the link will open the scheduled conversation in the Visitor Desk.
|Since visitors can only redeem PIN invitations in the integrated visitor UIs, but scheduled conversations open in the Visitor Desk, you can’t use PIN invitations for scheduled conversations.|
Public links open the Visitor Desk on the Request admission page:
Once the visitor has filled in their details and clicked on the Join button, they must wait for an agent to review their admission request. If no agent has the conversation open in the Agent Desk, the page will ask them to wait for an agent to join:
If an agent is viewing the conversation, the visitor will be asked to wait for the agent to admit them to the conversation:
In the Agent Desk, a modal dialog lists the visitors who have requested to be admitted to a conversation:
If more than one visitor wants to join the conversation, you can grant or deny each request individually.
Visitors who redeem personal invitations are automatically admitted to the conversation.
If you invite another agent to join a scheduled conversation as a secondary agent, the invitation will appear in their queue. Once they click on Answer, the conversation will open, just like any other conversation they answer.
If a secondary agent closes the conversation tab and wants to return to it later, they’ll find it in the Ongoing tab in the scheduled conversations overview, not in their inbox.
Scheduled conversations don’t start automatically. An agent who is participating in the conversation has to start it explicitly.
To start a conversation, open the conversation by clicking on it in the overview, then click on the Start button in the upper right-hand corner of the screen:
Alternatively, open the Conversation details fly-in page and click on the Start button at the bottom.
Invitees who try to join a conversation before you start it are shown the Waiting room page in the Visitor Desk. The page includes a message asking them to wait until the conversation has started:
Once an agent starts the conversation, visitors see the usual UI elements, such as the chat view and collaboration layers.
The conversation history provides access to all of your conversations. It includes both ongoing conversations and those that have already ended.
The list items in the conversation history show a lot of information about the conversations:
The date the conversation was created
The type of conversation, i.e. chat, video call, audio call etc.
The name of the assigned agent
The name of the context person or primary visitor
The state of the conversation, e.g. Active, Ended, Offboarding etc.
How long the conversation lasted
How long the visitor had to wait for an agent to answer their request
How the visitor rated the conversation
Here is an example of what an agent’s conversation history might look like:
In the picture above, most conversations have ended. The agent, Charlotte, was the assigned agent of all but one of the conversations. The conversation that Charlotte is participating in as a secondary agent is still active.
The Export data button at the bottom of the list allows you to export the data as a CSV file.
Clicking on a conversation displays its contents. See Viewing a conversation below for more details.
Like the inbox and the queue, the conversation history provides various filters to specify which conversations should be listed. You can filter by
|A conversation will be displayed even if its start and end times were outside the display range. For example, if a conversation started at 09:00 and ended at 12:30, setting the display range to 10:00—11:00 will return the conversation, even if there was no activity in the conversation between 10:00 and 11:00.|
The display range filter is always displayed. You can specify which other filters to show by clicking on the more icon . In addition, the filter field allows you to enter text to search for in the conversations' information.
The maximum number of conversations displayed in the history is defined in the configuration property com.unblu.conversationhistory.limit. The default value is 100.
Clicking on a conversation’s information icon opens a fly-in page with additional details:
Besides an overview of the conversation’s details, the modal page provides information on the conversation’s participants:
Who the participants are or were
When they joined and left the conversation
What the nature of their participation in the conversation was
Whether they are still participating or not
How they rated the conversation
The information on the modal page can be exported as a CSV file using the Export data button.
The Open button displays the conversation’s contents.
When you open a conversation in the conversation history, the conversation is opened with the agent participating as a ghost. This is the case even if the conversation is still active and the agent is the conversation’s assigned agent. In this view, you can see everything that has taken place in the conversation so far, including the entire onboarding process. However, the view is not shown next to the plus icon at the top of the Agent Desk.
(Supervisors and admins can also look in on agents' conversations as ghosts.)
Contrast this view with the same conversation after clicking on the Open in normal view button at the bottom of the screen:
Since the conversation depicted above has not ended, the text input field is available at the bottom of the screen. The Agent can thus continue interacting with the visitor from this view.
Supervisors can monitor agents on teams they are responsible for.
Administrators can monitor any agent of the Unblu account they administer.
As you can see above, the content area of the agent monitor is divided into four panels:
- Service Status panel
The service status panel shows the capacity utilization by named area and language.
- Agent Availability panel
The agent availability panel shows each agent’s capacity and their availability.
- Active Sessions panel
The active sessions panel provides information about the active conversations, including the type and duration of the conversation, and the agent handling the conversation.
- Filtered Chat Queues panel
The filtered chat queues panel shows the filters that agents are applying to their queues. This can be helpful when trying to determine why a certain conversation is not being accepted by any agents: maybe none of them can see it because of the filters they have set in their inbox.
You can sort the contents of any panel according to any of the columns.
The agent monitor distinguishes between three different levels of availability: Available, Busy, and Unvailable.
In the service status panel, availability refers to all of the agents assigned to the various named areas. The numbers in brackets after the availability status show the total current capacity utilization for all of the agents.
In the picture above, the named area Banking has an available capacity of five conversations and a total capacity of six conversations. Because there are agents with capacity available to accept conversations in this named area, the availability status is Available.
The named area Credit Cards, on the other hand, has an available capacity of zero conversations and a total capacity of one conversation. Its availability status is therefore Busy.
The situation is even worse for the named area Auto. Its availability status is Unavailable. This means that there are no agents logged in at present who could accept conversations in this named area.
The availability information in the agent availability panel refers to each agent’s personal availability. The numbers in brackets show their personal current capacity utilization.
In the picture above, the agent Karl Dangerfield has an available capacity of zero conversations and a total capacity of one conversation.
An agent’s personal availability status can be affected in a number of ways:
If an agent sets their online status to Away in the Agent Desk settings, their availability status switches to Unavailable.
If an agent has reached the maximum number of concurrent active sessions allowed, their availability status switches to Busy.
If you’re using automatic queue dispatching with priority band busy limits, the agent availability only reflects an agent’s availability status for priority 1 requests.
If automatic queue dispatching is active, agents can change their availability status independently of their online status. Check our article on automatic queue dispatching for further information.
Selecting an entry in any of the four panels will filter out the entries in the other panels that are relevant to the entry selected.
In the picture above, the agent Karl Dangerfield has been selected in the agent availability panel. This is indicated by the vertical orange bar to the left of the entry. As a result of this selection, we can see at a glance which entries in the other panels apply to this agent. Entries that don’t apply to him are grayed out.
The effect of clicking on Answer for a conversation in the queue depends on the type of conversation request you are answering. As a rule, the Agent Desk opens the most appropriate view for the request at hand. For example, if you respond to a live chat request, the conversation panel will display the messaging view.
In the upper left-hand corner of the conversation panel, you will see one or more avatars representing the other participants in the conversation. Clicking on a participant’s avatar displays a dropdown containing information on that participant:
In the image above, you can see the following information in the About section of the popup:
The user’s approximate location. This information is determined based on the user’s IP address.
In the image above, the user is located in Zurich, Switzerland, or thereabouts.
The user’s language, which in this example is English.
The operating system and browser the user’s running. In this case, the user’s computer is running version 95 of Mozilla Firefox on Ubuntu Linux. The version of Ubuntu is unknown, so it’s referred to as "Other".
The user’s email address.
The user’s phone number.
The user’s propagated username or their user ID next to a closed lock icon .
The Remove participant button allows authorized users to remove individual participants of a conversation from that conversation.
If a user hasn’t been authenticated, the popup displays an open lock icon with the label Unauthenticated:
The picture above doesn’t include the user’s email address and phone number. However, this information may be available for unauthenticated users, too, for example if they provided this information during the conversation xref:../docs/glossary/onboarding.adoc#,onboarding process.
The popup shows device-related information for each device the user is currently accessing the conversation with. This means that there can be multiple entries for both the operating system and browser as well as the language. Because of the method for determining a user’s location, there may even be multiple entries for their location.
The collaboration layers a participant can provide are listed in a section entitled Collaboration options beneath the About section. Ongoing collaboration layers are highlighted in the list. In both pictures above, there’s a single embedded co-browsing session running, and the user in question isn’t providing any other collaboration layers such as screen sharing.
Some controls are available in the conversation panel regardless of the conversation type. These controls are located in the agent conversation bar between the top bar and the action bar of the view opened to interact with the visitor.
The controls here affect the participants or the type of collaboration taking place in a conversation.
The forward conversation icon opens the following menu:
Selecting one of the menu items opens a modal page where you can select whom exactly the conversation should be forwarded to. The following picture shows the modal page after selecting Forward to team in the menu.
Once you click on the person, team, or named area that you want to forward the conversation to, a text field will appear where you can add a message for them. This message is optional, and it is not visible to visitors in the conversation.
Once you have finished your message, click on Forward.
When another agent answers the forwarded request, you automatically leave the conversation. For further details about what this entails for you, please review the documentation on the conversation life cycle.
If you try to forward a conversation that you’ve already forwarded, the following dialog appears:
If you select OK, the modal page will open and you can select a different user to forward the conversation to. The agent, team, or named area you originally forwarded the conversation to will not be selectable. They may, however, be shown beneath the list of potential recipients:
If you originally forwarded the conversation to an agent and now want to forward it to a different agent, the original agent will be displayed beneath the list of potential selections.
If you originally forwarded the conversation to a team and now want to forward it to different team, the original team will be displayed beneath the list of potential selections.
If you originally forwarded the conversation to a team and now want to forward it to an agent, the agents in the team you forwarded the conversation to will be displayed beneath the list of potential selections.
If you already invited an agent or a team to join the conversation, you can still select them to forward the conversation to them instead. Customer invitations are not listed.
The invite icon opens the following menu:
As with the forward conversation menu, selecting either Invite an agent or Invite a team opens a modal page where you can choose whom to invite and add a message to include with your invitation.
If you select Invite a customer, the modal page looks different:
You can choose whether to invite customers by generating a PIN that you can give them, or by sending them an email with a link, and you can mix the two invitation types if you want to invite more than one customer. It is also possible to issue more than one invitation without the modal page closing.
As you can see in the example above, by default, email invitations do not expire as quickly as PIN invitations. The invitations that have already been generated are listed on the modal page, along with a countdown to each invitation’s expiry time and a link to revoke the invitation. When an invitation expires, its Revoke link is replaced by a Renew link which lets you generate a new invitation.
The Collaborate menu provides access to the various ways for you to collaborate with other participants in a conversation.
Browser tab of <visitor> is used to initiate an embedded co-browsing session. The browser tab shared is that of the visitor in question.
Websites initiates a universal co-browsing session
Documents initiates a document co-browsing session
My screen initiates a screen sharing session. The screen shared is the agent’s.
Screen of <participant> initiates a screen sharing session. The screen shared is that of the participant in question.
Note that you can collaborate on any participant’s screen, but only on a visitor’s browser tab.
The participants' names and avatars in the menu are particularly important if there are more than two participants in a conversation. The extra information helps to ensure that you don’t start a collaboration with the wrong contents.
The picture below shows the collaborate menu from a conversation involving two visitors and two agents:
If you are already collaborating with someone, a blue dot to the left of the appropriate icon indicates which type of collaboration is currently in progress.
Selecting one of the menu items does not immediately launch a co-browsing or screen sharing session. It sends a collaboration request to the context person>> and to the participant whose content would be shared.
Suppose that in the conversation pictured above, Sven is the context person and Ren is a secondary visitor. If you want to share Ren’s browser tab, then both Ren and Sven will receive a collaboration request. This is what the request looks like in the floating visitor UI:
Both Ren and Sven must give their consent for the embedded co-browsing session to go ahead. If they fail to do so within a certain time, the collaboration request times out.
In the Agent Desk chat panel, you can cancel their collaboration request:
Once collaboration starts, the main content area of the Agent Desk displays the appropriate collaboration layer in the collaboration space. The following picture shows the collaboration layer of an embedded co-browsing session:
The type of collaboration you are currently engaged in is visible in three different places:
If you open the collaborate menu, a blue dot next to the appropriate icon indicates which type of collaboration is currently taking place.
The icon of the current type of collaboration is displayed to the left of the collaborate menu.
The icon and a brief description of the current type of collaboration is shown immediately above the collaboration pane’s main area.
Besides the hint about the type of collaboration in progress, the bar at the top of the collaboration space contains two controls.
The maximize content icon on the left hides the top bar, the agent conversation bar, and the sidebar. This increases the space available for the collaboration space.
When the content of the collaboration space is maximized, the icon is replaced by the minimize content icon . Clicking the icon displays the top bar and the agent conversation bar. It doesn’t reopen the sidebar.
The Stop button on the right allows you to stop collaborating. Doing so doesn’t end the conversation.
To the left of the collaboration pane is the layer action bar. It gives you access to different collaboration tools.
Toggle mark mode. This allows you to highlight areas of the screen for the other participants.
To mark an area in mark mode, press and hold your primary mouse button in one corner of the area you want to highlight, then drag the cursor to the opposite corner of the area. While you’re doing this, Unblu shows a dashed rectangle around the area selected.
Once you release the mouse button, Unblu highlights the area you selected.
If you just want to point to a particular feature of the page or document in mark mode, click on it. Unblu displays a pulsating circle to draw the other participants' attention to it.
Toggle remote control of the cursor in the co-browsing layer.
The remote control tool gives you control over the mouse pointer in the collaboration layer. You can click on page elements and fill in text fields, provided they aren’t protected.
While somebody other than the person sharing the collaboration layer has control of the mouse pointer, their display name appears next to the pointer.
Turn scroll lock on or off for the co-browsing layer.
Scroll lock ensures that your view of the co-browsing layer remains fixed, even if the other participant scrolls up or down on the page you’re viewing together.
Mark mode and remote control mode may be configured to require the approval of one or more participants in a conversation. If that’s the case, they’re automatically asked for their approval when you attempt to activate the tool in question. You can see the approval message in the list of messages, and you can cancel the request there, too, but you can’t approve the request on someone else’s behalf.
|Which collaboration tools are available depends on the type of collaboration and the configuration of your Unblu installation.|
To the right of the collaboration space, you can see the conversation’s messaging view. You can collapse it to give the collaboration pane more room by clicking on the hide chat panel icon . If you want to expand the chat panel, click on the show chat panel icon .
The End conversation button does exactly that: it ends the conversation. Depending on how Unblu is configured, it may display a dialog asking you whether you really want to end the conversation. The control is only available to users entitled to end the conversation. Who may end a conversation can be configured in the conversation templates and in the conversation itself.
Once a conversation has ended, you can no longer add content to it. You can review conversations that have ended in the conversation history.
For further details about conversation states, please consult the documentation on the conversation life cycle.
It is possible to collaborate in different ways at the same time. For example, a visitor could share their browser tab with you while you share your screen with them. You can only ever see one of the collaboration layers at any one time. However, running more than one collaboration layer simultaneously makes it easier to switch between them.
When multiple collaboration layers are active in a conversation, you will see icons for each collaboration layer to the left of the collaborate menu in the conversation top bar. The collaboration layer currently in view is highlighted:
In the example above, there are two collaboration layers, an embedded co-browsing layer and a screen-sharing layer. The screen-sharing layer is the one currently visible in the collaboration space: its icon is highlighted with a shade of blue.
If the visitor requested a live chat, an audio call, or a video call, the appropriate view will open when you answer the request. The views in the Agent Desk are almost the same as the ones in the floating visitor UI. The main differences are that there is no option to pop the conversation into a separate window, and that the Agent Desk provides canned responses.
For more information on the other view controls, please consult the Floating Visitor UI guide.
Canned responses are predefined text snippets that agents can use in live chats. They can include predefined placeholders for the agent’s and visitor’s names, as well as custom placeholders that must be replaced by the agent. Canned responses can be defined at the account, team, and user level.
To use a canned response in a conversation, type the appropriate keyboard shortcut in the text input field. Alternatively, press the Canned response icon adjacent to the field. Both actions open the canned response dialog above the text input field:
The dialog lists all of the canned responses available to the agent in the same language as the current conversation. The list is arranged in alphabetical order by title. Titles need not be unique, so each item also displays the beginning of the text of the canned response. This makes it easier to distinguish between canned responses.
The filter box above the list allows the agent to search through the titles of the canned responses. It doesn’t search through the texts of the canned responses.
Canned responses with one or more custom placeholders have an arrow to their right. When you select such a canned response, the entire text of the canned response is displayed in the dialog, and you can replace the custom placeholders with content suited to the conversation.
There are two types of custom placeholders: text placeholders and multiple-choice placeholders.
Multiple-choice placeholders are shown as buttons with the label Select and the placeholder title. When you click on the button, the available options are listed for you to select from.
Text placeholders appear as boxes in the text. The box contains the hint defined for the canned response. When you select a text placeholder, the hint disappears and is replaced by the text you enter.
As you can see in the picture above, you can use the same placeholder more than once in a canned response. When you select or enter a value for the first occurrence of a placeholder, the same value is applied to all of the other occurrences of the same placeholder. You can only assign a value to the first occurrence of the placeholder. By default, the first occurrence of a placeholder is pale blue. Subsequent occurrences are gray.
Once you are satisfied with the text, click on Add to insert the text in the text input field. The button will only be available if you have assigned a value to every placeholder.
You can continue to edit the text, either manually or by adding another canned response.
|If you want to add a new line, use Alt+Enter or Shift+Enter.|
When you have finished editing the message, press the Enter key or click on the Send message icon .
If you decide to start collaborating during a call, the visitor will have to confirm they consent to it. They will receive a message that they must confirm in the chat panel. Although the chat panel icon in the floating visitor UI will show a badge, visitors might not realize they have to go there before they can start collaborating with you.
Refer to the relevant section of the floating visitor UI guide for further details.
Clicking on the avatar in the upper right-hand corner of the Agent Desk opens the Agent Desk settings menu:
At the top of the menu, there is a brief summary of the agent’s details: their avatar, name, online status, and the team they’re in. The status shown here is mirrored by the badge in the lower right-hand corner of the avatar.
In the picture above, you can see that the agent logged in to the Agent Desk is called Charlotte Williams. She’s a member of the Call center team and is currently online.
The next element in the menu is a switch to toggle between the statuses Online and Away. The availability status you set here affects your status: if you toggle the switch to Away, your status will be changed to Unavailable, and the badge on your avatar will turn yellow to reflect this change.
Beneath the availability switch is a link labeled Set status message. If you click on the link and enter a message in the text field that appears, the message will be displayed with your status in the secondary conversations tab in the inbox and in the user details popup when somebody clicks on your avatar in a conversation.
Once you enter a status message, the link is replaced by one labeled Edit status.
|The status message appears in visitor interfaces like the floating visitor UI, too.|
The last entry in the first section of the settings menu shows your notification status. The agent in the screenshot above currently receives notifications. She’s also authorized to change her notification status herself.
If you’re allowed to pause notifications yourself, the menu entry opens the pause notifications menu:
If you select the option Automatically, Unblu automatically pauses notifications when you’re on a video or voice call in Unblu, or when you open a collaboration layer. Depending on how your organization’s configured Unblu, it may also pause notifications for other reasons, such as meetings scheduled in your calendar, or calls in other applications like Skype.
Once the reason for pausing notifications ends, Unblu automatically resumes sending you notifications.
You can also choose to pause notifications for a specific duration. The pause notifications menu includes options for various predefined durations.
Selecting the option Custom time opens a dialog where you can enter the date and time Unblu should resume sending you notifications:
The entry in the Agent Desk settings menu looks different depending on the option you choose:
If you pause notifications automatically, but notifications aren’t paused at present, the menu entry displays the notifications on icon and looks like this:
If you pause notifications automatically and notifications are currently paused, the menu entry displays the notifications paused icon and the reason notifications are paused:
If you haven’t paused notifications, the menu entry displays the notifications on icon, but it looks like this:
If you pause notifications for a specific duration, the menu entry shows the notifications paused icon and the time Unblu will resume sending you notifications:
|Pausing notifications only affects web push notifications and mobile notifications. You still receive notifications in the Agent Desk.|
The next section of the menu is dedicated to various settings that will only affect the agent logged in.
The Manage profile link takes you to the General tab of your Agent Desk user profile.
You can change a number of personal settings, including your first and last name and your username. To change your avatar, upload a new picture with the Select Image button.
If you want (or need) to change your password, select the Define password radio button. This will display two additional text fields for you to define and confirm your new password.
Remember to save any changes you make by clicking on the Save button. If you don’t, Unblu will display a dialog asking whether you want to discard the changes.
If you want to save your changes, click on Keep editing to return to the main window, then click on Save.
|You can’t edit your profile if it’s managed in an external system. You can still edit your canned responses and deputies, though.|
Canned responses are predefined text snippets that you can use in chat messages. They help reduce the amount of typing you have to do for frequently used messages.
Canned responses can be defined for entire Unblu accounts, teams, or individual agents.
The Canned responses menu item opens the Canned responses tab of your Agent Desk user profile.
In the picture above, the agent has access to five canned responses:
The canned response under the heading My canned responses is only available to the agent.
The three canned responses under the heading Team canned responses I have access to are available to anyone in the agent’s team.
The agent has access to canned responses from two different teams. This is because sub-teams inherit the canned responses of their parent teams.
The canned response under the heading Account canned responses is available to all of the registered users of this Unblu account.
The overview pictured above belongs to a supervisor of the team "Cards help desk". As a result, he can duplicate and delete not only his personal canned response, but also the one owned by the team that he supervises. That’s what the Duplicate and Trash can icons are for.
Whenever you click on the Canned response icon or type the keyboard shortcut associated with canned responses in the text input field of a chat conversation, Unblu will display the canned responses available to you. If you continue to type, Unblu will only display the canned responses whose titles match what you entered. The picture below shows the canned response dialog after clicking in the icon and typing
re in the text input field:
For more information on how to use canned responses, see the section on using canned responses above.
There are three different ways to influence the overview of canned responses.
If you enter a term in the filter text field, only canned responses with that term in the key, title, text, or owner field will be shown.
By default, canned responses are sorted by the name of their respective owners. You can choose to sort them by title, text, owner type, or key instead.
If the canned responses are grouped (see below), the sorting is applied separately within each group.
By default, the canned responses are grouped by owner type, that is, agent, team, or account.
It is also possible to group them by owner. In that case, if you are a member of a team that is a sub-team of another team, the canned responses of your team and its parent team will be grouped separately.
Alternatively, you can choose not to group the canned responses at all.
The New canned response button opens a fly-in page where you can create a new canned response.
The Choose language dropdown allows you to specify which language the canned response is in. When you open the canned response dialog in a conversation, it will only display canned responses in the same language as the conversation you are in.
If you are a supervisor or administrator, you can specify the scope of the canned response with the Choose scope dropdown. This allows you to create new canned responses for a team or the Unblu account, not just for yourself.
You can only specify the scope of a canned response if you opened the overview of your canned responses. If you open the canned responses of a user, a team, or an account, the new canned response is automatically created within the scope of that user, team, or account, respectively.
It isn’t possible to select another user as the scope of the canned response.
The Title is displayed in the canned response dialog above the text input field when you are in a text conversation. It should be a brief description of the canned response no more than 250 characters long.
You can use the same title multiple times, or create a personal canned response with the same title as a canned response provided to your entire team. This is especially useful when you create canned responses in multiple languages but always want the title to be in your own language.
The Canned text is the text that should be inserted in the text field. If Markdown is enabled you can use it in canned texts.
If you type the dollar sign $ in the canned text input field, Unblu will display a list of the placeholders available for the canned text. The two predefined placeholders
$(agent_name)allow you to include a visitor’s name or your own name in a canned response without having to enter them. However, you can also define your own custom placeholders.
You can also add a placeholder by clicking on the button sporting its title shown beneath the canned text input field .
Once you are happy with your new canned response, click on Save to save it. It will appear in the canned response overview immediately.
Placeholders make it easier to adapt the text of a canned response to the particular circumstances of a conversation. This can make the same canned responses useful in multiple contexts, thereby reducing the number of canned responses you need to maintain.
To create a custom placeholder, click on Add placeholder beneath the Canned text field of the New canned response fly-in page. This opens a menu where you can choose the kind of placeholder you want to create:
Text field placeholders create a field in the canned text that you can fill with any text you like.
The Placeholder title is used to select the placeholder and add it to a canned text. For example, to select the placeholder defined in the picture above, you would enter
$(account number)in the canned text.
Within each canned response, every placeholder title must be unique.
The hint appears in the canned response dialog in a chat conversation. Its purpose is to remind you of the content you should enter in the field.
Multiple-choice placeholders let you choose from a predefined set of choices.
As with text field placeholders, the Placeholder title is used to select the placeholder and add it to a canned text. It can be up to 200 characters long and must be unique within a canned response.
In the section entitled Choices, add the different options agents should be able to choose from. Enter the text that should appear and click on the plus icon .
You must add at least two choices, and can add up to 16 choices. The choices must be unique within each multiple-choice placeholder.
Once you have finished setting up your custom placeholder, click on Update to make it available for use in the canned response.
|Custom placeholders aren’t shared between canned responses. If you want to use the same placeholder in more than one canned response, you’ll have to add it to each canned response individually. Alternatively, you can duplicate the canned response from the canned response overview and then adapt the text.|
If your organization uses Unblu’s deputy delegation system, you may have a Deputies item in the Agent Desk settings menu. The menu item opens the Deputies tab of your profile.
The tab displays all your deputies and shows which delegation level they’re assigned to. If you don’t respond to a conversation request sent to you personally, your level 1 deputies are notified. They can then accept the request in your place. If your level 1 deputies don’t respond to the request, your level 2 deputies are notified, and so on.
If you’re authorized to manage your deputies (like the agent in the picture), you can add and remove deputies, and move them to a different delegation level:
To add a new deputy, click on Add deputies. Next, select whether you want to add an agent or an entire team as your deputy. This opens a fly-in page where you can choose the deputy you want to add.
The fly-in page only lists agents or teams that aren’t already your deputies.
Use the My team filter to limit the list of possible deputies to agents in your team, or to your team and its subteams.
To move a deputy to a different delegation level, click on the options menu next to the deputy in question. Next, select which level you want to move the deputy to.
Unblu moves the deputy to their new level immediately.
To remove a deputy, click on the delete button next to the deputy. Doing so removes the deputy immediately.
If you can’t manage your deputies, the Add deputies, delete button, and options menu won’t be present. You’ll still be able to see who your deputies are.
Keyboard shortcuts are an alternative way of completing an action that you would normally use your mouse for. The Agent Desk has an extensive set of keyboard shortcuts to make it easier to navigate the UI and to access various controls. UI elements that have a shortcut associated with them display that shortcut as a tooltip when you hover over the element with your cursor.
The Keyboard shortcuts item in the Agent Desk settings menu open a modal dialog that lists the current keyboard shortcuts.
The modifiers displayed are based on those available in your operating system. The default keyboard shortcuts have been chosen so as not to clash with the keyboard shortcuts used by common browsers.
If you prefer to specify your own keyboard shortcuts, click on Customize shortcuts to open the editing modal dialog:
To change a keyboard shortcut, first click on the modifier or modifiers that you want to use and enter a letter or number in the text field at the end of the line. It isn’t possible to change a shortcut by placing the cursor in the text field and pressing the combination of keys you wish to assign to the action in question.
Bear the following points in mind when customizing shortcuts:
Make sure that the shortcut you want to assign to an action isn’t already used by your browser.
You can create keyboard shortcuts with up to three modifiers. In the picture above,
Open the queue uses one modifier, Alt;
Open the inbox uses two modifiers, Ctrl+Alt; and
Open the secondary inbox tab uses three modifiers, Ctrl+Alt+Shift.
We strongly recommend that you include at least one modifier in each shortcut. If you create a shortcut without modifiers, or only with the Shift key as a modifier, you won’t be able to enter that character in a text field.
If you assign the same shortcut to more than one action, the Agent Desk will carry out all of the actions.
Suppose you wanted to ensure that you always turn your camera and microphone on or off together. In that case, you could define the same keyboard shortcut for both Toggle the camera in a call and Toggle the microphone in a call. When you use that shortcut, it will carry out both actions, turning both the camera and the microphone on or off.
Once you are happy with your custom shortcuts, click on Save.
If you wish to revert to the default keyboard shortcuts, click on Restore defaults.
Supervisors, Administrators, and Superadmins have an additional Administration section in the Agent Desk settings menu. All three user roles will find a link labeled Manage account, which opens the Account Configuration Interface.
Superadmins will also have a link labeled Manage global server settings, which opens the Global Server Configuration Interface.
Both interfaces are described in separate sections.