Everybody knows. Artificial Intelligence (AI) can no longer be ignored. AI is used in chatbots, voice assistants and all kinds of robots. This is why today’s designers need to be able to design conversations for all these new interfaces.
Won’t AI be a lot of extra work? Isn’t AI too complex for us? How do you design conversations for multiple channels and interfaces? These are the questions both designers and businesses from around the world ask us the most. Everybody wants to know how your build successful conversations. The key to success? Design voice first.
If your conversation works well in a darkened room with just the sound of a voice, it will work anywhere. Designing dialogue based on speech means you need to focus 100% on the core of an interaction. It is the most difficult and advanced challenge for designers because you don’t have any visual elements to support your design. No text, no images, no buttons. Anyone who can design voice first, can design anything.
If you’d then want to use your voice conversation for a chatbot, you can add buttons to help the user navigate more easily. But, if you start the other way around by designing the chatbot conversation first, you have a problem. Because if you’d like to make your chatbot service available in a voice assistant via Alexa or Google home, you will have to redesign everything.
On top of that, designing voice first will make your conversations feel more natural. A sample dialogue is the best way to start designing for voice. It is like two people doing improvisational theatre. One person plays the user, the other plays the bot. Record the conversation and analyse your findings. This way, you’ll have a great starting point for designing a natural conversation.
The latest development is the multi-modular interface. This combination of speech, chat and screen is used more and more often. Your voice first design comes in very handy here, because if your designs were designed voice first you can add visual elements for an even better user experience.