The flourishing of messenger bots, also known as chatbots indicates the irreversible trend of more integrated and seamless user experiences with conversational UI. It’s actually about conversations, something we participate in every day but doing unconsciously. The following passage lists 4 steps on how to design a conversation for chatbot, starting with a clear understanding of customer needs. Appropriate conversations are thus designed to improve customer engagement with human touch and to satisfy users’ demands.

It takes you 5 minutes to read.

It is so natural for the designer to put himself in the user’s place before setting to work. In his mind, there is a conversation going on between the designer and the user – That gets the designer closer to know what the user needs and wants. There is also the real talk with users giving designers a clear understanding of customer needs. Conversation of this sort can help the designer hit it off with the customer when real conversation begins.

As computers gain the human touch to converse with users, we are on the threshold of transforming conversation into a user interface. Design is more of a dialogue now, with messages flowing back and forth with the customer. There are so many questions daunting the minds of designers while they set out to design conversations. How to create a perfect chatbot experience and design conversation?

1. Understanding the Goals of Customer

Before designing conversation for chatbot, identify and understand the goals of the customer. To be more specific, understand why the client wants to build a chatbot and what does the customer want his chatbot to do. Finding answers to this query will guide the designer to create conversations aimed at meeting end goals. For instance, let us take the case of a customer aspiring to build a hotel bot.

While charting the scripting course, the designer comes across conversation of the type given below.

● User lands on the hotel chatbot
● User wants the best hotel matching his criteria
● The chatbot replies with the name of the best hotel
● User wants to make a booking
● The chatbot asks the user for the stay dates
● User gives out the stay dates
● Chatbot makes the booking

When the designer gets to know why the chatbot is being built he is better placed to design the conversation with the chatbot.

2. Simulating Conversations for Inspiration

The frontrunners that have used simulated conversations with systems or computers have given us an idea as to how these conversations are build. A company selling products trains employees to communicate with clients. Dialogue simulations to help an employee communicate with the client in different situations and prepares him to handle real-life situations.

Take the case of a car salesman taking part in this simulated dialogue with the customer. This is one of the many scenarios created within a system prompting the salesman to communicate with the customer based on what the customer wants.

Simulated dialogues with customers help the salesman grow in confidence to handle different situations. Designers can take a cue from simulated conversations created for systems that sound like normal conversations.

3. Designing Chatbot Interactions

The designer can model the conversation flow based on the type of interactions between the user and a chatbot. These are segmented into structured and unstructured interactions. As the name suggests, the structured type is more about the logical flow of information including menus, choices and forms into account. For instance, a customer buying a product is prompted to fill an order form. Similarly, a buyer ordering item at a restaurant chooses the item from a list. The unstructured conversation flow includes freestyle plain text. Such as conversations with family, colleagues, friends and other acquaintances fall into this segment.

It is important for the designer to understand capabilities of messaging platforms while designing structured conversations. Designing interactions for a platform supporting plain text only (SMS) is altogether different from interactions for a platform supporting custom keyboards. In this case, the designer considers the possibilities of a single field response and multiple field response, and leverages the right one as the case may demand. For unstructured conversations, the designer makes sure that the chatbot supports minimum vocabulary essential for the user to complete his tasks.

Structured conversation can be built using interview method for single field response or embedding a URL in the message taking the user out of the platform for richer messages.

4. Developing Script

Designing interactions for structured and unstructured messages is just the beginning. Developing scripts for these messages will follow suit. While developing the script for messages, it is important to keep the conversation topics close to the purpose served by the chatbot.

The designer ought to keep the sole purpose at the core of scripting, think of all the possibilities before scripting conversations. Take the case of a designer scripting conversations for a chatbot built to assist hotel reservations. The possibilities are overwhelming, from a user asking the chatbot to suggest the best restaurant in the neighborhood to a user asking chatbot’s opinion about the best item in a restaurant.

For the designer, interpreting user’s answers is important to develop scripts for a conversational user interface. The designer also turns his attention to close-ended conversations that are easy to handle and open-ended conversations that allow customers to communicate naturally.

During the script work, if the designer wants the conversation to be as natural as it can be, open-ended questions prove to be the right fit to achieve this objective. With the user saying and typing whatever he wants to, the designer needs to be diligent in framing questions and processing responses. Moreover, when the chatbot doesn’t come with AI brains, creating relevant responses to user queries will become an ordeal. The designer would be better off in framing open-ended questions only when it is extremely essential. Moreover, open-ended questions can steer the conversation away from end goals.

This is how an open-ended conversation can develop between the customer and a chatbot.

The designer is not short of ideas or inspiration even before he sets out to design conversation. The curious nature of designers can help them learn from observation, hearing and experience. For one, simulated conversations can spark ideas in designers – like the conversation between salesman and the customer. Designers now have a task at hand for the reason that conversation is going to be the future of design.