You don't have to be a coder to build a bot! At least not on our platform. Our easy to use , no-code platform can help you build a bot in just a few clicks with great customer experience.
- If you're getting started or looking for more practical end-to-end guides, checkout Cookbooks, which provides end to end bot building guide.
✔️ Before moving forward, we're assuming you've already completed the process of setting up your bot project.
Studio consists of following modules:
- Flows: Here conversatinal flow can be designed using journeys and nodes.
- NLU: The heart and soul of a bot "Natural language Understanding" capabilities can be added to bot in this section. You can train your bot here to respond to various user utterances, detect entities etc.
- API: To integrate APIs in a bot this module can be used.
- Code: Bot builders can write your code functions here if needed and attach them using function node to workflow.
- Code module should be used as a last resort only. Please note that this is a premium feature and access is provided on a application basis. To get code access for you bot please reach out to us at [email protected]
Learn: This module improves your bot training over-time by self-learning.
FAQ: This module can be used to add FAQs in a bot.
Document Cognition: Using document cognition capabilities, bot can answer user queries through unstructured documents.
Event hub: This module helps you add and handle various events.
Chat logs: This module detects patterns in selected conversations and tags them with actionables that can directly help you improve your bot flows.
You can design and test all your conversations in the design section of bot - mainly using Journeys & FAQs. But let's not rush ahead.
There can be multiple use cases for which you started building a bot - in most cases to automate some service/business function, with a conversational layer on top of it.
ℹ️ To take an example, suppose you wish to build a travel assistant bot, and these are the things you wish to handle-
- Flight tickets booking, cancellation and rescheduling
- Hotel bookings, cancellation and rescheduling
- Flight status
- Customer support with live human
Above points describe the scope of your bot, ie, what can your bot help your customers with. Once you're clear with what you want to do, first thing you will start with is designing the flow of conversation, right from where the conversation starts.
- First step of this flow design is configuring the Welcome prompt you see above. Welcome prompt helps you configure the first message your bot shows to user.
- Now that your welcome message is set, next you can plan is designing the flows we discussed above. We do this with the help of Journeys. You will learn more about Journeys in next section.
- In above flow you also wanted to answer customer FAQs, these FAQs can be added in the FAQ section.
- Moreover you can also handle contextual questions.
- Lastly as in above screen, you can configure your fallback options. When user says something that bot doesn't understand, or is not in scope of the bot, bot goes through fallback flow.
Fallback consists of three parts:
- Suggestions : When bot doesn't understand user input, it will help by trying to suggest journeys close to what user might want (if available). You should enable suggestions for a good user experience!
- Document Search : Another fallback bot can resort to if no journey comes up in suggestions as well, is searching and answering from the documents added in knowledge base.
- Finally if none of the above works, bot will go to fallback prompt. In fallback prompt you can set a Fallback message, Trigger a agent journey etc.
Another thing, the heart and soul of bot building "Natural language Understanding" capabilities can be added to bot in this section. You can train your bot here to respond to various user utterances, detect entities etc.
These sections are often used / needed in bot building journey.
- Code: You can write your code functions here if needed and attach them using function node to workflow.
- Events : Events section helps you add and handle various events .
- API : You can add and manage your APIs in API section.
Using document cognition capabilities, bot can answer user queries through unstructured documents.
It's hard to predict everything upfront for a project and create perfect project in one go. Continous improvements in any project are part of building great things, and chatbots are no exception.
Our last sections in studio , ie,
- Learn section helps your improve your bot training over-time by self-learning.
- Conversation logs section detects patterns in selected conversations and tags them with actionables that can directly help you improve your bot flows.
This is so you improve both conversational flows + Training overtime.
We will dive deeper into above points in upcoming sections.