Hypothesis of the design trend of chatbots in the coming year!
The technology underneath is not new but 2017 and 2018 have been the most prolific year for chatbot development. If we don’t want the industry to die, the next one year should be the year of massive user adoption.
Many users feel disappointed with chatbots because they can’t fulfill what’s expected from them. However, there a few players doing some really good proposals. It might be true that NLP (Natural Language Processing) is not as advanced as we all would want, but there have been successful chatbots since 2001 with SmarterChild!
We can expect that the trends in the next year will be around the following topics: Use cases, Personality, Context, Webviews and Omnichannel.
Use Cases
With all the available tools in the market anyone can deploy a chatbot in a matter of minutes. That’s why this year we will see every possible use case being implemented. Do you remember those days when you were able to find and app for anything you could imagine? Back then, Apple registered the brand “There’s an app for that”, and we are going to be living exactly the same situation with chatbots, and “There’s a bot for that” will become a real thing.
Designers have two important challenges to deal with
First of all, the designers need to adapt all these ideas to the new format. The designers need to know what’s the best way to deal with the situation in a pure conversational model? Learning piano lessons with Mozart or changing the channel in your TV will be possible soon. So they need to find out how to design all these new interactions.
Secondly, regarding product design, the results of all that experimentation must be analyzed and the scenarios which are better for a profitable business model must be realized. Looks like the best way to maximize the return of investment nowadays is related with developing chatbots that deal with easy-to-answer questions, leveraging the volume of requests to human call centers. Hopefully all the ongoing experimentation will show us some other interesting models.
Personality
Two of the most famous chatbots in history, Eugene Goostman and SmarterChild, gave us some thoughts about how personality could be the keypoint for success.
The personality of the chatbot is one of the most important points to take into account if we want our assistant to succeed. A research must be made about who our early adopter will be and in which situation they will be talking to the chatbot. Once a clear picture is obtained, a personality that fits perfectly with the user and with the specific situation must be made. Defining in advance how our chatbot is going to be will help us eventually to decide how the bot will talk and act in every situation.
Personality is not only a way to engage users and create empathy but it is also a way to reduce the number of topics and the words that will be used in the interaction. Personality is at the forefront of user experience. It’s lovely that the designers can draw on experiences from design, storytelling, film, and psychology to bring our bots to life.
Context
Being able to understand when and where the conversations take place and its context, will help the designers to have a better understanding of the user needs. Understanding the chatbot context will increase the success ratio of their chatbots and reduce the frustration they tend to produce on users. In the coming year, we will see solutions working on this side, which leads us to an interesting dilemma: How much a chatbot can know about us? Wouldn’t that be scary?
Webviews
Creating successful conversation flows for every scenario is a really difficult task, and that’s why most of the messaging platforms count with rich UI elements, like buttons and so on. Webviews are included so that any interface can be loaded as if it was any other app.
Despite we know that this is going to be a trend,webviews can be considered as a bad practice in the conversational challenge and designers must be encouraged to avoid its use. Using webviews is a shortcut that won’t work with voice assistants and with other interfaces willing to deal with the conversational problem.They must know that if they need to use a webview for their use case, then maybe they need to reconsider the use case itself.
Omnichannel or Multimodality
Against the use of webviews, an Omnichannel approach is recommended for those “impossible” interactions. Any time soon a lot of chatbots will have arms that will extend its use outside the messaging platform, showing relevant information in our SmartTV or our fridge.
Conclusions
Chatbots are here to stay but the next year needs to be the year of maturity for them. We will still see a lot of experimentation but soon they will need to face the reality and the results. Until the killer chatbot arrives, we predict that these 5 topics will be holding the conversation around chatbot design.
This story was originally published on the YellowAnt blog.
