Coping with the overload
Service design, predictive design, and artificial intelligence (AI) in design. Where we are and how we got here.
In the age of ubiquitous computing, when computers are embedded in the world that surrounds us — from wearable devices to houses, cars, all kinds of connected devices and services — technology might quickly become quite demanding and overwhelming.
Designers should predict the user’s needs in certain situations to achieve a normality and unobtrusiveness of the technology.
Predict and simplify
Anticipatory design is — according to Sophie Kleber from agency Huge — a design discipline that deals with algorithms taking the decision on the user’s behalf. Netflix was early with a film recommendation algorithm, Amazon for books and everything else, Spotify recommends songs, and Nest thermostat learns from your habits and anticipates your preferred room temperature. Virtual assistants understand conversations, our devices recognise us by fingerprints, eye scans, and face recognition. Products and services utilising artificial intelligence (AI) and machine learning have been with us for quite some time.
If we want to simplify the user experience (UX) and the user interface (UI), and slide the tech into the background of the user’s perception, we should reduce the number of choices users make while using our products and services. To reduce or even eliminate the choices, we must take the decision on behalf of our users.
Sophie Kleber argues that, “First of all, designers need to understand whether eliminating choices will actually make the user’s lives easier,” and points out a great example of this: “Uber, for instance, eliminated three choices from ordering a car service. Instead of asking users when they’d want the car, they assumed that users want it now, and from where they are right now, not another location. Two choices immediately were eliminated by anticipating the most common use case. Similarly, they streamlined payment methods and eliminated the hassle around tipping. All these options that were removed were a burden to begin with, which is why people like using Uber so much — it’s a relief from the paradox of choice.”
The probability of predicting the correct decision and the cost of getting it wrong should be taken into a consideration. There might be cases with only bad outcomes. Consider the case of autonomous cars. The software steering these vehicles has to be programmed with a set of rules by which it operates the car and takes the decisions on the go. Consider a scenario in which the car would be deciding between options to avoid the pedestrian on the road, drive into a wall potentially killing its passenger, or hit the pedestrian, killing him, but saving the passenger. How to determine the value of human life? Is the passenger by default worth more than the pedestrian, since we don’t usually want to kill our users? Is an old woman’s life worth less than that of a young man in his prime? Is a pregnant woman’s life worth more than the non-pregnant one? Even to consider these questions hypothetically makes us uneasy, but similar decisions have to be made by people designing the rules for the decision-making.
One way of dealing with this kind of moral dilemmas is by ‘teaching’ the AI that makes decisions to react more like a human. Some companies are feeding the machine learning artificial intelligence vast amount of data how real people respond to specific situations, trying to ‘recreate’ a human instinct that drives human subconscious decisions.
At the same time, the risk and cost of getting the decision right should be considered as well. Will you get the user in trouble, or anger, or scare her instead of delighting her? The US retail chain Target employed a predictive algorithm that proposed special offers on products to their users on the basis of their previous purchases. The algorithm correctly predicted pregnancy and sent offers for maternity-related products to the user’s home. The problem was, however, that the 16-year-old hadn’t announced her pregnancy to her parents yet.
The Brand Ecosystem Design Blog is a part of a book in the making. In the coming months, we’ll build up our case from looking at where we are and how we got here, to develop the framework, tools, and language to understand and manage the brand ecosystem.