Market research indicates that conversational user interfaces (UIs) and, by extension, virtual assistants will be the preferred UI of digital natives. There are two main reasons why this new way of interacting with computers is so promising. First, conversation is the oldest and most intuitive way for humans to interact. So, conversing with machines as we do with humans makes the interaction very natural and powerful. Second, conversational UIs are an opportunity to make discoverable the rich functionality that was previously hard, if not impossible, to find.

Let’s talk about it

With a conversational UI, users can communicate with a computer system, most often in the form of a chatbot or a digital assistant, as they would with another human being. It’s no surprise that younger people feel more comfortable with technology and are thus less self-conscious about conversing with a computer than their generational predecessors. But many occasional or inexperienced users also feel more at ease with a conversational UI because it takes the “techy-ness” and guesswork out of the interaction. This simplification of the customer experience through natural language, via text or speech, also translates into less need for training.

Conversational UIs rely on machine learning (ML) and natural language processing (NLP) to enable users to communicate with software in an intuitive way. “Natural” because it contrasts with “unnatural,” unambiguous ways of interacting with computers such as coding in a programing language or clicking on a graphical user interface (GUI). Human language is complicated and polysemic. The task at hand is therefore not a small one. The nuances of human communication must be carefully addressed to simulate person-to-person conversation. Although we have achieved considerable advancements in NLP and ML, there is still a lot of uncertainty about just how to design the dialog between humans and machines and how human-machine relationships might evolve.

It is clear though, that to be successful a conversational UI must be just that, a conversation. One-way communication, whether human-initiated or machine-initiated, is not very satisfying. In the context of the enterprise, a virtual assistant must also be able to react to commands and questions initiated by the user and initiate conversations on its own. In business scenarios, a conversation with a digital assistant is most useful when it draws upon full knowledge of the user’s enterprise data. In this way, it not only provides data the user has explicitly requested, but in addition has the necessary information to propose a suitable course of action. Another requirement of a digital assistant is that it carry the context of previous interactions from one conversation to the next. It must “remember” our preferences and past interactions. Imagine how frustrating it would be to have to reinitiate a relationship from scratch with a co-worker each time you speak to him!

Getting more out of technology

In addition to being a more natural way of getting things done, conversational UIs allow us to focus on the task at hand. Currently, we interact with an ecosystem of digital things at home and at work as well as in our cars and through the gadgets we wear. We see a proliferation of applications in app stores and catalogs that are available to us for play and work. The task of finding the right app for the right purpose at the right time is increasingly overwhelming. Worse, each of these apps has its own interface.

In short, the GUI has reached its limits. It is no longer sufficient for users to manage their ecosystem of apps and devices. We need new interaction models to help users reduce their cognitive load and offset the increasing complexity and diversity of our digital world. Conversational UIs allow users to focus on the task at hand instead of wasting time remembering where and how to find what they need.

The integration of conversational UI capabilities into digital assistants, whether they be for consumer or enterprise systems, humanizes (and thus improves) the way we interact with machines. Conversation-enabled digital assistants allow the user to take full advantage of what artificial intelligence, machine learning and other technological advances can offer – ultimately making humans more powerful. Now that’s something to talk about!

Maricel Cabahug is Senior Vice President, Global Head of Design, User Experience, and SAP Fiori Product Management at SAP. You can follow her on Twitter @MaricelCabahug.


Not logged in