Current Issue
This Month's Print Issue

Follow Fast Company

We’ll come to you.

long read

How Google Finally Got Design

Google’s transformation into a company that creates beautiful software is the story of how tech itself has evolved in the mobile era.

[All Images (unless otherwise noted): courtesy Google]

It would have been crazy to say just a few years ago. But today, Google produces better-designed software than any other tech behemoth. If you don’t believe that, then set down your Apple-flavored Kool-Aid. Take a cleansing breath, open your mind, and compare Android and iOS.

Start with push notifications—a crucial feature that's telling precisely because we take them for granted. How’s Apple doing? Your push notifications appear on your lockscreen. Careful. If you want to follow up on one but don’t swipe exactly on the notification itself, you’ll dismiss it. Where did it go? To find it, you drag your finger down the top edge of the screen. Here they are: An unruly list crudded up with headers and dividers, organized by app even if you wanted them arranged in a timeline, with nary a hint that you tap to open an app.

Examples like these happen everywhere in iOS, and they're painfully obvious when compared to Lollipop, the latest version of Android. There, your notifications appear in a drawer, again from the top of the phone. But every one takes you directly to an action inside an app, making it foolproof to get into maps or Uber or Facebook. There’s intelligence behind what you see: A algorithm that invisibly figures out what notifications are most important to you, and serves those up first. There are hardly any chances to swipe wrong. You won’t end up in a place you hadn’t expected. In so many places, Android is so much more logical, the details so much more alive. Tapping any button sends a wash of color across the screen, like a ripple across a pond—a smart way of underscoring your taps, while hiding the teensy bit of lag that occurs as you wait for app to response.

Such attention to detail used to be Apple’s thing. Today, that distinction falls to Google. Unveiled last year, Material Design—Google’s evolving design language for phones, tablets, and desktop—offers relentless consistency in interactions; invisible rules that govern everything, so that every app feels familiar; and beauty in the service of function. It’s why so many designers will tell you, as they’ve told me, "I just like Android better." Whereas iOS is still inching along without improving much, Google is creating a coherent, unified language that easily scales across phones, with enough flexibility to jump to watches and cars. "It’s not even about composing a UI in one place," says Nicholas Jitkoff, who helped lead the creation of Material Design. "It’s about composing interactions from one device to the next."

Google has come so far, despite years of self-defeating battles over what constitutes good design. "When we brought up design at Google, people used to scoff," says John Wiley, a designer who, in nine years at Google, has seen the company transform. "It made it hard for us to hire great design talent because it didn't seem like we had the full measure of respect for design." Here’s how an organization that once crowed about testing 42 shades of blue and called that design created a user-savvy organization that even Apple could learn from.

Design Comes to Google Before Its Time

Eight years ago, Evelyn Kim was the first visual designer ever hired on a Google product team—a graphic designer, full of Bauhaus ideals about beauty and function, which she’d learned at the famed Rhode Island School of Design. She arrived believing that design should change the company. Her boss, the seminal web-designer Douglas Bowman, believed so too, and so he had her work on his clandestine project to redesign almost every product at Google.

Evelyn Kim

The effort had a compelling logic. Google operated as a bunch of tiny fiefdoms, to better allow new ideas to spring forth, unimpeded by bureaucracy. Thus, every Google product team had a couple designers on hand, to prettify their creations after the hard work was done, with little sense of greater cause.

This created problems. Shouldn’t all of Google’s products share a design ethos? Kim remembers one maddening example. Her team had gathered up all the instances of Google’s own logo, across dozens of products. In that assembled galaxy, you saw not one logo, but many, each a few pixels off. Anyone could easily imagine the whole mess transforming over time into something as shoddy as a Chinese counterfeit.

So in the course of several weeks, Kim and her tiny band of co-conspirators created a unified design language encompassing Mail, Maps, and Search, dubbed project Kanna (an Icelandic word for "explore"). Finally, they anxiously presented their work to Eric Schmidt, Google’s then CEO, and Marissa Mayer, an engineer with no design background who had climbed up to become head of Google’s user experience. It didn’t go well. The executives were overwhelmed at the options they were presented with. Schmidt and Mayer thought the passion project was interesting, but it was a non-starter.

This was the era when Mayer boasted that Google understood design, because the company had tested 42 shades of blue for hyperlinks, to find out which one people clicked on the most down to the decimal point. It wasn’t a design philosophy as much as a naked fear of screwing up Google’s cash machine. The company was focused on growth, not beauty. It was focused on speed. And so one of the earliest attempts to instill a design philosophy at Google died.

Just four years later, in 2011, Evelyn Kim found herself on an eerily similar project, at the behest of co-founder Larry Page, who’d been reinstalled as CEO. Almost immediately, Page confidently told Google’s rank and file that the company now cared about beauty and user experience. For insiders, it was an almost hallucinatory moment. This was Google. And this was Larry Page, a man who, when asked by one designer what Google’s aesthetic was, responded, "Pine." That is, a command-line email system common during Page's college years, whose main draw was its speed.

Page’s answer spoke to a philosophy that still dominates in the minds of many engineers: That the best design is no design at all, because speed is the only metric that matters. Adding anything charming to a computer interface simply slowed down. For many years, that made sense. In the dawn of computing, and the dawn of the internet, it didn’t matter of the computer spat out something ugly, so long as it spat out something as soon as you asked. This was a version of the so-called two second rule, formulated in the 1970s: If a computer didn’t respond within that time frame, humans naturally drifted away. For a computer to actually augment your mind, it had to respond almost instantaneously.

Pine. The very word represented decades of ingrained wisdom about computers. Yet in the four years between 2007 and 2011, something happened not just within Larry Page and Google, but in the broader culture of technology. Google’s transformation, from a company that shut down a well-needed redesign to one that creates beautifully designed software is really the story of how technology itself has evolved in the mobile era.

Matias Duarte

Searching for a New Kind of Design Management

For the role he plays as Google’s most high-profile designer, Matias Duarte cuts an unlikely, garish figure. With a tightly cropped goatee and gelled hair that swirls into to a glossy forelock, he looks like the geeky younger brother of Mephistopheles. The last time I met him, he was wearing a ruddy plaid shirt and red pants, which was restrained for him. I’d seen him twice before in a crushed velvet jacket spun with silvery accents, and a shirt printed with a galaxy of swirling colors, like a well-used painter’s palette. He seems to love paisley.

When Duarte came to Google in 2010, after leading the design of Palm's visionary but doomed Web OS, its products were a mess: They had largely remained stalled, design-wise. The Don’t Fuck It Up ethos of the Mayer era still reigned. Duarte's mandate was to fix Android. But soon enough, his mandate began to grow. Showing me a slide of Gmail, Duarte says, "It’s kind of painful for me to look at Gmail like this. But it wasn’t that people didn’t recognize that it was bad. It’s that Google didn’t know how to institute design."

As Duarte himself admits, Google seems almost purpose-built to create messy products—in fact, that was the essence of Google’s appeal to many people working there. There was a viral web cartoon that summed everything up, in jokey org-charts. Amazon, the creation of the hyper-structured mind of Jeff Bezos, a former management consultant, was represented by a fastidious series of branches, descending two at a time from the CEO on down. Apple, bent around the vision of Steve Jobs, was shown as a single angry red dot touching a ring of anonymous blue dots: Micromanagement and undiluted vision, both at the same time. Microsoft, always riven by strife, was shown as separate clusters of branches, each with a hand pointing a gun at the others.

Google, meanwhile, was simply a bundle of lines crazily pointing everywhere, like a bunch of pick-up sticks touching in too many places to count. Google, in other words, was built to encourage the messiness that breeds new ideas. But that same messiness doesn’t much encourage the coherence that marks great design.

Manu Cornet/Bonkers World

Think again on that cartoon diagram of Apple’s structure, with Steve Jobs as the angry red dot. Jobs wasn’t a designer, but he did design a company built to adhere to a singular vision of how a product should be. Even Duarte admits that he held onto that bias. "I had this experience, that unless you’re able to centralize creative decision making, you end up with sloppy results. I thought, the best you could do was structure individual units," which would each have their own designer as visionary, Duarte says. "But good luck trying to bring coherence."

Looking for an alternative, Duarte didn’t have much to go on. Companies that consistently produced outstanding designs—Braun, Olivetti, Apple—were almost always marked by CEO’s who had close, conspiratorial relationships with their lead designers. But it’s folly trying to design as well as Apple by designing like Apple—other companies can’t recreate its mixture of history and personal relationships. Good design isn’t just a product. It’s also an organization, and a story that organization tells about how it came to be.

In Google’s case, the company probably had no choice but to make design a priority in 2011. Owing to its relentless design perfectionism, Apple was on the cusp of becoming the most valuable company in history. To compete with Apple’s tech cachet, Google’s products had to be well-designed. But Page’s design awakening reflects some broader trends in technology that have been brewing for a decade.

As Brett Lider, Google’s design lead for Android Wear, points out, web design during Google’s ascendance in the mid-2000s was focused on utility. Being homegrown and DIY lent a certain credibility on the web, especially in the valley. Conversely, most well-designed sites were marked by a painful lack of performance. In that brew, ambitious design actually suggested a lack of seriousness about engineering. Google’s obsession with tech geekery, visible in details like the Android logo, and the functional but unimaginative language of stripped down simplicity happened to fit both the valley’s DIY self-regard, and an ancient precept in human-computer interaction: That the most user-friendly thing you could do is to make a computer fast, because if it were fast enough, it would hold people’s attention. Faster speeds inevitably made people spend more time at a computer.

This all changed, of course. Computing power eventually became a secondary draw to user experience. That's partly because broadband exploded, making sheer speed less of a selling point. But mobile is what really forced design to center stage. Unlike desktop computing, which took decades to become household mainstays, the iPhone ushered in a new era of invention that was geared toward computing experts and computing novices—from software developers to grandmothers—at the same time. Everyone was learning about mobile, all at once, forcing both engineers and designers to think about usability on unprecedented scales. User experience, once a discipline that evolved at a pace dictated by Apple and Microsoft, was being pushed ahead by every new app that did things just a little bit better.

Like 13 Colonies, Rising Up Together

Once Page made his announcement, dominoes began to fall at the company. He tipped the first one over, by bringing together a small group of designers—including Wiley, who headed design at Google Search; Nicholas Jitkoff, a UX lead for Chrome; Michael Leggett, at the time Gmail's design lead; and Kim, at Maps—to once-again try to standardize and beautify Google’s desktop products. The internal name for this undertaking, Project Kennedy—as in John F. Kennedy, father of the moonshot—hinted at how strange the effort was for Google. A beautiful product may as well have been the Sea of Tranquility.

But with Page’s blessing, it worked: Just a few short months later, Mail, Calendar, Maps, and Search had all been cleaned up, modernized, and brought into some semblance of a unified UX. Not only were all those products cleaner, they also finally shared their design principles, ranging from where menus lived to how colors were used. And, to quiet the grumbling engineers, voluminous user testing had proven the success of the new design.

This emphasis on consistency would eventually become the spirit of Material Design. But perhaps the most important outcome was the personal ties that began to knit together Google’s disparate project groups. The company’s hive mind was starting to self-organize. In Duarte’s search for a new kind of design organization, he hit upon two key factors: Grassroots connections and the sense of a greater cause.

As Kim points out, designers like her started getting better about describing what design even was. "To convince people about design, we had to say, ‘This is going to solve user problems.’ It’ll take less steps, or people will find that perfect place for a romantic dinner," says Kim. "You always have to frame it as these are the people we’re trying to help. You try to say, ‘This is important as a company to help not ourselves but something bigger than that.’" As Wiley, longtime head of design for Google's search products, says, "Beauty itself has utility. That was a big part of our internal recognition. What beauty brings to function is hierarchy, what's related to each other, and how things are related."

Looking back at that process, it’s striking how different it was than that when Evelyn Kim had tried to reinvent the company’s design language behind closed doors. To hear her tell it, the reason Google was eventually able to build a well-designed organization was that it had failed so many times before. "We practiced trying this so many times, we knew how to do it," says Kim. What Kim and the others on Google’s design team had discovered was that they couldn’t work on an agency model, where a cloistered group of designers prepares a solution unveiled at a grand presentation. The agency model simply couldn’t work in an organization built to foster so much autonomy. "Maybe I’m a romantic and maybe it’s because I’m an American, but I believe in this vision of a bunch of people rising up to together to create change. Just like 13 colonies banding together," says Duarte.


In that formulation then, Duarte would be George Washington, and his coup lay not in commanding Google’s product teams to march in lockstep, but in convincing them all his vision aligned with their own best interests. The strategy wasn’t to tell everyone what Google’s new design would be. It was to convince all the myriad product teams at Google that they were constantly solving the same problems, duplicating each other’s work needlessly while simultaneously not letting the best ideas spread far enough. One example that Lider points to is an early design exploration—one that preceded Material Design by a few months—into how animations should be used in mobile interfaces. Looking for a metaphor to emulate, the project designers came up with the idea that animations should work like choreography on a stage.

On stage, if an actor walks off stage on the left, he wouldn’t suddenly reemerge on stage, on the right. And yet that’s exactly how so many avatars within Google’s own mobile apps behaved. On stage, such jarring leaps that defy physical conventions take you out of the moment, suspending your belief in the world before you. So too with the virtual world. Lider calls that insight—about visual continuity and consistent choreography—the "proto Material Design" that eventually bubbled up to become a guiding principle of the finished system. By surfacing the best work of every team and working it into a system, Duarte and his band of designers created common cause, instead of a singular vision, which turned out to be just as unifying.

The question then remains: How do you make any of this stick, once people inevitably change jobs or move on? How do you make turn a moment of good design into a culture that can outlive any of its authors? Here, there is a delicate push and pull between vision and organization. As Kim points out, Duarte’s own access to Page, and his ability to sell a single design story to the executive team, have cleared the way for designers like her to take bigger risks. They no longer have to explain that design is important. Page said it was, so it is.

But Duarte still hopes to keep that band of 13 upstart colonies growing, no matter who’s in charge. So he created a Congress, of sorts. Today, Material Design exists as a dedicated team inside Google, coming to the aid of the myriad product-design teams, helping solve any problems they encounter. Those solutions are then integrated back into the whole of Material Design. It’s a Borg-like arrangement, sending emissaries across Google's universe, and assimilating the best facets back into the collective. And it works both ways. Designers within the various product teams at Google are slated to do tours of duty inside that Material Design team—so that they’ll spread the gospel just a little bit better when they eventually return to the broader organization. If Google does go on to durable design success, then Duarte will have created something that sounds oxymoronic: The self-organizing design organization.

Can Google Stay Focused on Design? The Case for and Against

For Google, all of this change arrives at a telling moment: The enormous amounts of data that the company hoovers up about us—whether it’s our dinner reservations or commuting patterns or work relationships—offer the potential to unlock a new era of computing. You already get glimpses of it in the anticipatory gestures you find in Google Now, which can anticipate your commute, providing traffic estimates when you usually leave work, and can send you shopping reminders timed to when you arrive at the grocery store. As Brett Lider points out, "Computing is getting more human-centered. We’re getting closer to what people want as opposed to the constraints of the technology." And yet, if designed poorly, newfangled interactions can be jarring, unsettling, even scary. It’s like design’s own version of the Uncanny Valley: smart enough to be freaky, but not good enough to be friendly.

Lider, who works on Android Wear and thinks all day about how to bring computing more quietly into the fabric of our daily lives, knows this probably better than anyone. And he knows that you make technology human-centered not with grand gestures, but with one tiny insight at a time. One of Lider’s examples would be easy not to notice, and was created because the designers realized that smartwatches are either a no-hands or two-hands device: That is, you either glanced at them, or reached to fiddle with your watch using your off-hand. But what if your hands were full, and you needed a different piece of information?

One of Android Wear’s guiding principles is that it be a "no touch" interface. So Lider’s team came up with the idea of flipping through information on the watch face using a flick of the wrist—which, if you already wear a watch, recalls the almost instinctive gesture of jiggling your wrist when your watch gets misaligned to your glance. Instinct is the holy grail of interaction design, and it’s one Google is particularly well-placed to understand, simply because of the massive amounts of data the company can bring to bear on any design problem.

Another elegant example of that is how Android Wear handles emojis. Obviously, if you’re getting a message on a watch and want to respond with a frown, it would be ridiculous to have to scroll through an enormous list of emoji. So instead, Android Wear let’s you draw an emoji with your finger tip. If you can’t draw particularly well, no problem: The software simply guesses which emoji you were trying to draw, and plugs it in for you. That sounds like it would be hopelessly inaccurate, but Google managed to do this by asking 100,000 people to draw each of the most common emoji. Your every drawing is compared against that data, and mapped to the right emoji using a best-fit algorithm. True to the Google way, hidden behind a very human insight about users lies an enormous amount of computing and data.

Yet it remains to be seen whether all this investment in human insights and ambitious design will last at Google. This, after all, is a company that has proven fickle in its passions: For every bet that Google makes, there are many others quietly forgotten, and others still abandoned when the winds changed or the prime movers moved on. Put another way, you can create good design in sprints, but creating great design is a marathon, run over and over again. Google still has so many miles to run.

For one, the company still has a broad, structural challenge in getting its best designs in front of its users. In fact, less than 10% of all Android devices actually have Lollipop, the first operating system to use Material Design—even though it was first released last fall. The countless devices and operating system flavors that exist out in the wild prevent Google from being able to push out updates to all its mobile uses, en masse. Perhaps in time, Google will solve this problem, by forcing greater adherence to standards in its ecosystems. And indeed, that seems to be the goal: one of Google's top designers, Jonathan Lee, who served as the lead visual designer on Material Design, now spends a huge amount of his time educating app developers on how Material Design works, and how to apply in a million different ways. For any of this to be truly successful, Google's commitment can't waver. But Google’s designers also believe that Google’s culture has changed—and culture tends to last.

"We feel the effects on our jobs. There’s this trickle down effect where I’m involved with more strategy, whereas I could never do that eight years ago," Kim says. Moreover, Google's product teams are flocking to the example provided by Page, and his newfound love of design. "When you have Matias having high level conversations with Larry, people look and say, ‘We need to have that.’ When people see those behaviors on other teams they model after it, and if it produces awesome work, they want to do it more."

Related: Google’s New Logo Is Its Biggest Update In 16 Years