Velaro chatbots provide a flexible solution that can augment your live agent team as well as handle fully automated chat engagements.
Chatbot agents, profiles, and groups
Velaro chatbots allow the creation of multiple automated agents that can be assigned to one or more agent skill groups. This provides the flexibility to configure multiple topic- and purpose-fit chatbots and facilitates easy maintenance of the chatbot dialog and activity flows.
Chatbots can be assigned to greet, extend invitations, and respond to website visitors. Chatbots can monitor visitor behavior and send a proactive message to greet the visitor or invite them to engage. When a bubble-style invitation is used with the bot, the invitation triggers a new chat session with the bot immediately and the invitation messages are captured as the first message lines in the conversation.
Availability, triage, and routing
Chatbots know when agents are available and triage visitors using both path-based menus and natural language understanding to provide all-hours service, triage visitors, solve basic service needs, route to other purpose-fit bots, or escalate to a live agent.
Conversation paths and response buttons
Conversation paths allow you to configure automated responses and conversations with images and media. Provide multiple choice options to visitors for one-click answers that take them down the conversation path needed. Data capture and workflow rules Capture data elements within conversations as custom data and chatbot-specific workflow routing rule conditions. Use captured date to qualify leads and contacts and perform actions.
A chatbot path can be configured a variety of actions and perform back office fulfillment functions or interact with your CRM or ERP systems to trigger and drive transactions, record creation and completion, lead qualification and enhancement, preferences capture and profile building.
Natural Language Processing (NLP)
Natural language understanding allows a visitor to enter a request using their own words, and the chatbot will match the intent of the request and respond accordingly. Chatbots can be trained to understand the intent of a visitor’s message (or “utterance”) and deliver a message or perform an action. Please refer to the document “Velaro Chatbots with Natural Language Processing” for more detail on using this service.
Knowledgebase article delivery
A chatbot also serves the live agent by providing a keyword search of your company’s content library with the Velaro Knowledge Base. Content assets and topic-specific responses are housed in the knowledge base, where responses and content assets can be edited and maintained. Articles of knowledge are delivered to both live and chatbot agents through keyword lookups. Chatbots return articles by responding to a matched intent.
The Velaro Knowledge base can also be searched directly using a keyword search widget made available to a visitor or offered as a search task by a chatbot.
The Velaro agent console will feature a knowledge base search panel that returns keyword searches and allows an agent to select a response or article and insert it into the live conversation. This feature can be further automated with a real-time keyword detection service.
Setup chatbot profiles and groups
Bot Profiles allow you to set different profile pictures and names for your group. This is the identity of the automated agent. A profile is assigned to one or more groups. A Chatbot profile can be assigned to one or more group. However, only one chatbot profile can be assigned per-group.
Setup chatbot invitations
Chatbot bubble-style invitations can be configured to reach out to your visitors based on workflow rules and group affiliation. Each message send by the chatbot is counted if the message exceeds three bubbles by default. When the chat window is opened, the entire chatbot invitation conversation is visible. When the chat is started by the visitor responding to the invitation, the text of the invitation is archived at the start of the conversation with either the chatbot or live agent. Multiple chatbot bubble invitations can be configured for one or more groups based on workflow rule configuration.
Chatbot bubble invitations are configured on the Design > Invitations screen
Conversation path actions
The Velaro bot allows you to design an automated chatbot conversation using path-based conversation mapping and follow-up actions alongside natural language processing controls.
The deployment controls allow you to:
• Select the group to which the chatbot is deployed.
• Copy another chatbot
• Preview the chatbot
• Enable/disable the bot for the group
|Path Manager||Add/edit/delete conversation paths. Each time you add a path, it is listed in the Path Manager. Click on a path to edit or delete it.|
|Path Designer||Controls to automate messages, present reply buttons, capture visitor input and custom data, and configure follow-up actions.|
Design your bot conversation
A path usually consists of one or more messages from a bot, but it can also perform an action without posting a message. It can also include the capture of custom data conversationally by the bot, which can then be used in workflow rule conditions. Each path is ended with reply buttons or follow-up actions. The paths together create the chatbot dialog flow.
|Path Name||Enter/edit the name of the path. The path name should be aligned with your chatbot dialog conversation planner.|
|Message editor||Add/edit/delete a message and its contents. Each message appears in its own chat “bubble”.|
|Add message or message or data field||Insert a new message or data input field. Messages can be dragged to reorder.|
|End this path with reply buttons||Add/edit/delete one or more reply buttons|
|End this path with follow up actions|
• Wait for NLP Input (Note: NLP requires Natural Language Processing to be enabled for the account)
• Close chat
• Route to agent
• Route to group
• Route to bot
• Go to path
• Check agent availability
• Check group availability
• Knowledge Base
• Show calendar
|Save Configuration||Once a change has been made in the Velaro Bot, the Save Configuration button is enabled. Be sure to save changes to your configuration before exiting the chatbot designer.|
Path message editor
The message editor allows you to customize the bot conversation messages and visitor input.
|Emoji||Select and insert an emoji into the message|
|GIF||Search for and select an animated GIF and insert into the message|
|Data display||Select a custom data field to display in the message|
|Image||Upload an image to be displayed in the message|
|Edit/delete the message||Once you click on the message, you can edit the text, and delete the message.|
|Reorder message||Once you click on the message, your pointer will turn into a hand, and if you click on it again, the hand will close to “grab” the message and you can drag it to another position in the path order.|
|Data input||Adding a data field into a path allows you to accept the visitor's response as the data field input. The Edit option allows you to customize the message when validation fails for the data field. Data input into the bot can then be displayed back to the visitor within any path in that chatbot. Data can also be used as workflow rule conditions in the chat engagement designer advanced workflow. The data input field supports unlimited custom data.|
End Path with reply buttons
Ending a path with reply buttons allows you to focus the conversation by providing specific visitor response options.
|Delete button||Click on the button to select it and the x will appear to allow you to delete the button|
|Reorder button||Click to select the button to drag and reorder the button.|
|Edit button label||Click on the button to edit the button name. Buttons are named “Reply Button by default when they are first added|
|Select path||Select the path that the button that will be followed when the button is clicked. The menu lists all the paths in the designer|
|Add||Add a new button|
|Delete||Delete all buttons in the cluster and display the original option to end the path with buttons or actions.|
End path with follow-up actions
Each end path action has unique configuration controls, listed below.
Wait for NLP input
Wait for the visitor to type a natural language message input, and AI logic will detect the intent and respond with the message path mapped to that intent.
Note: the NLP module will need to be enabled for your Velaro chatbot, and the NLP intents will need to be configured and trained to properly meet the expectations of the visitor typing a natural language request to the bot. See the section below title “Natural Language Processing for a full description of NLP features and controls.
Exit the chat from the chatbot agent side and present the End Chat button to the visitor.
Route to agent
Route the chat from the bot to a specific agent. If the agent is not available, then the system will respond based on workflow rule conditions.
Route to group
Route the chat from the bot to a specific group. The chat can be answered either by an agent or bot in that group. If the group is not available (i.e.: the group is not scheduled, or no agents are available in the group), then the system will respond based on workflow rule conditions.
Route to bot
Route the chat from the bot to another bot group and path.
Go to path
Route the conversation directly to another path in the current bot.
Check Agent Availability
Check for the availability of a specific agent and route the bot if available or unavailable.
Check Group Availability
Check for the availability of a specific agent and route the bot if available or unavailable.
Search the Knowledge
Base Search the Velaro internal knowledge base (KB) for articles with specific search tags or allow the visitor to type a keyword and search the knowledge base, and the bot will return one or more articles from the KB and display them in a carousel in the chat conversation. When the visitor clicks on an article thumbnail in the carousel, the article opens in a new browser window. Paths can be configured for the helpful/not helpful follow-up buttons.
Display the calendar control in the bot conversation, specific to settings in the Calendly account. Set the paths for when events are scheduled and not scheduled.
Enter a webhook that provides data or performs an action. An authentication header may be supplied for additional security. The webhook can be customized to fulfill a variety of functions, e.g.: “Reset Password” or “Check my Order”.