How effectively is to pretend that face/off UX is full of endless possibilities.
It is well known that the graphical interface appeared to simplify the people’s tasks of communication with the machine, and at the same time it became possible to work with a computer without special knowledge.
On the other hand, in everyday life we do not go with a set of pictures at hand to fulfill our communicative tasks, usually we just use words.
Why then in the digital world visual support has become an integral part of communication?
Probably the answer is that of all methods available until recently, the optimal UX occurs in the case of a visual feedback.
With the advent of conversational UX, there is again the question whether it is possible to solve the user’s problem just as effective without any help of graphics.
But when you look at it, what exactly means ‘effective’ here? First of all, it means speed, relevance and a positive UX feedback.
In fact, these same criteria can be applied to the process of solving any problem in real life with the help of another person — we want this person to react to us quickly, offer the best possible solutions, and at the same time seek to be very nice.
Imagine that instead of visiting the service web page, you came to a meeting with a consultant of this service.
Highly likely, you will not immediately ask for the dialogue navigation.
You’ll say hello and ask your first question in the hope of hearing a prompt and competent answer; depending on this answer, you will ask the next question and so on.
Also, the consultant will offer additional options that suit you best based on the information received.
For example, if you come for the same service every six months, the consultant will most likely specify if you need a regular order or something new.
If you came for the first time, for sure the first question would be about your preferences or the popular new services presentation.
That’s the point — in the process of dialogue the consultant not only is able to help you, but he also collects information that you would not have shared in the process of interacting with the graphical interface, and this opens up new possibilities for personalizing the offer.
In this type of communication a context shows up.
If the consultant is experienced and qualified, you will solve your task even faster than using a graphical interface.
Yes, GUI gives us certain freedom and assumes multitasking, but on the other hand it knows almost nothing about us and hardly takes the context into account.
The visual component plays another important role, namely it creates a certain atmosphere and even translates the brand values.
In the case of conversational UX, the only way to do this is text and its tonality, so the value of these parameters for constructing a successful CX is hard to overestimate.
When designing conversational UX, you should also remember that when interacting with a graphical user interface, as a rule, the user takes the initiative to search for the necessary information. In the case of a dialogue, we rather expect the ‘consultant’ to have a proactive position and the ability to clarify the information needed for solving our task successfully.
So what determines a good conversation UX from the user’s perspective?
In short, this is how accurately the program can understand user’s talking and how sensitive reaction is given in response.
It would seem that it is all about the opportunities associated with technology, and to a lesser extent refers to UX.
Nevertheless, as we know from real-life examples, the level of intelligence and communication skills is not the same at all.
Similarly, in the case of conversational UX, it is not enough to apply a super-smart technology; you also need to build a dialogue with the user, which will be most comfortable for him, while providing the user with the requested information and offering the information that matches your goals.
The ability to recognize natural language does not yet mean the ability to create a natural conversation.
In other words, we should try to construct a dialogue depending on the context and thoroughly understand the user’s behavior, and then we will have a key skill of orienting in user desires and intentions.
In general, we try to do the same in graphical interfaces, but there we have more tools to influence the user and a greater credit of trust, especially due to the fact that we leave the active role to the user and almost do not impose anything on him.
Accordingly, to gain this skill, we need not only to be technologically adequate, but also masterfully know the subject area.
For example, if we are talking about booking a hotel, we know exactly what information we need to book in the room, and what information is still important, but secondary.
Just as we would build a visual hierarchy on the reservation web page, now we build a dialogue structure with the user, forming conversational UX.
There is, however, a popular belief that the real development of conversational UX is hampered by the insufficient development of artificial intelligence, and as a consequence, the inability to provide the necessary level of communication with the user.
But what if we do not shift the entire responsibility for poor UX to AI, and instead, try to improve the quality of communication with the user by using already available UX knowledge?
Maybe to achieve better results when designing conversational UX, we must lead the user, as a more experienced partner does in the dance.
Perhaps, we even have to limit user in some movements, but as a result together we will create a beautiful dance pattern, which was originally conceived as a successful result of our communication.
Hence, my main question is why, when creating a conversational UX, we should try to emulate the communication with a real person as much as possible, thereby confusing the user, and pretending that he can say whatever he wants and in any manner?
After all, we do not pretend that if the user starts knocking the pen on the monitor, our button will respond to this as a click, right?
So why do we have to do our best to pretend that the dialogue is driven by AI with unlimited possibilities and there are no rules?
There is an opinion that a human is running scared when talking to a robot, and therefore the main task is to pretend that he is communicating with one of his fellows.
I would say that it is much less scary and more ‘UX positive’ for a human to act in accordance with certain intuitive rules.