Sample Chapter: Closing the Loop

This is a sample chapter from Sheryl Cababa’s book Closing the Loop: Systems Thinking for Designers. 2023, Rosenfeld Media.

Chapter 1

The Shortcomings of User-Centered Design

As a designer working in technology, I never thought I would see Black Mirror, the BBC show focused on a future of dystopian technologies, used for product inspiration.

I was in an ideation workshop with a client team that was working on a design strategy for augmented reality. We were talking about potential features and adding sticky notes with ideas to a whiteboard. During our discussion, we started talking about potential unintended consequences to features and design decisions. One of my colleagues brought up an example from the show Black Mirror in which soldiers, implanted with an augmented reality system, saw other humans as monsters that must be killed.

We discussed it for a bit, and everyone was quiet. Finally, one of our clients spoke up.

“Yeah, that’s a good idea—add it to a note on the whiteboard.”

“Add what?” my colleague asked.

“You know, the idea that people can use avatars and disguise themselves.”

It was one of the key moments in which I realized that our methods—our user-centered design methods—were failing us.
We reminded our client that, no, Black Mirror wasn’t meant to be a feature inspiration—rather, it’s a cautionary tale. We reminded him that it’s a good example of showing the ramifications of technology—that not all scenarios are good. And that even if he were thinking of it strictly from the technical problem-solving perspective, that the horror of it should give him pause.

It showed me that the ideation process was too myopic, too idealistic, and way too technology-solution-centered.

And so is the rest of user-centered design.

The design practice is experiencing a critical moment in time. Designers design products and services, especially in technology, that often have millions, and even billions, of users, yet they often fail to see design beyond individual users and the immediacy of their interactions with the products and services they work on. They often fail to anticipate and design for the impact on those who are not the direct users of their products, or for long-term effects on those they design for. And before that, they fail to clearly understand the problem space and the context in which their products will live.

In order to address the problems of user-centered design, you first need to understand what it is, why this is an approach that is widely used, and why it’s so problematic in the first place.

The Beginnings of User-Centered Design

If you’ve ever seen a Dutch bike, known as an omafiets, you might notice that it’s got quite a different design than the typical racing bike or modern commuter bike. Its handles are swept back, curved toward the rider in a way that keeps your arms and wrists free of pressure when you are sitting upright on the bike (see Figure 1.1). This type of bike is a good example of user-centered design: it’s meant to make the act of riding the bike more comfortable and enjoyable. It’s designed for the context in which these types of bikes are used, such as getting to work, carrying kids, running errands, all while wearing street clothes, which is quite a different context than, say, a racing bike. It’s a design decision made more than a century ago that prioritizes how the rider experiences the bike. It does not appear to have prioritized a more efficient manufacturing process, or cheaper materials, although perhaps with the popularity of this design over time, these processes may have responded to the demand. Ultimately, it’s a design that puts the user first.

Figure 1.1
A woman is riding a Dutch-style omafiets.
Image: Todd Fahrner

This type of design, which prioritizes the user’s experience, is certainly not new. However, a user-centered approach has not been inherent to, nor codified within, the design process, particularly in digital technology, until fairly recently. Much of the user-centered approach to design in the technology industry was pioneered by designers in the 1980s, and the spread of its ideas can be attributed to the writings of Donald Norman.

In his 1988 book, The Design of Everyday Things, Norman referred to a conceptual model that has three parts: a designer, a user, and a system. The interaction between the designer’s decisions and a user’s actions is facilitated through what he called a system, which, in this case, are objects and products. This book popularized the notion of conducting user-centered research and framed “good design” as that which is intentionally directed toward, and considerate of, a user’s mental models of how things should work.

These ideas shaped the tenets of the modern user experience design practice. Designers in recent decades have rallied around and emphasized the importance of these ideas: Designers must develop an understanding of end users by engaging with them directly through the course of their design decisions! They should emphasize ease of use and efficiency as it maps to a user’s expectations!

These were much needed advancements in the philosophy of product design, particularly as many products entered the digital realm. For example, think about many products—particularly electronics—that existed before the popularization of user-centered design: the first personal computers or VCRs when they initially appeared on the market. They were barely usable, with buttons and interfaces that were impossible to decipher. The approach that has brought us the iPhone, and the obsession with user-friendliness, was a much needed shift that users of digital products have all benefited from.

This approach has been built upon and articulated in a process known as design thinking, popularized by the Hasso Plattner Institute of Design at Stanford (also known as the d.school). The process was borrowed from a method called challenge mapping developed by Min Basadur, which sought to emphasize problem generation and conceptualization prior to solution development.

The five parts of the design-thinking process are typically articulated in the following way:

• Empathize: This phase emphasizes gaining empathy with your user and their situation, in order to understand the context in which your product or service will potentially solve something. It involves conducting foundational user research.
• Define: This phase is focused on taking your insights from the Empathize phase and synthesizing them to create problem statements, user journeys, or other aspects that will serve as a foundation for your ideas for solutions to your user’s problems.
• Ideate: This phase is where designers brainstorm on possibilities for solving for the problems identified and articulated in the first two phases. Often, designers ask “How might we [solve for X]? to spur creative thinking around potential solutions.
• Prototype: Core to the design-thinking process is experimentation. This phase occurs when designers create quick versions of design solutions—they could be paper prototypes, digital, physical, or anything that is lightweight—to take to end users for testing purposes.
• Test: In this phase, designers test the prototypes with end users, so they can continue to learn, iterate, and refine.

This process has helped many organizations take a user-centered approach to product and service design and development. It has also gained broader appeal as business leaders can see their return on investment for user-centered design. Apple is a good example of a company whose products have had wild success due to their user-friendliness, and many apps and services, such as Airbnb, which was founded by designers, have become prominent and widely used. User advocacy has become core to business decision-making, which has been an advancement from the bad old days in which users of products were poorly understood, seen as just buyers, or considered to be an obstacle to a business’s profits. An understanding of users has become essential.

And yet … this philosophy, with its thoughtful approach to how people interact with products, has led to a myopia in which designers—and others within organizations—have failed to recognize other consequences that fall outside the realm of the direct use of the products they are designing. There have always been outsize effects to both good and bad design. However, there is a growing sense among user-centered designers that the process does not address contextual understanding of people beyond just users, nor take into account impact at scale, nor acknowledge the complexity of the technology for which they are designing.

To clarify, there is nothing inherently wrong, per se, with the design-thinking process articulated previously. And, in fact, much of the criticism directed at the design process is actually related to how this process is practiced. For example, for the Empathize phase, the process itself does not necessarily prescribe who you need to empathize with. But in the practical and common execution of the Empathize phase, most practitioners are focused on one type of stakeholder only: the end user of a product. And this particular way, in which design thinking and UCD are commonly understood, informs how design is typically practiced far and wide.

Ultimately, there are three key problems with the approach and process of user-centered design that contribute to its shortcomings and lead to shortsightedness and potential unintended consequences.

• Users are viewed as nothing more than users.
• A user-centered approach does not acknowledge or address potential harm, and it limits the potential impact of design.
• A user-centered process does not inherently take into account the systemic forces.

Users Are Not Just Users

Users are multifaceted humans who affect others and are affected themselves by contexts that go beyond their relationship with the products they are designing.

Yet the approach in design is often oriented toward only the direct benefit of usage. Think about the tools that designers often use to create understanding about people: personas, user journeys, and user stories are common frameworks for synthesizing insight about users into something that can be used to make design decisions (see Figure 1.2).

Figure 1.2
A typical user or customer journey map articulates the direct relationship between a person and the product.
Image: Nielsen Norman Group

These tools, such as the customer journey, provide nuances about an individual’s contextual situation as it applies to how they might perceive and interact with your product. They help articulate customer pain points when it comes to that product relationship, and they can be used to identify opportunities for making the product experience better. The commonality is that they tend to focus exclusively on an individual who might buy or use a specific product.

Take a mad lib that is popular in the product design process: As an [individual], I want to [do something] so that [I can achieve something]. Although it seems to take an outcome into account (the “so that” part of the statement), this mad lib is most often used to define software features. In most design processes, people are defined solely by their relationship with your product. This is the most efficient way to think about people—as users—but it narrows your purview of the ways their lives and situations interconnect with other people and circumstances.

In addition, designers tend to only design for a “typical” user. By focusing on a typical user, and trying to solve for their immediate needs, they can go through the motions of being a user advocate. Let’s take a designer working on a social media platform. She’s been tasked with designing a fun experience that allows users to look at their past year and enjoy photos and posts they may have forgotten about. Surfacing this experience keeps the users engaged and makes it fun and interesting for them to keep returning. She creates an experience: fun celebratory illustrations that include text that says, “It’s been a great year! Thanks for being a part of it!” She engages evaluative testing to see which versions resonate most, and the feature goes live.

Although this is hypothetical, it is actually close to a feature that Facebook launched and continues to run. You may be familiar with the fallout from the platform’s “Year in Review” feature. In 2014, designer and developer Eric Meyer wrote a blog post addressing his experience with this feature. Rather than surfacing a fun party, an amazing trip, an excellent meal, or any of the other scenarios that Facebook’s designers may have considered, the Year in Review feature surfaced Meyer’s most interacted-with post: the photo that he posted when his six-year-old daughter, Rebecca, died of cancer. Instead of celebrating, he was forced to relive his grief over and over again every time he logged into Facebook, because this feature continually surfaced in his feed for weeks on end. In writing about this experience, Meyer said, “The design is for the ideal user—the happy, upbeat, good-life user. It doesn’t take other use cases into account.”

The designers who worked on this feature almost certainly did not intend to cause this type of trauma for Meyer or other users like him. But they failed to consider people at the margins, who might not be having what they assumed to be the typical user experience.

The tools that designers use contribute to this “flattening” of people into users viewed in isolation of others. In addition, the narrow focus on ideal users, and a failure to recognize additional contexts, contributes to unintended consequences time and time again.

User-Centered Harm and the Limited Impact of Design

In November 2020, design researcher Erika Hall tweeted, “This is all too often how UX design is considered and practiced.” The image she referred to was an anglerfish, with its glowing lure marked “UX” and its mouth marked “business model” (see Figure 1.3).

Figure 1.3
The shiny beacon of design!
Image: Erika Hall on Twitter (@mulegirl)

When you consider many tech industry products, particularly those based on ad tech such as Facebook, this is an apt and depressing metaphor. Designers often create shiny, wonderfully interactive experiences that can lead to detrimental outcomes.

The UCD process does not take into account the impact that designers have on their users beyond their direct use of their product, and in fact, there are no codified tools that help designers think specifically about harm. The repercussions of this are vast, particularly in tech.

Take, for example, Instagram’s infinite scroll. Aligning with tenets of persuasive design, Instagram’s feed provides what twentieth century psychologist B.F. Skinner referred to as variable rewards. The three components are described by writer Nir Eyal, a former proponent of addictive technology. He outlined them as rewards of the tribe (validation by others), rewards of the hunt (satisfaction gained by searching for information), and rewards of the self (gratification through stimulation). Instagram’s feed provides all of these things, with a touch of randomness that keeps its users engaged and interested.

Although users are constantly entertained by scrolling through for seconds, minutes, and hours, other outcomes aren’t so wholesome. Teens are less happy, misinformation thrives, and users experience addiction to these products.

If you asked designers whether Instagram’s infinite scroll constitutes good design, well, it would be hard to say no based on what makes user experience design successful. Is it easy to use? Yes. Does it satisfy and keep users engaged? Yes. Does it give people what they want from the product? Yes.

And yet, this feature, in keeping users glued to the product, can lead to harmful behaviors. Aza Raskin, who is often credited with (or blamed?) for the concept of infinite scroll, has said, “One of my lessons from infinite scroll: that optimizing something for ease-of-use does not mean best for the user or humanity.”

In another example, Tony Fadell, a former executive at Apple and founder of the thermostat company Nest, once remarked that he now regrets the design choices that his team made while designing the iPhone because of the destructive patterns of distraction, interruption, and digital addiction with which the device has been associated: “I wake up in cold sweats every so often thinking, what did we bring to the world?”

From a user-centered design perspective, those very decisions for the iPhone are wildly successful: engagement, ease of use, its fulfillment of users’ needs and desires. On a societal level, it’s a much more nuanced story: there are problems of distraction and addiction, as well as a pattern, particularly with teens, of deteriorating mental health associated with relationships with their smartphones. Unfortunately, the tools and methods of user-centered design make designers ill-equipped to address those nuances.

A key aspect worth acknowledging is that these examples are not about design gone wrong. It’s very different when a product is created, and people are harmed because of flaws in the product or misunderstandings of how people might use it. These examples are not like Samsung phones catching fire. The technologists who worked on that did not intend for exploding phones to be an outcome. The difference between the Samsung phones and something like infinite scroll is that technologists intended the outcomes of infinite scroll to be a user spending more and more time in the product. In prioritizing the short-term goals that benefitted their product financially, they failed to account for the long-term harmful outcomes for individuals as well as for society.

It’s important to recognize that the responsibility to do better does not just sit on the shoulders of designers. Everyone in an organization, particularly those empowered to make fundamental business decisions, is responsible for whether or not products have a positive impact on humanity. To return to the anglerfish metaphor, a fundamental problem is that a designer’s influence only extends to the glowing lure, and the lure sits within the context of a harmful business model. It’s not just a designer’s job, but everyone’s job, to consider context and actively strive to create less harmful products and services.

If organizations were shaped differently to prevent this deterministic use of design, then designers could be more empowered to use their skills to question the purpose of their work and think holistically about it.

UCD and Outside Forces

When conducting work with communities with whom I lack familiarity, whether in global health or regionally, I typically partner with practitioners from those communities to help give cultural context to our research work. This approach is not inherent to the user-centered design process, and adherents to the process are often willfully agnostic to the problem spaces and contexts for which they will later be tasked to design. This is hubris.
Although you can learn a lot by using the design process to “empathize” and “define,” designers often fail to acknowledge their own blind spots and assumptions.

In his Medium article entitled “Stop Designing for Africa,” designer Chris Elawa discussed the failure of the XO laptop, the device from the famed “One Laptop Per Child” initiative in the early 2000s. The initiative failed for a myriad of reasons. One key reason was that the designers and distributors did not understand the context of ownership: “The XO laptop was rejected by adults in African [low income economies] because it focused on empowering one child as opposed to providing value to a family or entire community.” In addition, “Factors such as marketing, distribution, and funding do play a role in the relative success of a product designed for the socio-economic development of African [low-income economies].”

These factors are not typically considered through the course of user-centered design. An understanding of individual users can only take you so far. Now that designers, particularly in tech, are often designing in contexts for millions, and in some cases, billions of individuals, the practice needs better tools and processes to consider these challenges of consequence, context, and scale.

A Change in Perspective: Systems Thinking

Knowing about the variety of pitfalls in user-centered design, what is a designer to do? Given that so many tools of the practice represent the narrow philosophy inherent in UX and user-centered design, it’s difficult to shift your mindset without also shifting the tools you use for problem-solving. Broadening your lens requires both a mindset shift and a shift in the tools that you use. Systems thinking is that mindset and contains your requisite set of tools and methods.

So, what is systems thinking? In Systems Thinking for Social Change, David Peter Stroh describes the practice as “the ability to understand… interconnections in such a way as to achieve a desired purpose.” What he acknowledges through this definition is that systems often do the opposite of achieving a desired purpose: sometimes the system achieves a purpose that is unintended, undesirable to most, or both unintended and undesirable.

For example, picture, as a system, a city’s desire to decrease homelessness within its borders. The city spends money and resources toward attempts to solve the problem. Yet, unfortunately, both homelessness continues to increase in the city, and despite its efforts, the city continues to increase its spending in this area. In this instance, the stakeholders within the city’s government may need to ask themselves: Is our spending having an impact? Are there other interventions, or addressing of root cause, that we need to engage in? Are there actions that are exacerbating the problem? If homelessness continues to increase, then, harkening back to Stroh’s definition, the system is not achieving the desired purpose.

Designing for Greater Impact

One of the most famous myths in animal lore is that lemmings collectively kill themselves by jumping off cliffs. It’s a fanciful idea that definitely fascinated me in elementary school, leading my vivid imagination to picture cute little hamster-like animals dutifully marching toward a cliff edge, robot-like, and quietly plummeting to their deaths. (I never imagined them screaming because I had zero notion of what a lemming might sound like.) In case you were wondering, according to the Alaska Department of Fish and Game, this myth began after the release of a Disney nature documentary in the 1950s. Apparently, during the filming, “The lemmings supposedly committing mass suicide by leaping into the ocean were actually thrown off a cliff by the Disney filmmakers.” Monsters!

Despite its questionable origins, the lemming myth is often used as a metaphor for human behaviors. I find it useful to explain the benefits of systems thinking, a thought exercise around solving more broadly for certain problems. If, as a designer, you were faced with the problem of lemmings throwing themselves off a cliff en masse, you could take a couple of different approaches to reducing the harm. You could set up a lemming clinic at the bottom of the cliff and treat lemmings that were injured, or had survived the fall, maybe even catching some as they were falling. Or, you could build a barrier, a fence, at the top of the cliff to prevent them from falling in the first place.

It’s a good parallel for the way that many of our institutions, both public and private, often approach complex problems. Lots of investment is spent at the bottom of the cliff, treating symptoms or the results of problems, rather than at the top of the cliff, on preventive measures. This plays out in many domains that have “wicked problems,” a term coined by Horst Rittel, a design professor at Ulm School of Design in 1973. Wicked problems include unique, multidimensional issues such as climate change, homelessness, public health crises, and extreme poverty. As the Interaction Design Foundation described it, they are often “problems with many interdependent factors making them seem impossible to solve.”

Often in the design practice, the course of problem-solving is already determined. I’ve had multiple client organizations clarify—as my team and I were conducting foundational research to understand a problem space—that they already knew that the solution would take a digital form, that it would solve a specific acute problem, and that inquiry into other spaces was not needed because of that. This narrow and deterministic focus often leads to solutions that have minor impact.

Take, for example, solutions that are meant to reduce youth homelessness. One can treat the issues that young people face once they are homeless: a lack of shelter and safety, inaccessibility to healthcare and other resources, and a myriad of services that are confusing and dispersed. In the lemmings metaphor, this would be positioning the solutions at the bottom of the cliff. A preventive approach would be to address the issues that would keep young people from becoming homeless in the first place: increased housing or family stability, financial and educational support, a culture that protects LGBTQ youth, an end to institutional racism, and many other interventions, both narrow and broad.

The beauty of systems thinking is that it allows designers to access multiple forms of potential intervention as possibilities, even if, as a designer, you may have the skill set to design for only some areas of intervention. Acknowledging the diversity of possible interventions within a problem space, both preventive and treatment-focused, enables designers and their stakeholders to take a broader lens, and perhaps consider alternate areas of investment or direct ideas oriented toward policymaking and other domains, involving the appropriate experts. In the meantime, if a designer is, for example, a digital experience expert, she can more accurately focus on the most meaningful execution of her intervention space, while understanding the broader context in which that digital solution or experience will sit.

What Systems Thinking Is Made Of

In relation to the ability to understand interconnections, there are several methods that comprise systems thinking. The act of systems thinking, according to researchers Samir Patel and Khanjan Mehta in their comparative study on design frameworks, is focused on “not any one thing, but a set of tools, habits, and practices that help in mapping dynamic complexities. Systems thinking focuses on the cyclical cause and effects, as opposed to linear cause and effect relationships.” Often, in the methods employed in the design practice, designers are essentially required to engage in linear thinking. For example, in product design, think about the typical product lifecycle process. From a high level, it consists of four steps: introduction, growth, maturity, and decline. Similarly, many expressions of user journeys or experience mapping follow a linear process of product use. Systems thinking, through its focus on cycles, as well as radiating effects, extends designers’ thinking beyond just the linear format. It requires you to recognize that problems aren’t solved and done: in fact, solutions feed back into a cycle of effects, and could cause problems themselves, or create cycles of positive outcomes.

If you connect these ideas to some of the outcomes of user-centered design, this approach of focusing on system goals and interconnectedness can help avoid the shortcomings—the narrow view of “users,” the limited impact, the unacknowledged outside forces—and create very different conversations from the outset, which automatically prioritize multiple views of a problem space. And, in fact, when designers consider systems, they can utilize their skills in design thinking to further and more thoughtfully think about how to change the system.

Takeaways

• User-centered design, when it was first developed, was a much-needed advancement in prioritizing the users of products and services. As time progressed, and products and services became more complex, its limitations became apparent.
• Some of the shortcomings of user-centered design include: an assumption that users are only users, that the practice of user-centered design fails to acknowledge potential harms and limits the potential impact of design, and that it fails as a process to account for wider forces.
• A response to the shortcomings of UCD is combining design with systems thinking, which prioritizes a focus on understanding interconnections, recognizes multiple forms of intervention, and articulates whether system goals are meeting the outcomes we want as a society.

back to Closing the Loop