Behind every good brand is a stellar copywriter, and the same goes for conversational experiences. Conversation design is an amalgamation of design processes, thoughtful copywriting and language and linguistics principles, and bypassing any of these would be remiss: at the detriment of the product and the user experience.
One of the interesting things about Conversational AI is that it brings together the creatively minded: writers, designers, and linguists (Conversation Designers come from all walks of life), with the technologically oriented. In a highly technical field, there is now a need for creatives like myself to create experiences that hinge on our most intrinsically human trait: language.
Conversation Designers work with language and user experience (UX) to make conversations with AI natural and intuitive; while developers work magic behind the scenes to bring our wild ideas to life. We are two halves of an apple, and we need each other. Yet while the tech enables conversational AI experiences, language is the driving force; and to design quality experiences, the conversational aspect needs to be at the forefront of the design process.
We Need to Talk
While we all know how to have a conversation, most don’t know how a conversation actually works. For Conversation Designers—as the job title suggests—this needs to be the first consideration. Luckily for us, people have been having conversations for millennia and there is plenty of research on the subject.
While conversation can seem messy and random, it is actually an organised type of chaos. Language is always evolving, with new words constantly emerging and each generation with their own unique expressions—yet despite this, the patterns of language stay the same. In the book Conversational UX Design, Moore and Arar discuss that as Conversation Designers, we don’t need to reinvent the wheel; rather we can draw on decades of linguistic research to design better experiences.
While tech advances are taking conversational AI to new heights and capabilities, it’s easy to foresee a throng of mediocre conversational experiences in circulation, that have utilised the tech but failed to understand the user’s needs or the patterns of natural conversation. Most conversational software now has generative AI features that use AI to fill in the gaps and generate content; however conversational experiences can’t be built solely on ad-hoc AI content.
Chatbots and voice experiences still require considered design, copywriting, and Natural Language Understanding (NLU) training to ensure we are getting it right on all levels. From initial design: discovery, brand tone of voice, user research and bot persona; to testing and ensuring a diverse range of people with varying genders and accents are used to maximise the bot’s comprehension; to ongoing training through NLU analysis and iterating as necessary once the product is live.
All these processes have language at the core. After all, conversation is the interface that will carry the experience—and with any conversation, the right language is key. Instead of a quick fix, it takes a nuanced, iterative approach to create a product that is in alignment with the brand: a smooth, enjoyable user experience that is as accessible as possible for a wide audience. So, rather than leaning on tech to create conversational experiences, put language first to design an experience that speaks for itself.