In 2016 we saw continued excitement and hype around ‘conversational interfaces’ and ‘chatbots’. Several trends are contributing to this phenomenon – according to Business Insider, we are now spending more time in messaging apps than on social networks, and artificial intelligence and natural language processing technology is progressing rapidly. This had led to lots of experimentation and excitement around the idea that we can carry out a range of interactions, such as shopping or getting our boarding pass, in threaded conversations, often in a way that mimics messaging. This represents an interesting shift in how we think about user experiences and interactions, more as a text based ‘conversation’ that helps users to achieve their goals than a visually based interface. Conversational UX also includes audio based conversations, such as those we have with Siri, Cortana, and Amazon Alexa, but for this post we will focus on text based conversations. Conversational UX is about conversations that we have with computers and interfaces in order to achieve our goals.
Best of the Best
This trend is really at the very beginning of its evolution, and there is much critique of the less than great conversational experiences out there. Designing for conversation, with all of the complexities and limitations of humans interacting with a computer in this way, is a challenge! So what did we see in 2016 that really spoke to us?
Quartz News App
The year opened with a real winner when Quartz launched their news app in February. The app reimagined how we consume news, introducing a messaging style interface that gives users the choice of which stories to dive deeper on and which to skip. Asking for more on a story results in follow-up messages that give a summarized version that include key facts. The conversational UX also makes use of emoji, GIFs, and data visualizations such as graphs, making news feel interactive, fun, and fresh. Links out to the story sources allow users to learn more. The user experience of this conversation is very smooth, and meets users where they are (in messaging interfaces) in a format they are familiar with. It also mimics the idea of asynchronously diving in and out of conversation – checking in with the news when you have a moment on a commute, standing in line, or reading in bed. Part of the value proposition is also the curation of multiple sources, allowing user to feel well-informed while staying within the app conversation.
Having ‘conversations’ with brands is an opportunity area for conversational UX, and the Insomnobot-3000 is a prime example of this. Launched in September by mattress company Casper, Insomnobot-3000 is a friendly companion that you converse with via SMS, designed for those nights when you find it impossible to go to sleep. This bot is a fairly low tech, in that it uses a finite database of categorized responses, rather than incorporating machine learning. User research with people informed the copy writing and topics of conversation – showing the importance of understanding human interactions and emotion in creating a compelling conversational UX. In particular, with a chatbot designed to create brand affinity, the experience feeling like messaging with a friend keeping you company late at night is no small feat!
Duolingo is a language learning platform which leverages gamification and personalization to make learning a new language fun and effective. In October, Duolingo introduced Bots within the app to its French, German, and Spanish learners. The Bots feature allows users to practice their language skills by texting with a ‘Bot,’ which takes on different personalities and topics as a way to explore a range of conversations, such as going to a restaurant, going through border checks, or visiting a zoo. This is a very smart application of conversational UX – allowing learners to mimic the experience of having a conversation, often one of the most challenging aspects of learning a new language, in a very low pressure way. Duolingo Bots have their limitations, but the UX is very smooth, with options to get help with replying as well as differing levels of points depending on the robustness of your answer. This is a really exciting application of conversational UX, with huge potential – imagine when you could practice this verbally with a Siri type application!
Keeping the conversation going
Quartz news, Insomnobot-3000, and Duolingo Bots are a few best in class examples from 2016 of what can be done with conversational interfaces. It’s interesting to note the versatility of these examples – from serious news to late night companionship to practicing a new language. All of these experiences leverage the power of text-based conversation and experiment with our frequent use of messaging platforms. Conversational UX relies heavily on copy-writing, nuance and interaction tropes such as emoji, GIFs and SMS to build an interactive experience. Expect more of this in 2017!
Oh, and guess what?! Want to learn about UX through a conversational interface? There’s a bot for that.