I am searching for a new doctor on my insurance company’s website. A button to select my plan launches a pop-up that shouts at me in red caps. The search bar returns hundreds of providers without meaningful filters. A large area of the page lists vague and irrelevant pricing—“Mammogram (one breast) $100–$480, get my cost.” Then, in small type, on the center of the page, it states: “Please call a provider before you visit to make sure they’re in-network.” Call? But that’s why I’m here! The site forfeits its very function. I take a deep breath and close my laptop.
My symptoms now include irritability.
My typography professor warned me this would happen. Not that I’d have trouble finding a clinician. But that studying design would leave me crazed. “You think design will be fun,” she said. “But you will become obsessive.”
To teach us to discern good from bad, she assigned a seminal text from the 1930’s, The Crystal Goblet, or Printing Should Be Invisible. In it, Beatrice Warde, a typographer, poses a question. If you were a connoisseur of wine, would you choose a clear, crystal goblet to drink from, or a gold, ornate one, studded with jewels? You’d choose the crystal goblet, she answers. Why? Because the design reveals the content. It elegantly and succinctly solves a purpose. It guides the user toward the objective, without distraction. Good design is transparent.
My typography professor warned me this would happen…“You think design will be fun,” she said. “But you will become obsessive.”
My professor, Sharka, with her Czech accent and stern sensibilities, taught us be exacting in design. “Why did you bold and italicize?” she gasped at me once. “Why did you add color?” “Why these lines, when lines are created by your columns?” Every detail must contribute to the function.
So her prediction came true. I began to see bejeweled, overornamented designs everywhere. Not just in typography, but in processes too, like student loans. Websites, like those, later, of my insurance company, were common culprits. Functional design was hidden by garish noise (mammogram pricing for half your breasts!) which obstructed the intended action (to find a primary care provider).
Yet, learning design was fun too—transparent design could also be discovered. When I spent a summer studying in Copenhagen, I saw how the Crystal Goblet applied not only to posters and chairs but to the design of behavior. Walking to class, I was shocked that the Danes beside me stopped at crosswalks, even when cars weren’t approaching. Standing on the corner beside them, I realized their behavior was part of a larger, transparent system design, a collaboration that let cyclists come up and whiz by us: an intended action. Rounded curbs, protected bike lanes, tiny human-sized lights and audible walk timers worked almost invisibly to shape individual behaviors into a prosocial society, which, in turn, supported the individuals.
Why didn’t Americans have similar habits? I wondered. Our system is messy. It’s hard to see the benefit of biking or patience or health. I realized then that I didn’t want to design more lamps or chairs to plop on the planet. I wanted to design behaviors. To remove some of the muck. To help us be our best selves.
After school, Penn Medicine, a large academic medical center, hired me as an experience designer in the Center for Health Care Innovation. We were like a little SWAT team comprised of physicians, health economists, and designers. Our leadership handed us the big, sticky problems of the health system. Why is the liquid-oncology bed flow in the hospital backed up? Why is our clinical trial activation process such that clinicians no longer want to run trials? Why are employees suffering heart attacks and strokes when they have access to excellent care?
Simply, the systems needed better design. They were bejeweled with processes, boards and forms, requirements and barriers that burdened the primary user. The patient, a clinician investigator, an employee—was rarely the sole, primary focus.
We applied an innovation design process and behavioral science to these problems. Deep user ethnography, root cause analyses, and prototyping led us to understand and test the design criteria. Findings from the literature, like precommitments, loss aversion, and social belonging, strengthened our solutions. Together, our designs stripped processes down to their core functions, to make the complicated clear.
In our modern world, it’s easy to junk things up. Simple is hard. We’re quick to add more questions to research surveys, more buttons to a digital interface, more burdens to people.
Our hypertension project, for example, addressed the problem of employees with untreated high blood pressure. The existing system made employees initiate, in their own time off, going to the doctor for a mostly symptomless disease, drive to a lab to get blood drawn and to a pharmacy to get a prescription, and then remember to take that medication, every day, without feedback on its efficacy or importance. Yes, they had “access” to great care, but that access was a seemingly minor prize down a long and loopy obstacle course. Our solution eliminated these steps, instead offering turnstile checkpoints for screenings around the health system where employees worked, and a central location where they could receive care, prescriptions, and labs all at once, for free. Our clinical team also gave them home blood-pressure monitors, trained them how to use the monitors, and then chatted with and coached them from home over an app where they logged their adherence and readings until they were titrated. The program built in self-efficacy, education, defaults, and delight.
In our modern world, it’s easy to junk things up. Simple is hard. We’re quick to add more questions to research surveys, more buttons to a digital interface, more burdens to people. Yet more words, screens, pages, boards, or tasks are seldom the answer. In design school, critiques—the practice of pinning up work and having peers discuss what is and isn’t working—helps us learn to chop and evolve ideas. We step back to see what the content or objective is, versus the design, the vehicle. Successful businesses and research labs do this too, to build creativity, clarity, and a muscle of refined restraint. It takes good leadership—like that at the health system—to zoom out and approach problems in this way. Though effortful, it lets better outcomes, such as health, shine through.
Transparent design doesn’t mean a solution can’t be complex. At Penn, physician-researcher Shreya Kangovi’s community health worker model, IMPACT, is built from a decade of randomized controlled trials, grants, and iterations. Yet the design is a clear, straightforward workflow for nonclinicians to treat underlying determinants of health. In the school of engineering, designer Bethany Edwards created Lia, a flushable pregnancy test, which appears to be a simple paper product, the equivalent of 6 squares of toilet paper. Yet it incorporates demanding design criteria of being discreet, flushable, biodegradable, and accurate. Both designs now also have the outcome of being disruptive.
It’s even possible to add more information and still make your product clear. Physicians in our hypertension program worried that too many patient-logged blood pressure readings would be burdensome. But the app, Twine Health (since acquired by FitBit), presented the information in such a clear graphical snapshot that twenty readings proved much more useful than two.
When I saw Danny Kahneman speak at a meeting last year, he shared a similar principle to the Crystal Goblet he took from psychologist Kurt Lewin, who, like Beatrice Warde, was active in the 1930s. Imagine, Lewin said, you have an object with forces pushing against it from opposite sides. Human nature would say if you want it to move one way, add more force to one side. But Lewin advised against this. A much stronger solution, Kahneman said, is to remove the force blocking the user’s way. Eliminate some of the muck.
I read a designer’s job description recently which requested: “You are an elegant simplifier.” Yes! I thought. That’s what Sharka meant, and Warde and Lewin meant, and what designers and behavioral scientists aim to be. Elegant simplifiers create designs that are crystal clear.