Author: Mirthe van Gaalen
Conversation designer at Crossphase
Published on February 14th 2020
Last week, the 6th international chatbot summit took place in Berlin and lucky me was there! These are my 5 key takeaways for those who couldn’t attend. The speakers and attendees came from all over the world and all sorts of backgrounds, so these 5 come from very different directions as well. The first two are about the current state of the market and the direction it’s going, number 3 and 4 deal with conversation design, and finally number 5 is a more personal takeaway. Here we go:
In the early days of chatbots, there was, as with most breakthrough technologies, a lot of hype around how they would transform business as we knew it. Now, we see that this hype is over and reality has set in. Chatbots need to actually start solving real business problems and start showing their worth, otherwise it’s just another technology craze that failed to find practical application in the organization. In his keynote: ‘State of the Bots’, Yoav Barel (founder and CEO of the chatbot summit) shared his predictions. He confirmed the hype surrounding bots has dissolved, but we are also already passed the ultimate low point. The market has started a steady ascent “up the slope of enlightenment, and towards our vision — a vision where billions of people are going to use their own natural language in order to interact through technology”.
But there is more. In this steady way up towards mass market use of chatbots, we come across one big hurdle: the chasm. This is where those who fail to find a practical application of chat or voice will have to admit defeat and start anew. As conversation designers, team leads and businesses as a whole, we are faced with the task of crossing this chasm. How? By identifying and delivering real added value for the user. Barel: “We have two critical years in front of us where we are required to work together as an industry to replicate the current success stories in large enterprises, small businesses and disruptive direct-to-consumer start-ups”.
The second key takeaway is kind of a by-product of this steady ascent towards mass market use of chatbots Yoav Barel talked about. Through talking to experts in the field, from product owners to platform developers, the effect of this market change from hype to reality became clear. Businesses are beginning to realize that chatbots and voice assistants can be a powerful technology, however the focus needs to be on solving real problems instead of simply investing in a chatbot just to be part of the craze. As discussed above, it becomes more and more clear that businesses need to be specific about the real-world use cases their technology will solve in the near future. With the industry having entered this level of maturity comes the acceptance that we need to moderate our expectations and accept that only certain chatbot use cases will survive. This sense of realism means the days of quick fix catchall chatbots are over.
This key takeaway is about conversation design. In his talk ‘Designing for Scalability’ Jared Strawderman, head of UX Framework for the Google Assistant, focused on multimodality. The idea behind multimodal design is to combine multiple user interaction modes, such as speech, text or visuals, to improve the user experience of a chatbot or voice assistant.As conversation designers we are familiar with the notion of writing dialogue voice-first. Once you’ve got it right for voice use the rest becomes easy (your know, relatively). We write these dialogues by imagining what you might say in a human-to-human conversation. The next step is to imagine how the conversation would change if one of the participants was holding a touchscreen. What details of the conversation can now be dealt with visually? And what does this mean for the copy in the prompt? Typically, the spoken prompt can be significantly shorter since the user can just as easily understand the information through the visuals or the display prompt. The degree to which voice and visual are used in the same conversation depends on the device and therefore the context of the user. Strawderman made this clear through the Multimodal Spectrum:
Many businesses take on a quite ambitious use case for their chatbot or voice assistant. For example, the digital, always available customer service colleague that is to relieve its human counterparts. Of course this is not impossible to achieve, however in one of the last talks of the summit the speaker briefly brought up micro-moments as an alternative source of use cases. Micro-moments are those moments where people turn to a device to act on an impulse to do something, learn something, watch something, buy something etc. In baking a cake for example, a micro-moment could be double checking the amount of sugar to add. Google describes these moments as “intent-rich moments when decisions are made and preferences shaped”.
With the idea of micro-moments, we have new touchpoints to consider, understand and design for. If we as conversation designers can identify them, we can create the content or tools to enable people to achieve whatever they want right in that particular moment. These small every-day moments can provide a use case that delivers real-world added value for the user (and business). To identify these moments, we do need to truly understand a user’s needs in the moment and use this context to deliver the right experience.
This last one is a more personal opinion on what is happening at summits and conferences in our industry. We all know the phrases and quotes that say that the greatest lessons are learned at the worst times and from the worst mistakes. And many a management book teaches us to ‘fail fast, fail hard and fail often’ so we can learn from them and get better at what we do. Even the failures and mistakes of others can teach us so much; by others faults the wise correct their own, right? So why is the majority of events and summits like this only dealing everybody’s success? If all the knowledge we share is about how rainbows and sunshine got us to a successful chatbot, what lessons are there to learn? And how is that moving us forward as an industry? I feel like events, meant for knowledge sharing, have turned into a ‘my bot is better than your bot competition’. This, in my opinion false, sense of competition is withholding us from joining forces as an industry to invent this wheel together and truly create added value for users. So, let’s fail hard and share harder 😉.