Insights — August 10, 2017

Human Centered UX Design in the Age of Automation

As UX designers, how can we make the human experience more human in its interaction with technology?

by Lisa McGee

Customer Experience UX Design

A person sitting at a table with a red chair

“Is this tomorrow?” courtesy: Tomáš Müller

__

“The future is dark, which is the best thing the future can be, I think.” – Virginia Woolf

In 1816, at the height of the Luddite Rebellion, eight textile workers in England were sentenced to death after raiding a factory in the Midlands, destroying all fifty-five of the factory’s lace frames (what they dubbed “engines of mischief”) in an act of protest against the machinery that, among other practices germane to the industrial revolution, was threatening their trade and their livelihood. The severe sentencing put the final nail in the coffin of what had been a five year long rebellion against the industrialization of the textile industry in England.

Contrary to popular perception, the Luddites were not opposed to textile machinery in an of itself. Rather they were opposed to certain types of machines, particularly what were known as the “wide frames,” lace and wool frames that produced cheap goods of poor quality, which, the Luddites felt, debased the profession and the reputation of their work. Nevertheless, the term Luddite has since become a euphemism, synonymous with reactionary curmudgeons who are afraid of technology.

And while textile machinery did not replace textile workers all together, it did fundamentally change the trade, requiring lower paid, less skilled labor, and resulting in widespread lower quality goods and a reduced quality of life for many workers.

The industrial revolution, of course, wasn’t all bad. It’s the reason why cars, and cell phones, and many of the other conveniences we enjoy today exist. The mass production it made possible also paved the way for the spread of ideas, making everything from travel, to books, to a college education more accessible. Such is the paradox produced by all technological eras. You win some and you lose some.

__

Fast forward 200 years, and we stand at a similar impasse as the threat (and promise) of full automation looms ever larger. In 1970, the industrial sector employed 1,000 robots. By 2016, this number had grown to over 1.6 million. And according to an Oxford University study, 47 percent of today’s jobs will be automated by 2034 (click here to see how likely it is you’ll be replaced by a robot).

Such trends used to inspire a certain optimism, like that of John Maynard Keynes, who in 1930 famously predicted that by 2030, the pace of automation would reduce the workweek to 15 hours, making more time for leisure, family, and pleasure for everyone. In fact, as of 2017 the work week has remained relatively steady in most places, actually increasing by 30 percent in Europe.

For almost half of the population whose jobs are slated to be replaced by a robot in the next 20 years (not to mention those whose jobs have already been replaced by automation), automation may feel like a death sentence, even as we relish in the constantly refreshing list of new tools and apps that allow us to automate the most tedious aspects of our lives and our jobs.

Responses for how to deal with the looming threat automation poses tend to err either on the side of inevitability (the Elon Musk’s of the world, who see universal basic income (UBI) as the only solution) or on the side of the neo-Luddites, who are trying to preserve or even revive outmoded industries or ways of performing work (think the coal industry).

Both approaches are wanting. While UBI has gained steam amongst the technical class, not only does it fail to account for the fact that many workers don’t just work for a paycheck, they work for fulfillment, and take pride in their work, but this view, especially as it is spouted by tech industry moguls like Musk, also fails to hold the tech industry accountable for its own role in destroying jobs (while also going to great lengths to avoid paying taxes), placing the impetus solely on the government and the taxpayer to solve the problems it helped create.

At the same time, trying to preserve a dying industry or job in the face of automation can be seen as regressive, futile, and sometimes dangerous. Such practices can put countries and individuals at economic and social disadvantages, and like the Luddite Rebellion, most efforts to do so are ultimately doomed to fail.

Can automation be humane?

While there may be plenty of reasons to ring the doomsday bell right about now, I’d argue that we also have reasons to be optimistic. In this time of transition and rapidly accelerating technology, we as UX designers have the ability (and IMHO the responsibility) to affect both short term and long term outcomes when it comes to how automation will impact people’s lives, for better or for worse.

Rather than approaching the issue of automation (and the jobs it will destroy) from the two extremes of resigned inevitably and reactive neo-Ludditism, as designers, we can use our UX prowess to do what we do best: solve human problems, the chief among them being the looming threat of the loss of meaningful human work at the hands of robots.

Several large tech companies are already thinking through the ethics of AI. Google’s UX community recently coined the term “human-centered machine learning” and last month Google Brain, the company’s AI research division, launched a new research program called the People + AI Research Initiative (PAIR), dedicated to driving a “humanistic approach to artificial intelligence.”

As tech companies strive to articulate their own set of ethics when it comes to automation, it’s important (as I have written elsewhere) that UX designers, as individuals, start to do the same.

Below are a few of the techniques I believe designers can bring to the table as they work with clients and consider the repercussions of automation.

1. Remember that employees are users too

One of the most common and frustrating oversights when it comes to notions of human-centered design is thinking that the humans at the heart of the design process are only and always customers. While customers, of course, are an integral (and indispensable) piece of the equation, truly human-centered design considers all of the humans impacted by a design–from front-end customers all the way through to internal stakeholders and a company’s own employees, who may have to interact with software on the backend.

Consider a retail ecommerce store. While you may design an amazing digital experience for a shopper that improves sales and customer satisfaction, the experience on the backend for the employee who has to manage the inventory, the fulfillment, the website updates etc. might be a nightmare. In our experience, the design that is better for everyone is almost always the smart business decision. Not only is there a direct relationship between employee satisfaction and customer satisfaction, but designing with employees in mind can increase job retention and make a business run more smoothly, further contributing to the bottom line.

Software companies like the ecommerce platform Shopify are tackling this issue head on. By making the employee experience a key focus of their business model and design efforts, they are solving a major problem in the industry–that of the ecommerce platform backend that is too complex, not intuitive, and requires a high level of expertise. While the platform may make some development/programming jobs obsolete, it also drastically improves the experience for the employees who are managing the ecommerce store, which in turn can contribute to greater customer satisfaction and overall growth for the company–a recipe for job creation.

2. Reframe the problem

As interface designers and systems thinkers, we are often asked to step in when an organization wishes to streamline a process. For business owners, “streamline” is often code for dollar bills saved, and sadly, this often means that people get laid off.

By definition, the role of a UX designer is never to simply “streamline” a process. Our work is always about solving problems and designing based upon human needs. Streamlining may be an outcome of this work, but it is not the goal.

Reframing the problem, and challenging your clients to think less about streamlining, and instead about creating a better, more human(e) experience, can, in the end lead to a solution that isn’t necessarily about replacing jobs, but about making a job that already exists better, which can also be “good” for a company’s bottom line. If a client asks for a solution that will make their employees work faster, ask them to instead consider where their employees add value to their product or service, and start solving from there.

Possessing the business savvy needed to advocate for users in this way, and reframe problems for your clients is a part of your role as a UX designer, but it takes practice. Learning these skills can help you establish credibility with clients, and give you the authority to steer projects in the right direction.

3. Do the right kind of research

As UX designers, there are different modes of research we conduct depending upon the specific project. Sometimes we conduct in-person interviews, while other times we’ll do our user research remotely, moderated or unmoderated.

Choosing to conduct the right kind of research for the specific project you are working on is essential. While there are certainly times when you can learn what you need to know through surveying or remote unmoderated testing, there are also times when you can only gather the insights you need by doing your research in person.

In person research can give you insight into things like when human touch points are essential for customers in a DTC solution, what environmental factors may be competing with the digital experience you are designing, what components of a process are the most tedious and burdensome for different types of users, the limitations of a user’s workplace technology, or the way a user physically interacts with their device.

This is especially true when designing enterprise software where users often have little say in big changes to organizational processes, and (if they want to keep their jobs) won’t have any choice in the software they use to complete their work. By meeting with them in person, we can understand first hand where they may be adding value in the form of a human touch in addition to the drudgery that they wish they could automate. Playing these insights back to our stakeholders, we can make a case for software that increases the value of human services, influencing how and when to strategically employ automation. You may even find that there is a simpler, un-automated process that might accomplish the same thing, and make the humans involved happier.

Once your wonderfully human-centered designs have been implemented, follow up with research and testing to benchmark your results. This will help your client see the value in continuing to solve their own problems with a human centric lens, and give you ammunition to sell these design principles to future clients who will inevitably want to “streamline” their own organizations.

4. Sell your services and your thinking

As a UX designer, you have a much bigger role to play than moving around boxes on a screen. You are a researcher, an analyst, and a scientist who uses data, human insights, and observation to make decisions about what is best for people. All of the tactics we’ve discussed won’t amount to much if you don’t have buy-in from clients and decision makers. It’s on you to convince your clients that it is not just your services that are valuable, but your thinking. And to demonstrate the value of this thinking.

As our lives become more and more entrenched in technology, ethical design questions that have been answered in one market will spring anew in others. See a place in the market that could use a human centered touch? Sell them your services too. Take benchmarks that you created for your past clients and use them to show the value in this kind of work to new industries. There’s all kinds of industries that can benefit from UX thinking that don’t necessarily seem like digital industries, and currently, may not even have digital interfaces.

Healthcare is one area that is ripe for innovation and user centered design, from the way we facilitate interactions between practitioners and patients to the design of hospitals and clinics themselves. It’s also an industry where it’s easy to imagine harmful effects of the wrong kind of automation. By intervening early with design thinking and facilitation, we can mediate technology’s impacts and influence outcomes toward human centric goals.

Lastly, if you know you’re going to be working toward a solution that will displace people, ask your client if they have any plans in place to help those who will be laid off. Educate yourself on programs, solutions, and success stories within their market to share with them, and encourage investment in job retraining and ongoing professional development for employees whose jobs may be at risk. It might be an uncomfortable conversation, but by holding each other accountable, we can earn the trust and respect of users and clients alike.

__

At the end of the day, automation will continue, whether we like it or not. It is certainly not our role to stop it. Taking this line of thinking ultimately leads to primitivism, denying us the possibility of imagining an optimistic future where technology really can help to solve big problems and support the growth of humans. Unchecked, full automation could lead to a technocratic dystopia where robots do our bidding and everyone but the one percent is bored out of their minds and subsisting on a meagre UBI.

So are designers responsible for the future of automation? Not exactly. But we can (and should) play a role. As UX designers we not only have the opportunity to reduce harm, but we can also lead the charge in holding automation ethically accountable to the humans it is designed to serve.

It all starts with the same question we ask ourselves at the beginning of every project: how can we create the best possible experience for the human beings who will use it?

In the dark future of automation, this question will be our guiding light.

(Articles only appear in the frontend.)