This article explains how to set up and write instructions for a chatbot to ask questions in a specific, step-by-step sequence.
By asking questions in a clear and structured order, the chatbot ensures a smooth, conversational flow, which can help collect necessary information, guide users through complex processes, or improve the user experience.
Sequential questioning is useful in scenarios like collecting order details, troubleshooting problems, processing refunds, or guiding a user through a multi-step process. Writing clear, structured instructions helps the chatbot understand the right order of questions, enabling it to collect all the necessary information without overwhelming the user.
What does sequential questioning mean?
Sequential questioning refers to the process of asking a series of questions one after the other, where each question builds on the response from the previous one. This allows the chatbot to gather information progressively and use the answers to refine or direct the next questions.
The benefit of sequential questioning is that it prevents confusion by making sure the user answers each question in the correct order. It also makes it easier to handle complicated workflows where multiple pieces of information are needed, like processing a return or a technical support request.
Example scenario: returning an order
Let’s say a user wants to return an item they purchased from your online store. The chatbot needs specific details from the user, such as the order number, the item to be returned, the reason for the return, the email address used for the purchase, and the shipping address (including zip code and house number). These details should be collected step by step, with each question following logically from the previous one.
For example, the chatbot should first ask for the order number, and only after receiving that should it ask for the item to be returned. After that, the chatbot would inquire about the return reason, then the email address, and finally the shipping details.
Instruction to add
When a user wants to return an item, you should sequentially ask for the following information:
- Order number
- Item they wish to return
- Reason for the return
- Email address used for the order
- Zip code and house number where the order was delivered
Additionally, it is recommended to include instructions for what happens next after receiving all the required information. Will the user receive an email with return instructions? Will they be contacted by phone? Clarifying the next steps will help manage user expectations.
Note: You can set up an Action for the chatbot to automatically handle the return request, or use a Handoff feature to transfer the conversation to a human agent after the required information is collected. You can use these articles to do so:
How to use Actions
Example of sequential question flow
Here’s how the interaction might unfold:
- User: “I’d like to return something.”
- Chatbot: "Please provide your order number so I can help you with the return."
- User: "123456"
- Chatbot: "Which item would you like to return from your order? Please provide the item name or description."
- User: "Red leather jacket."
- Chatbot: "What is the reason for the return? (e.g. damaged, wrong size, not satisfied, etc.)"
- User: "The size is too small."
- Chatbot: "What email address did you use to place your order?"
- User: "order@watermelon.ai"
- Chatbot: "And what zip code and house number was the order delivered to?"
- User: "3511 AB, 12"
- Chatbot: "Thanks! We will process the return and send you the necessary instructions. Thank you for your patience!"
By following this sequential process, the chatbot ensures it collects all the necessary details without overwhelming the user, leading to a more efficient and user-friendly experience.