Before you jump in to develop a Bot, it would be a good idea to spend some time in understanding what the Bot would be and how it would help you and your business. This understanding would bring clarity and efficiency to the Bot development.
Chatbots are artificial intelligence systems that the users interact with via text or voice interface. These interactions can be straightforward, like asking a bot about the weather report, or more complex, like tracing missing entry in your bank account. Further, the interactions can be structured with user choosing options from a list of items presented or unstructured freestyle flow similar to a conversation with involving a human agent.
Whatever be the type of user interaction, good design helps in building an efficient bot. A good design allows for answering most of the user queries, anticipates all the conversation flows and expects the unexpected.
Post ver 7.2 release, platform has an option to help you design your conversations using the Storyboard feature. It is an intuitive conversation designer that simplifies and streamlines the bot blueprinting process, without the need to have external flow charts, tracking and versioning tools. See here for more.
Platform Recommendations: The following steps can be considered while designing a Bot:
- Understanding the user needs to set the scope of the Bot. The business sponsors, business analysts, and product owners play an important role in identifying the user requirement by gathering market requirements and assessing internal needs.
- Setting the chatbot goals helps create a well-defined use case. This would involve converting the above-identified scope to a use case. It would be advisable to involve the Bot developer in this phase.
- Designing a chatbot conversation to define chatbot behavior in every possible scenario in its interaction with the user. Simulating conversations go a long way in identifying every possible scenario.
Once the bot capabilities and ideal use case are well-defined, Bot developer can begin the process of configuring bot tasks, define intents and entities, and build the conversational dialog.
Things to keep in mind while designing a chatbot: Try to answer the following questions (some if not all):
- Who is the target audience? Technical help bots targetted for a tech-savvy customer need a different design when compared help bots for a layperson like, say, a bank customer. Hence assessing the target audience is always important.
- What bot persona will resonate the most with this group? This will help define how the Bot will talk and act in every situation.
- What is the purpose of the bot? The goal i.e. the customer query that the Bot needs to address will define the end point of any conversation.
- What pain points would the bot solve? The purpose and scope of Bots can be set by identifying what the Bot will address and when the human agent needs to take over.
- What benefits would the bot provide for us or our customer? The main benefit of using a Bot would be time-saving. The user need not waste their time waiting for a human agent to be available for answering their query. You, as the business owner, need not worry about not being there to cater to all customer needs.
- What tasks do I want my bot to perform? Simulation of user conversation would help identify the tasks that need to be catered by the Bot.
- What channels would the bot “live” in? This would to some extent drive the way the Bot would be presented, the various options available for the chatbot would be limited by the channel/medium it would be used.
- What languages should my bot speak? When catering to a multi-lingual community the language support would be imperative and building the dictionary simultaneously would be useful.