Designing for voice

According to a recent Google study, conducted by Northstar Research, nearly half of users in the US use voice search on a daily basis. A catalyst for this growth is also the use of voice-activated home assistants such as Amazon’s Alexa and Google Home. These home assistants help you order milk when you’ve run out, change the music or lights in your house for a different mood, and even give you a heads up when traffic is backed up before you head out to work.

Designing for voice involves writing prompts and responses that help the user accomplish these tasks, quickly and effectively. UX designers already possess the ability to design for voice – it’s not a revolution, but rather a natural extension of what UX designers have been doing all along for other mediums, by guiding their users through a visual journey while navigating experiences. In the case of voice, though, it happens through seamless conversations.

Words matter more than ever, as there are no accompanying visual cues when designing for voice. The designers are relying fully on what the user perceives the chosen words or phrases to mean. Yes, a standard set of commands or keywords exist for the range of voice assistants, but think about it, to tell your users to ‘click here’ has no meaning in the indivisible world of voice. The most fundamental web design element ever, the clickable link, no longer has a place in the future of interface design.

Consistent interpretations of commands between voice and visual interfaces will become the key concern for UX designers, particularly in web experiences. Consider how visual and voice interfaces handle a common digital interaction we are all used to – opting in for a monthly email newsletter. In the traditional visual interaction, a user will fill in their email address and click the sign-up button. It’s easy and straightforward. Now think about how this same interaction can go for in voice: “Subscribe me to this newsletter”, “Add my email address to the mailing list” or “Opt me in for the website newsletter” and so forth. There are almost infinite ways to articulate the same basic intent via voice, which means UX designers must make sure they ask the right questions to prompt the correct verbal responses from the user.

Once the novelty of voice interaction wears off, UX designers will be challenged to maintain user engagement. Designing variability into the voice actions, through different moods and personalities, will prevent users from getting frustrated and keep engagement during conversations. This also places emphasis on crafting various potential responses such as, meaningful error messages that help steer the conversation back on track without being annoying. Users will quickly lose interest in communication with a robotic voice that keeps repeating, “I’m sorry, I didn’t quite catch that?”

Visual interfaces have one advantage over voice, which is being able to display multiple options to a user in a hierarchal manner. UX designers need to ensure that when a user asks for a list of options, that the responding voice does not just list the options like a shopping list, but rather responds in a natural conversational way. A list-based conversation can look something like this: “What are the closest Indian restaurants to me?”, “Would you like to walk or drive, John?”, “It’s a nice sunny day, I would like to walk.”, “Great, Banjaara Indian Cuisine is a quick two-minute walk, but The Curry Place is cheaper and only a five-minute drive away.” Voice interactions need to be treated as natural conversations, otherwise it will always feel like we are talking to machines.

Voice will be the next giant leap in UX design and it will rapidly grow within the next few years. The way we have been interacting with devices and the world around us are about to change a whole lot. As UX designers, we need to ensure that we keep conversations as natural as possible and keep innovating the responses to ensure user engagement does not falter. For decades, we have interacted with technology in a two-dimensional manner, yet as humans we operate three-dimensionally. Designing for voice can ultimately return us to the original goal of what UX is about, by treating people like people again.