- Types of bots
- Proactive bots
- Reactive bots
- How to create a reactive chatbot
- End of a chatbot flow
- Setting your chatbot live
- How to create a proactive chatbot
- Chatbot Flow Conditions
- HTTP Request and Customer input
- Supported triggers and actions
- Chatbot Fallbacks
- Chatbot undo-redo
- Chatbot Inbox
With Rasayel Chatbots, you can streamline responding to frequently asked questions you receive on WhatsApp. This can significantly enhance your ability to assist customers promptly and effectively while freeing up valuable time for your support team to address more intricate inquiries.
Important to know
- Currently, bots are exclusively supported for WhatsApp numbers provided by Cloud WhatsApp API and 360dialog.
- If you want to use bots, you’ll have to disable your auto-replies.
Types of bots
We have 2 types of bots, proactive and reactive.
Proactive bots are those that initiate actions or conversations without waiting for a user's input. They use pre-programmed rules to engage with users, offering assistance or information before being asked. An example might include a customer service bot that reaches out with a special offer or reminder.
In contrast, reactive bots wait for a user's input or command before taking action. They respond to direct queries or instructions, such as answering a specific question or performing a task when prompted.
How to create a reactive chatbot
On your Rasayel dashboard, choose Bots → Create a bot → Select “Receive message.”
First, fill in the Trigger information: select the relevant channel (i.e., your WhatsApp number), and add a condition, for example, “Contains specific keywords.” You can add multiple keywords that trigger the bot.
You can add a second condition, for example, “Contains specific data attributes.”
Once you create the trigger, a new window will open, and you can start building your chatbot flow.
You can put your chatbot nodes anywhere you like. You can link them back to previous nodes, too. This will make you more flexible when designing your chatbot experience – from mapping the flow to nailing the details.
Next, compose your first message, add buttons by switching the Buttons toggle and type your buttons’ names:
Afterward, you can choose what happens if your customer presses one of these buttons. In this example, we have two buttons: Yes and No. If a customer presses Yes, the bot will send them the below message and connect them to an agent on Rasayel (by switching the Connect to an agent toggle).
And if a customer presses No, we will set up the bot to send them a different message and close this conversation on Rasayel (by switching the Close conversation toggle).
End of a chatbot flow
To end a chatbot flow you’d need to toggle one of the following options:
- Connect to agents - Assign the conversation to a teammate
- Connect to a team - Assign the conversation to a team
- Close conversation - It will move this conversation to the Solved folder. You can also close a conversation permanently. And if that customer reaches out to you again, a new conversation will be created.
- Tag conversation - You can tag a conversation by clicking “Add tag” and selecting one of the existing tags.
Setting your chatbot live
After adding all the steps, you can review your flow. And if you’re happy with it, press Set Live. From that point forward, the flow you've set up will be activated whenever a new WhatsApp conversation begins. This means bot responses will be automatically triggered.
How to create a proactive chatbot
The bot can initiate conversations proactively. This can be achieved through pre-designed templates within campaigns, individually via the inbox, or programmatically using our API, making the process of lead conversion and qualification faster and more efficient.
- On your Rasayel dashboard, choose Bots → Create a bot
- Select the “Send a template” option
- Select a channel
- Pick the appropriate template
- Click “Create trigger”
- Drag and drop the "Simple Message" not onto the flow.
- Write the chatbot message, and if you would like to assign this conversation to an agent when the client responds, toggle "Connect to agents" and select the name of the agent.
Chatbot Flow Conditions
You can create more accurate flows by adding conditions to specify who can see particular parts of it by dragging a "condition node" onto your canvas.
Here are some examples of how our team will be using it:
- Segment your flow per language using the “User tags” condition
- Segment your flow per type of customer using “Conversation attributes”
- Segment your flow per type of customer using “Data attribute”
HTTP Request and Customer input
The HTTP Request feature enhances the customization options of a typical webhook. It lets you designate a specific URL from which data can be fetched or compiled. This proves particularly useful if you're employing your own software solutions, such as a Customer Relationship Management (CRM) system, as it allows you to extract and leverage data from that source.
The Customer input feature allows you to create nodes that can pull specific customer data and use this information as a variable in subsequent text nodes. For instance, you could configure a node to request a customer's unique identification number. Once the customer provides valid information, the system then stores it. Subsequently, you can use an HTTP Request to cross-verify if the provided identification number corresponds to an existing entry in your database.
Supported triggers and actions
What triggers are currently supported
- A brand new inbound message (new conversation)
- A new inbound message in a closed conversation
What actions are supported:
- Sending a message
- Sending a message with interactive buttons
- Redirecting the conversation to an agent
- Closing conversation
- Permanently closing conversation
- Multiple actions per message
- Assigning to specific users
- Tagging conversations
- More filter options
If none of the conditions are met, we have a fallback option.
You can define what happens when the user doesn’t reply for a defined period or responds outside the predefined flow:
- Interact outside flow: Define the response based on the number of user messages (1, 2, or 3).
Change conversation status:
For each user message count, change the conversation status to:
- Open conversation
- Close conversation
- Close conversation permanently
You can also choose an action:
- Connect to agents
- Connect to a team
- Tag conversation
If there's no user response, select when to stop the flow:
- 15 minutes
- 30 minutes
- 1 hour
- 4 hours
- 12 hours
- 24 hours
If the user doesn’t respond within the chosen time, you can decide how to modify the conversation status:
- Open Conversation
- Close Conversation
- Close Conversation Permanently
Tailor your chatbot's behavior precisely to suit your needs.
You can undo previous actions in your chatbot flow. You can quickly correct errors and save time when building your chatbot conversation structure.
All conversations with a Bot will be created in a separate Rasayel folder - Chatbot.
You can move conversations to your main Inbox folder at any time. The bot is designed to get out of your way whenever you want to take a conversation over.
Automatic redirection to an agent occurs in the following situations during a conversation:
- If a customer sends a message that isn’t one of the buttons
- If an agent sends a message / presses the button to take it out of bot control
- If the bot fails to send a message