Why strong sound design is critical to successful products

Sound design should not be an afterthought at the end of a design process.

By Amber Case and Aaron Day
September 8, 2017
Sound. Sound. (source: Pixabay)

A brief survey of sound design

Given its broad range of uses, and central role in the formation of our culture and intellectual traditions, is it any wonder that we eventually bent ourselves to the task of turning sound into a full-fledged tool of communication and influence? The history of human civilization is also a history of increasingly complex sound design. How we got there offers some crucial insights into what we’ve come to expect from sound, and how to meet or confound those expectations.


Throughout human history we have devised alarms that alert us to danger, or convey some other kind of status, over a greater physical range than could be achieved through visual signalling. Initially this was something as simple as a stick hitting a log, but as humans gathered into larger groups and more permanent settlements, we refined them to be louder, more distinctive and more customized. It was no longer enough to let 40 people know that a raiding party was coming—now you had to let 2,000 people know that a house was on fire, floodwaters were rising, or a thief was about. The story of alarms is the story of civilization, to the point where we hear dozens a day, almost without realizing it, from car horns and police sirens to school bells and smartphone alerts.

Learn faster. Dig deeper. See farther.

Join the O'Reilly online learning platform. Get a free trial today and find answers on the fly, or master something new and useful.

Learn more

Musical instruments

The earliest musical instruments found are flutes, over 40,000 years old, meaning they predate written language by many thousands of years. We were shaping each other’s emotions with sound long before we learned to do so with text. As best we can tell, early musical instruments were created for use in ritual and were played in communal settings, and in many ways they’re still used this way in modern times.

Today, your daily ritual might incorporate a specific upbeat song into a wakeup or workout routine. Romance and relationships may be associated with particular songs. Outdoor concerts are spaces for coming of age as well as showing off and socializing. They are also places for the establishment and communication of tribal signatures such as fashion, identity, beauty and mating readiness. A popular, catchy summer song (Daft Punk’s Get Lucky comes to mind for 2014, or George Michael’s Faith for 1988) may define an entire summer, not just in one country, but around the world. Together these songs represent rituals of summer or certain moods.

Digital recording and playback marks an advancement over analog instruments in that it lets us produce and share music even more easily. Digital is portable: We can play it anywhere we like. It’s also accessible: We can sign up for a music streaming service and play it at any time.

This ease of access comes with drawbacks though, such as the loss of communal listening, and the loss of sound quality due to compression. Today it’s rare to hear full, uncompressed music as we might in a chamber music hall. Instead, we most often hear it post-produced to make it listenable at low bitrates, making it relatively easy to stream, but at diminished quality. This puts a hidden cognitive burden on our brains, which must work to fill miniscule gaps in the music left by the compression process.

In the short term, this is a small price to pay for instant access to millions of songs, but in the long term it turns music into a sort of virtual junk food, providing a semblance of satisfaction while fatiguing our auditory processing abilities. Still, this temporary fatigue is generally preferable to the random jarring noises present in everyday urban and suburban environments.

Messages at a distance

Among musical instruments, the talking drum is of particular interest to sound designers. Originally from West Africa, this drum is unusual in that it was specifically designed to make a variety of sounds that emulate human speech, giving it a rudimentary vocabulary. This made it effective as a method for long-distance communication between villages, with other drums picking up signals and repeating them further along.

Although not the first device to convey variable messages at a distance, the talking drum was far beyond any other audible communication device of its day, in its combination of flexibility, resolution and range. It communicated a variety of messages, not just a single alarm or status report, by taking advantage of the wide tonal range of languages commonly spoken in the area. Its resolution—the amount of detail that can be conveyed in a given period of time—is still considerably lower than that of human speech, but much higher than the one-bit resolution of an alarm, which is either on or off.

It turns out that many group-level communications tasks fit neatly into this resolution. If you wish to announce to your neighbors that there will be a council meeting next Monday, you need more than a gong. But sending a person several miles to the next village simply to say, “Meeting next Monday” is a lot of effort for a small amount of information. The ideal resolution is just a few bits, but it needs to stay accurate at a distance. The talking drum was an early, elegant solution to a very common problem.

Since then, humans have developed ever more sophisticated ways to convey messages over a distance, and a good fraction of them involve sound. Our first progression into electrically enabled distance communication was Morse Code. It employed sound, and enabled communication that turns out to be extremely useful for large groups like militaries, companies and governments.

Fast forward to today, and the variety of communication media we employ is astoundingly rich. Even in an era when millions have the ability to quickly record and send video at 1080p, we often resort to text messages just a few words long, or willingly submit ourselves to the 140 character limit of Twitter. There’s a certain wisdom to this: whether in sound, text or image, a limited amount of information is often preferable, especially when we’re asked to absorb thousands of such messages in a day.

Sound as instigator

Often, sound is deliberately employed to change the actions of individuals or groups, beyond the simple, direct “watch out” of an alarm. When properly designed, sounds can profoundly shape the behaviors of large groups of people, allowing a central actor to quickly communicate something to a large group of people and coordinate them.

Music is sometimes pressed into this kind of service. At shopping malls throughout China, the song Going Home by saxophonist Kenny G is played to indicate closing hours. In Taiwan, garbage trucks play a recording of Beethoven’s Für Elise as a cue for neighborhood residents to bring their garbage out for pick up. Numerous other examples exist on a small scale (one resort in Washington’s San Juan Islands plays Semisonic’s Closing Time when the spa is about to close), but each is leveraging the same advantages of recorded music: specific, recognizable, unignorable but also relatively unintrusive.

The downside, of course, is the danger of a formerly beloved song becoming irritating through constant repetition, creating a negative association with both the song and the event it’s indicating. Anyone who grew up hearing ice cream trucks playing “It’s a Small World” or the Mister Softee tune every summer day is familiar with this phenomenon.

A better solution in these situations would be to change the melody regularly, but always use the same timbre and instrumentation. This provides all the benefits of song-signalling, from recognizability to unobtrusiveness, but adds enough variety to avoid irritation. This type of subtlety in sound communication is still rare in our current sonic environment. One aim of this book is to make it the norm.

Why sound design is critical

We live in a world with a lot of undesigned sounds. Many are unavoidable or taken for granted. But many sounds can be designed, and when we can design them, we should.

Sound eases cognitive burdens. We build more and more products every year, and our environments are more the result of human than natural processes. Our attention is already overloaded, and rather than taking some of the cognitive load off our beleaguered brains, undesigned sounds add to the burden. This is the pragmatic argument for improving our sound design capabilities.

Sound is also a powerful brand differentiator. Implemented well, it helps a product stand out; implemented poorly—whether through inept design, hardware limitations, or poor integration—it detracts. Just as with visual aspects of UI, sound characteristics should be consistent across products and platforms, leaving us with no overt impression of any specific sound, but a deep sense of satisfaction and brand identity.

Sound is emotional. Because hearing is passive and immediate, it can produce an emotional response outside the will of the human experiencing it. Unless you block your ears or are hearing impaired, you can’t not hear.

Many of our sound reactions are hard-wired by our physiology. Hearing an unexpected and harsh alert from a mobile phone can trigger a startle response. These responses range from brief distraction to a fight-or-flight response, the result of millions of years of evolution. That snap might be potential prey. That growl might be a hungry bear.

It’s not all bears and beeps though. There many positive emotions that sound can release as well. Voice, music, the sound of a well-made machine in operation, and even Pavlovian stimuli such fresh toast popping up from a toaster can elicit an involuntary positive emotional response. Other times, it is the sound itself which elicits emotion. Beyond startling and alarming sounds, we have music, voice, environmental and architectural acoustics that can affect our emotions directly and without interpretation. For example: whether you speak Portuguese or not, try listening to Amália Rodrigues singing “Estranha forma de vida” and try not to slip into melancholy.

Finally, sound impacts productivity. Part of our brain is always attending to the sounds around us. Even if we’re good at compartmentalizing, a portion of our cognitive attention always goes toward listening. We are evolved to expect a certain level of background noise, and a certain level of variability: nature is rarely silent, but it’s also rarely loud, and loud noises in nature often indicate danger.

Because of this, sound can impact our behavior and productivity in a dramatic way. It’s difficult to focus on a task when part of your brain is sensing uncertainty or the implication of danger, but it’s also easy to dismiss these effects when no actual danger is present. For these reasons, sound has an extraordinary ability to influence our experience of nearly any task, promoting or sapping concentration, creativity and resolve.

How many benchmarks do we have for the visual components of user experience design? It’s time we had models and frameworks for sound design as well, that take advantage of existing design processes already developed. Otherwise we’ll be left with more of what we’re already facing—sound design as an afterthought at the end of a design process, or worse, sound determined by the default settings on hardware built in factories far away.

Where do we start?

Not every product needs sound design.

Maybe you build and sell valves for natural gas substations. Nobody really cares what valves for natural gas substations sound like, but a technician cares deeply about whether each valve is functioning properly. An audible alarm may be necessary to announce a problem, but only if it’s someplace the technician can hear it, not at the valve itself. In this case, the sound design challenge is extremely straightforward: there’s only a single use case, and a specific, known audience.

Other products don’t have sound as a design driver but still indicate their behavior and status audibly. Consider a baseball bat hitting a baseball, or an axe successfully splitting a piece of wood. Both produce sounds that give a lot of information about the action performed, and about the qualities of the objects themselves, but neither has been explicitly “sound-designed.” Still other products benefit from the tuning of their passive acoustic characteristics: weights are added to car doors in order to make them feel more solid when they are opened and produce a sturdy “crumpf” sound when they are closed.

This is why electronic devices increasingly look to the natural or mechanical world for inspiration in sound design; why your smartphone makes a subtle mechanical click when you unlock it. There was a time when designers debated whether camera apps should go entirely skeuomorphic (an actual recording of a real camera) or entirely digital (a monotone, artificial beep), but the sweet spot turns out to be somewhere in the middle, with a slightly pared-down version of the shutter sound. Human brains, it seems, like sounds that are familiar but not too familiar.

For the time being, it’s enough that you understand that sound matters, and that sound (like all aspects of UX design) is always designed, whether intentionally or not. In most cases, bad sound design is worse than no sound design at all. But if your product does make sound and you don’t design it, someone will design it for you.

Post topics: Design