Last updated: 1 December 2021
There are two major parts of any chatbot: its FAQs and its conversations.
FAQs are all the questions your chatbot must be able to answer quickly, on a one-to-one basis. Things like ‘where are your offices?’, ‘what are your opening hours?', etc. FAQs are built and trained using NLP, to ensure your chatbot can understand not only these questions but also slight variants of them.
Conversations are automated, multi-step processes. Chatbot users are onboarded into conversations through a specific trigger.
In this tutorial, we will show you how you can trigger a conversation from an FAQ. This essentially merges the best of both worlds. You can now trigger entire processes using NLP.
Step 1: Create an FAQ
Go to Knowledge → FAQs.
Create a new FAQ by clicking + Add FAQ in the top right.
Provide up to three variants of the same question. Make sure each variant is highly relevant to your question, while being different.
Step 2: Dismiss the FAQ answer
Usually, you would then write the FAQ’s answer. Because we’re going to actually trigger a conversation instead of the answer, for this specific FAQ you are free to write a placeholder answer.
The chatbot user will never see this answer.
Step 3: Create the conversation
We have the first half of our work done. Now, we can create the conversation that will trigger when the chatbot recognises the FAQ.
Go to Builder and click + Conversation.
Give your conversation a name, a description, and an (optional) tag.
In the trigger, we’re going to use the conditional
Trigger specific FAQ. Using the dropdown, find the FAQ you've created.
Step 4: Build your conversation
You can now build your entire conversation.
Moving forward, every time your chatbot recognises the FAQ’s questions (or its variants using NLP), it will trigger this conversation.