In this digital age that we’re living in, conversations with customers are no longer limited to face-to-face or phone interactions. We can easily see this change when we look at the popularity of channels like text messages and messenger apps.
Communication methods are evolving, and conversational AI interfaces are quickly becoming a mainstream channel for connecting with customers. The success of these channels is dependent upon good conversation design.
What is Conversation Design?
Conversation design is the process of applying the principles of natural human-to-human conversation to a conversational AI system’s interactions. Conversation design goes beyond simple voice commands and allows users to interact with a system in an intuitive back-and-forth way on more complex topics.
True communication with a system means that both parties (the user and the system) are taking turns exchanging information. Good conversation design makes the interaction feel as close to human dialogue as possible. This is possible only by collaborating with a specialist who has the experience, soft skills, and the desire to learn new things and stay up to date with technology. This is why companies nowadays are focusing more on learning how to hire a freelance designer that will help them build these experiences and be successful.
Benefits of Conversation Design
A well-designed conversation AI system can be integrated into a website or app to interact with and convert more leads in a way that benefits both you and the user or customer. Some of these benefits include:
- Increased user engagement — Some users are more willing to engage with an easy channel of communication that feels natural and allows them to quickly tackle questions and problems on their own.
- Better self-service — A well-designed system understands the user and is contextually aware enough to guide them through the experience without human help.
- Lower agent costs — Agents can devote more energy and productivity to users with more complex needs, and there are fewer transfers when the AI is more intuitive.
- Faster response time — Conversation systems can respond instantly and can be available around the clock for users’ needs.
- Scalability — Insights taken from conversations can fuel a company’s growth by providing data-backed research straight from consumers.
Users are beginning to expect this level of service from the brands they interact with online, and systems with good conversation design can be a valuable asset, both functionally and for the future growth of businesses.
Principles of Conversation Design
Learning how to design interactions that feel more natural and helpful is dependent upon listening to users and researching how human interaction flows. The natural human conversation consists of a few basic principles that can be incorporated into the design of a system’s interactions.
The following principles, outlined in Erika Hall’s book Conversational Design, are a great place to start when deciding how to give systems a more human-centric touch to their interactions.
Cooperative
An interactive system needs to be cooperative and able to provide the right information with little effort from the user. This means they should be designed to be intuitive and to respond in an easy-to-understand way that closely resembles the natural flow of human conversation.
Goal-oriented
Anytime someone interacts with a chatbot, they have a goal in mind for the conversation. The design should allow the user and the system to achieve this goal as quickly as possible. In order to accomplish this, there needs to be sufficient research to understand what users want.
Context-aware
Responses in human interaction naturally fall into the context of the conversation, but systems don’t always do this well if they’re overly automated. While designing, it’s important to anticipate what users want and expect during each stage of the communication process to help your system respond appropriately and in a current and context-aware way.
Quick and clear
Users often choose to use a chatbot over other available methods of communication because they want quick answers in a hassle-free way. Good design uses unambiguous language and proceeds in a logical sequence to help users get to the point quickly.
Turn-based
In natural conversation, participants take turns listening and sharing bits of information. The system should allow users to participate in the conversation and avoid long blocks of information that the user has to scroll through.
Truthful
Users should feel like they can trust the system they’re interacting with, which means delivering clear and reliable information. Your system should never try to lure in users with vague language, and users should be able to expect what they’ll find behind links your system sends them.
Polite
A polite chatbot not only speaks respectfully and pleasantly, but it also shows consideration of people’s time and needs. Users should be able to get their answers quickly without distractions and pushy sales tactics.
Error-tolerant
Human beings will inevitably make mistakes when chatting with an AI interface, and a system needs to be able to understand the intent and get the conversation back on track. Designers can plan for common misspellings and create prompts and buttons to give users options. Your system should also offer to transfer a user to a human agent if things get too complicated.
The Conversation Designing Process
Creating a more natural and human-like interaction experience is a complex process that involves many levels of optimization. Designers must decide what technology they’re going to employ, develop user and bot personas, and determine what actions the chatbot will help the user accomplish.
Once the core decisions are made, designers create sample conversations to help them identify the right flow that the average conversation with the bot will have. Conversation designers will typically work with a flow map that allows them to fully visualize and anticipate the way conversations should play out.
This flow map will help them identify potential responses and possible misunderstandings so they can optimize the chatbot’s functionality. Once the basics are covered, the dialogue will likely still require extensive rewrites to fine-tune everything. The final task before launch involves testing the dialogue on real users to ensure it flows as flawlessly as possible.
Conversational AI interfaces are only as good as their core conversational design, and how good (or bad) a chatbot is at natural communication can play a big part in whether a user continues to communicate with a brand.
Romy Catauta works in the marketing field and is passionate about writing on web design, business, interior design, and psychology.
Further reading
5 things conversation designers should be thinking about in 2023
Woebot: Case Study in Conversation Design for Mental Health Products
