The Problem
The biggest UX challenge when designing interactions with Voice UIs (or Conversational UIs) is that of discoverability: how can we show the user what’s possible without a screen? There are many clever ways that designers on the Google Home and Amazon Echo teams have addressed discoverability in the conversational format of these devices themselves, but they still assume that the user of the device is also the owner of it—but what if that isn’t the case? Someone can hook up dozens of connected smart home objects to their Google Home that a guest using the device will not know exist (and asking “what are you connected to?” is not an option…yet). Additionally, a guest may not know the names that the owner has assigned to various connected light bulbs, thermostats, etc. in the home—how would the guest know that the room they’re in is called “living room”?
The Solution
This project was my attempt at designing printed material to accompany a smart home hub (in this case a Google Home Mini) in order to help guests in an unfamiliar home discover the various functions the device can perform for them, via voice and sometimes, their smartphones. I did this by showing to the user a list of options in the format of variables that can be inserted into the proper syntax of voice requests to a Google Home, in this case.
Applications:
Short-term vacation rentals (Airbnb, VRBO, invisible service hotels)
Skills:
Interaction design, copywriting, illustration, visual/print design.
Measuring Results
By far, the most common request is (and always has been) to "play music" - therefore, the MVP for this solution is to get the music onto better sounding speakers, which will happen if the guests use the proper variables in their voice requests. The KPI for determining the success of this implementation will be to look at the requests and see if the guests are indeed specifying the speakers from which they want the music to play.
Using Google and IFTTT, I've been able to monitor guests' usage of this setup and evaluate the effectiveness of my solution. Given it's recent implementation, the sample size of users remains fairly small, but early results are promising: there have been fewer failed attempts to get the smart speaker to do their bidding and overall, guests are making a larger array of requests, including successfully controlling the thermostat and various lights connected to the smart home hub.
Summary
This project was an extremely fun and rewarding side project for me because it combined my interest in IoT and smart home automation with my skills in visual and print design. I have set this up as a longitudinal study and plan to check in every few months to look for any insightful results. I personally believe that voice has the potential to vastly improve the guest experience for short-term vacation rental properties, but it will require a lot more thoughtful attention to the challenge of discoverability, designers of these voice assistants must divorce themselves from the notion that the users are always going to be the people who set up the device.