In 2016, I began working on a site redesign with the Regulatory Assistance Project (RAP), an organization that helps craft climate-change regulations in the EU, China, India, and North America. They approached our agency with a few key project goals; one was to make their global reach easier to understand, so we created a “Where We Work” page to display all their key regional actions on a map (Fig 1.1).
I found a northern-hemisphere map on Wikipedia and edited the SVG so it would spin to display different countries when their name was selected in the sidebar. I thought it was a really cool idea, and the client was on board too. Win-win.
But just before launch, we got a stern email from the client telling us the maps were wrong. “Wrong? How can they be wrong?” I thought. “Wikipedia is always right!” But in my rush to get the vector file ready for development, I neglected to note that the map:
- didn’t include any countries formed after 2011, like South Sudan, and
- didn’t identify disputed territories, like Kashmir, the Golan Heights, or Crimea.
In fact, there are many parts of the world that don’t fit into nice neat lines, including the borders of China, India, and Pakistan—right where RAP does a huge amount of work. Oops. We quickly made the corrections (Fig 1.2).
Displaying only one view of those disputes meant that we were not participating in the dynamic, socially enacted process of culture—what Huatong Sun, an associate professor of digital media and global design at the University of Washington Tacoma, called “emergent, becoming, practiced, temporal, and thus contested” (http://bkaprt.com/ccd/01-02/). Human societies are always in flux: we endlessly remake our identities, our languages, and our methods of communication. We constantly negotiate our interactions, intentions, and meaning through language, technology, art, and the environment—and the web is part and parcel of all that.
Cross-cultural design is a complex topic. Cultural misunderstandings can lead to discomfort, embarrassment, and offense, hurting users and ruining brands. We don’t want to make choices that fail with a multicultural audience, so we often avoid thinking about it.
But every project is an opportunity to design for and with people from every niche, and to build loyal users who keep coming back to your content. A design methodology that is culturally responsive and attuned to what global audiences need and want is one that will win in the long term. Later on, we will look at facets of culture and how they can inform design decisions, but first, let’s start by defining modern digital audiences and their cultural experiences online.
Who makes up the web these days? What do they look for in the digital things they use? What role does the web play in their lives, especially when they span languages, time zones, and political systems? If we can answer these questions, we can see the web as a set of widely differing life experiences, a mix of cultures and technologies, instead of the monolithic “user” we so often default to.
Devices for all
A great place to start understanding those varied experiences is by looking at mobile phones, a common way people access the internet. The Pew Research Center (http://bkaprt.com/ccd/01-03/) conducted an extensive survey of device usage patterns in eleven countries and found the following:
- Phone ownership. Majorities in every country surveyed own their own mobile phone. Overall, about seven percent of adults share their phone, though that ranges from one percent of adults in Vietnam to seventeen percent of adults in Venezuela.
- Smartphone usage. In the surveyed countries, younger people use smartphones at a much higher rate than those older than fifty, except in Lebanon and Jordan.
- Dual device usage. Having a home computer and a tablet is uncommon for most internet users. Only thirty-four percent of those surveyed have access to either of these. In Jordan, half of smartphone owners report having no home computer or tablet; in Venezuela, only eighteen percent said the same.
This data on phone usage and ownership begins to show just how important it is to be connected for people in cultures across the globe. Mobile devices give people access to news and information, to each other, and to the world at large.
Access for all
The number of global internet users creeps upward every year. As of July 2019, the We Are Social and Hootsuite Global Digital Reports estimate that fifty-seven percent of all people use the internet (http://bkaprt.com/ccd/01-04/). That’s wild. When I was a child, such a thing couldn’t be imagined except in science fiction, and now there are 4.3 billion internet users around the world—and 366 million more of them have joined since 2018 alone!
But those numbers tell us there is still lots of work to be done to make the web a more welcoming place for everyone. Even as the number of users goes up globally, the benefits of connectivity are not being shared equally. The 2016 World Bank Digital Dividends report (http://bkaprt.com/ccd/01-05/, PDF) found several troubling statistics on internet access (Fig 1.3):
- In Latin America, fewer than one in ten poor households is connected to the internet.
- In the Central African Republic, a month of internet access costs more than 1.5 times the annual per capita income.
- In Africa, the median mobile phone owner spends over thirteen percent of their monthly income on phone calls and texting.
And yet, there is progress. For people who traditionally haven’t had access to the web, technology is gradually bringing them online. Mobile phones are a key driver in the growth of internet connectivity across the world. A full ninety-six percent of the world’s population lives in range of a mobile network. The number of people using mobile phones grew by about 2.5 percent last year, and internet usage has grown by 8 percent since 2018. With the right device (and enough money, of course), almost anyone can now get online. That is progress!
When people do get access, the amount of information they consume also keeps growing. A report by Ookla on connection speeds and data shows that from 2017 to 2019, average mobile data speeds worldwide increased 21.4 percent, to 27.69 Mbps (http://bkaprt.com/ccd/01-07/)! With that speed, users can stream video, download music, and play online games.
These people are the modern web. They are the audiences you are either designing for now or will be designing for in a few years. While the numbers may seem huge and abstract, keep in mind that behind each stat is a person logging on for the first time, sharing pictures of their child with family overseas for the first time, or even making their first website on a mobile phone. Your modern audiences come in all shapes, sizes, life experiences, and opportunities, and it’s our job to design with their needs in mind.
The obvious way
When it comes to the mobile money experience, things get even more interesting. Thirty-seven percent of internet users, or 2.8 billion people, shop online. They collectively spent about US$1.78 trillion in 2018, which is slightly more than the entire GDP of Canada in 2017. They are what experience designer (and “closet anthropologist”) Stephanie Rieger calls a “highly globalized group of digital-first consumers” (http://bkaprt.com/ccd/01-08/). To accommodate these new customers, business models need to rethink how people spend money and how they access the internet.
Look at the success of M-Pesa, a platform designed to do just this. The mobile money platform is an essential tool for millions in East Africa, allowing them to save, transfer, and pay money all from their mobile phones. The Economist said: “To Westerners, ‘mobile banking’ is a new way of doing something old. To many Africans, it is the obvious way of doing something new” (http://bkaprt.com/ccd/01-09/).
Chinese “super-apps,” so called because they provide an entire ecosystem of ecommerce and social services, show us yet another “obvious way” that modern audiences are handling money online. The app WeChat enables its one billion monthly active users to pay school tuition, transfer money to friends, and even take care of rent. Barclay Bram, a PhD candidate in anthropology at the University of Oxford, wrote about his experiences using WeChat for even the most mundane transactions:
I open the Meituan delivery app and scroll through all the coffee options around me. I order an Americano. I have my WeChat linked with the facial recognition scanner on my iPhone; when I pay, I just hold my phone up to my face and a green tick flicks across the screen. Seven minutes later, I get a message telling me the coffee is on the way, with the name and number of the delivery driver. It arrives at 9:53. (http://bkaprt.com/ccd/01-10/)
The web keeps evolving. Old barriers to access are falling, smartphones are taking the place of feature phones and computers, and mobile networks are handling more data every day. As access increases, so does the number of potential users our digital experiences can reach.
Interfaces as Cultural Products
Nathan Shedroff, a strategist and executive director at Seed Vault Ltd., describes digital interfaces as “cultural products,” things that mediate our social activities (http://bkaprt.com/ccd/01-11/). Databases, websites, mobile apps, social-media platforms, and (especially) intranets are all systems that reproduce, revise, and amplify different parts of our cultures—both good and bad. They illuminate our preferred methods of communication, our social rules and perceptions, and even our aesthetics.
The most well-designed interfaces reinforce the ways that people like to communicate with, and about, each other. How a society talks about itself can take many forms, but some of the most interesting examples are sites that focus on reproducing and communicating culture from a government’s perspective. These act as digital products that serve millions of people—all with differing motivations and goals, but all in need of a shared understanding of their country’s norms and laws.
GOV.UK is one such example. In 2012, the United Kingdom set out a radical new vision for how the government would interact with citizens. The Government Digital Service (GDS), a group of strategists, technologists, and designers, built GOV.UK to provide a single point of access to government services (Fig 1.4). No longer would information about value-added tax, or land registries, or immigration regulations be buried under poorly designed and overwhelming HTML pages.
For instance, the GDS worked with Her Majesty’s Land Registry to create a single database of all land restrictions on things like conservation areas and historic buildings (http://bkaprt.com/ccd/01-12/). Working with 316 local authorities, they replaced the inaccurate, manually updated registries with a new single format for the entire country. They rewrote the registry interface in plain English, and made sure all information was automatically checked against a mapping database and standardized. Along with many other transformations like this, the GOV.UK site helps citizens mediate their interactions with the government and eliminates the capriciousness and slowness that many cultures associate with bureaucracy—inspiring similar initiatives all over the world.
Culture can also change the way people perceive your design and content. Different societies may expect different things from digital interactions, interpret information differently, and hold different mental models than your own. Don’t assume those mental models are fixed within the same country or culture, either. (This is another reason why research is so important, which we’ll talk about more in Chapter 4.)
Your users’ varied mental models can impact the layout, microcopy, or even information architecture of a site. In a 2003 study, researchers compared the mental models used by Taiwanese and American consumers when shopping online for home goods (http://bkaprt.com/ccd/01-13/, PDF). Participants in Taiwan had better recall and made fewer errors during testing when using a thematically organized taxonomy, where products were grouped according to their location in a house (Fig 1.5). For the American participants, a functional taxonomy—where products were grouped according to their purpose—was more effective (Fig 1.6).
In a 2013 study, Bijan Aryana, a senior lecturer at the Chalmers University of Technology, analyzed how Turkish consumers interacted with the default contacts app of their Android phones and how that behavior was connected to their cultural norms (http://bkaprt.com/ccd/01-14/). Turkish culture is collectivist (we’ll talk about that more in a moment), emphasizing the power of in-groups—families and communities—to provide safety, stability, and connections (http://bkaprt.com/ccd/01-15/). Because of this norm, many Turkish users value the ability to quickly access their in-group contacts in an interface.
Aryana asked research participants to draw their preferred organizational system for their contacts list (Fig 1.7). In contrast to the alphabetical organization of the given Android contacts app, the participants wanted the ability to:
- organize their contacts into hierarchies, and
- create an in-group of contacts, based on frequency and importance.
When I spoke with Aryana about this research, he pointed out that users wishing to modify and adapt standardized taxonomies is a common theme, but can be missed by teams that don’t research cultural perceptions: “For any design project, there is a need for fresh research rather than relying on what we assume we know about a culture, community, or social group.”
These two examples show how our interfaces should offer humans the mental models they are familiar with; research can help teach us what those are. The more we can align our taxonomies and interface structures with our users’ expectations, the more effective our designs will be.
Culture has a huge, yet often overlooked, effect on what we consider aesthetically pleasing. It’s common for Western designers to point to concepts like rational type systems, clean lines, an absence of decoration, and mathematical layout grids as universally “good” design without realizing that most of those principles originated in the century-old Bauhaus movement.
Founded in 1919 in the city of Weimar, Germany, the Bauhaus school, and later movement, aimed to create a union of all the arts, from architecture and product design to typography and graphic design. It was a clear cultural response to the hugely destructive First World War; Western Europe craved order, simplicity, and a new way of thinking about modernity. In the New York Times, Nikil Saval wrote that “the Bauhaus began as a protest against the thoughtless direction of industrialization, the harm it did to mind and spirit” (http://bkaprt.com/ccd/01-16/).
Aesthetics are a function of the products and experiences we design. But rather than base them only on WEIRD cultural principles, you should base them on what modern cross-cultural audiences find pleasing. And that’s a challenge, since beauty and form are perceived differently across cultures; we cannot actually say that our design work, in and of itself, has aesthetic value (http://bkaprt.com/ccd/01-17/).
So how to apply this thinking to your work? Start by asking yourself some pointed questions about what constitutes “attractiveness” to your audience and what will present your content in the most aesthetically pleasing light. Evaluate those cultural preferences across major visual design elements:
- Space. What information density feels most appropriate for your audience? For example, in Japan, web pages with high information density are considered functional and aesthetic (http://bkaprt.com/ccd/01-18/). This preference goes against the Western aesthetic impulse to include ample negative space. Consider the homepage for Japan Post, the national Japanese postal service (Fig 1.8). The page in Japanese and the page in English both contain the same functional elements: logo, main navigation, package tracker, and so on. However, the information density of the Japanese page is almost double the English one, containing many more icons, text, links, and even two competing logos in the header.
- Imagery. What images reflect the impulses and biases of your audience? Attractiveness will rarely extend to putting literally attractive people in all of your site images, but it’s worth noting that people love to see themselves reflected. Some cultures may have a taboo against certain images or expressions of beauty. For instance, in Nigeria, using your left hand to give things to people is taboo; a photo of a customer using their left hand to give money to a shopkeeper would stand out as a faux pas.
- Color. What palettes create a harmonious effect in the eyes of your audience? In the Japan Post example, red is their primary brand color, but it also matches the color of post boxes all over the country, creating a familiar connection for the audience. Color also carries culturally symbolic meanings; the color green is sacred in Indonesia, for example, but in American culture, it often signifies money.
- Type. What typographic scales and hierarchies are most natural for your chosen culture? Your readers’ language direction may be left to right, or right to left; what does that mean for your interface? The density of certain scripts such as Japanese, or the relative length of words in a language like German, means that your type scales, letter-spacing, and line-heights will need to be adjusted to achieve optimal aesthetics. (We will look at this in more detail in Chapter 5.)
Ask yourself these questions from your own cultural point of view, but then take a step back. It is up to us, as designers, to understand and then account for those aesthetic preferences. What you value in a graphic design aesthetic will not always match what others think is beautiful.
Designing for a global audience goes beyond crafting culturally relevant aesthetics. You need to take your users’ identities—social class, gender, religion, and even personal preference—into account.
When we talk about people’s identities, I want to make it clear that they are not fixed. Identities shift and change over time, and culture, which defines those identities, changes constantly too. However, there are specific lenses we can use to examine how people view themselves, and how their cultures view them (http://bkaprt.com/ccd/01-19/, PDF). Humans of all stripes come “preloaded” with a number of identities that shape how we view ourselves and others—and how others view us. These facets include:
- social class,
- individual identity, and
- algorithmic identity.
As we design culturally relevant digital experiences, we will need to take these factors into account. Let’s look at them in more detail from a design perspective.
Race is based on physical characteristics: skin color, eye color, facial structure, hair color, and so on. Your racial identity is something imposed on you by society. It’s not a choice you make; it is how you are seen by others. In every society, racial identity is used as a tool to maintain the power structures of some groups, oppress others, and assign value.
Our design choices—particularly photos and illustrations—can either reinforce racial dynamics or confront them. Stock photo sites like Tonl make it easy to find images that affirm the diverse racial identifies of global audiences (Fig 1.9).
Ethnicity refers to an affiliation with a set of cultural markers—such as language, tradition, religion, and geographic region—that you see as part of your identity (http://bkaprt.com/ccd/01-20/). People often confuse racial and ethnic identity, but they aren’t the same thing; racial identity is something imposed and enforced externally by society, whereas ethnic identity is less institutionalized, and more about individual history and cultural participation. For instance, look at the ethnic identities of basketball star Giannis Antetokounmpo. He was born in Greece to Nigerian parents, grew up there in a Nigerian household, and now plays for the Milwaukee Bucks in the United States. He belongs to and understands his Greek, Yoruba, Nigerian, and American ethnicities. Identities can be a bit complicated, right?
Your gender identity is your understanding of how you feel in relation to being male or female. It comes from an internal place, and may or may not match your biological sex, or the gender role society has assigned you (http://bkaprt.com/ccd/01-21/).
Gender roles refer to the behaviors, attitudes, and expressions that society expects from people of the same gender identity or biological sex. Similar to race, these roles are externally enforced.
As designers, one of the most important things we can do is enable accurate representation of our users’ gender identities. In their article “Trans-inclusive Design,” Erin White, head of digital engagement at Virginia Commonwealth University Libraries, suggested several tactics for giving people control of their names and representation in digital spaces, including getting rid of “real name” requirements, building name-change processes into user profiles, and allowing users to erase their identity records from your database (http://bkaprt.com/ccd/01-22/).
A poor understanding of gender identity can warp digital projects. Even in markets where gender roles and identities are more rigid, it is still possible to choose design defaults and strategies that do not make assumptions. For example, the website for the health insurance company One Medical uses a registration form that lets customers add additional gender info as they see fit (Fig 1.10). The form field is accompanied by text reassurance that the information will not be shared; however, it would be even better to include additional context about how the data will be used, how it will be protected or encrypted, and what users can do to change or delete it in the future.
Social class is the level of society you belong to, by birth or aspiration. Every culture identifies social class in different ways, but they are all tied to wealth and power.
In the US, for instance, higher education is often marketed as a tool for members of lower-status groups to gain status (http://bkaprt.com/ccd/01-23/). However, the costs involved can be astronomical. The College Board, a nonprofit that helps expand access to college education, reported that average yearly tuition and fees at private four-year institutions was US$35,830 in 2018–19 (http://bkaprt.com/ccd/01-24/, PDF).
Paying for that typically means taking on huge amounts of debt. So when Wellesley College’s website emphasizes that a typical student will incur only US$16,000 of debt for all four years, that sounds like a pretty good deal for a chance to move up the social ladder (Fig 1.11). The site features images of young people wearing casual clothes, engaged in quiet study and discussion. They don’t display markers of wealth or power, like expensive electronics or accessories. The message of this page is that more aid and less debt will help students access “the world’s most powerful women’s network.” Access to influential people and social status is an explicit selling point for prospective Wellesley students, and it is marketed very effectively here.
Age identity can reflect how different age groups interact with one another, as well as how they communicate among themselves, building a shared identity to define their generation. I vividly remember my father sitting on our porch in rural Nigeria in the 1980s, listening to the Voice of America on a small AM/FM radio. These days, I do something similar, but I check my Twitter feed for news instead.
Younger generations are often quicker to adopt emerging technology, approaching it with different ways of thinking than the (often older) creators of that technology intended. Young people these days do much of their thinking and communicating in what researcher Julia Sefton-Green calls the “digital bedroom” (http://bkaprt.com/ccd/01-25/). For wealthy kids in WEIRD countries, there’s a good chance this maps to their actual bedroom, with access to high-speed Wi-Fi and connected gadgets. But for many other young people, the exploration of technology happens in different contexts.
In “Design Principles for Indigenous Learning Spaces,” researchers at Australia National University’s Centre for Aboriginal Economic Policy Research defined a set of design principles for Aboriginal youth as they explore digital technology. The researchers state:
There are few spaces in the public domain where Aboriginal people experience a sense of control… Although access to technology in many remote communities may still be mediated through a non-Indigenous ‘gatekeeper’, the emergence of affordable small, mobile digital technologies including mobile phones, [and] laptop computers has brought technology into the everyday lives of Aboriginal people, especially young people. (http://bkaprt.com/ccd/01-26/)
They recommend creating safe spaces for young people to learn and experiment with technology. As designers, our interfaces should create pathways for them to define themselves. The age of your audience can define how they interact with your digital experience, often in surprising ways.
Religious identity is how you define your spiritual beliefs, and how your community defines your spiritual responsibilities. In designing for people in religious (and secular) societies, you must be aware of the prevailing belief systems and mores, and how they impact social behavior and perception.
In 2018, the company ConvertKit announced a rebrand—a new look, a new brand strategy, and (their biggest change) a new name—to match their growing organization. Their CEO announced that ConvertKit would be called “Seva,” based on the old Sanskrit word meaning “selfless service.” Since the company’s mission was to serve their customers, the new name captured that ethos. ConvertKit had already spent close to US$500,000 buying a domain and designing marketing collateral, a design system, and a new identity system (http://bkaprt.com/ccd/01-27/).
However, seva (ਸੇਵਾ) does not simply mean “selfless service.” It is a holy practice, as Hindu and Sikh people quickly pointed out to company leaders; it means giving generously to others out of love. Using seva as a name for a capitalist service was offensive and out-of-bounds.
ConvertKit stopped the rebrand and apologized; the CEO acknowledged that he didn’t do his homework, and thanked the people who helped him learn, even though they were under no obligation to educate him (http://bkaprt.com/ccd/01-28/). Moral of the story? Don’t appropriate elements of religious identity—or you’ll end up spending a half million dollars to offend 1.18 billion people.
Your own view of yourself is shared by no one else. Individual identity is complicated, nuanced, situational, and ever shifting. As designers, we can manage this fragmentation by designing in ways that celebrate individual complexity. Don’t attempt to railroad your audience into preset identities. For instance, my LinkedIn identity is very different from my Twitter identity, and may even change day to day. Luckily, LinkedIn respects that difference; they don’t insist that I add a Twitter handle to complete my profile, or bombard me with warning messages when I don’t. Treating individual identity as fluid and contextual seems like the most honest way to go about building a web for all.
When individual identity moves online, it becomes filtered, categorized, and surveilled. John Cheney-Lippold calls this algorithmic identity. It refers to the “categories of identity covertly assigned to you by means of algorithmic analysis of the data that an organization such as a web analytics firm has amassed on you” (http://bkaprt.com/ccd/01-29/).
In Digital Dead End: Fighting for Social Justice in the Information Age, professor Virginia Eubanks talks about the experiences of women who got assistance via Electronic Benefits Transfer (EBT), a system meant to provide poor Americans with monthly financial assistance (http://bkaprt.com/ccd/01-30/). Because of the way the EBT system collects and tracks data about its recipients, it effectively creates shadow identities that are almost impossible for users to access, limit, or correct. This makes EBT recipients more vulnerable to the whims of police, caseworkers, employers, politicians, and any other powerful party with a stake in the EBT system.
As web professionals, we must be careful not to deliberately (or accidentally) enable this “hidden legislation.” Make sure users always have access to and control over their digital identities.
Just as individuals have identities, so do nations and cultures. Geert Hofstede is a Dutch cultural anthropologist who, in the early 1980s, interviewed hundreds of IBM employees in fifty-three countries about how their cultures were organized. Because all of the research participants were IBM workers, and therefore part of a single business culture, Hofstede felt he could put their differences down to something else—their national cultures.
Over the next thirty years, he developed his findings into a taxonomic view of cultural dimensions (http://bkaprt.com/ccd/01-31/). Hofstede’s work has been validating for me. As a biracial Nigerian who has lived on three continents so far, it’s comforting to see my own observations of different cultures backed up by a carefully researched framework. His research has helped me make better design decisions.
Let’s take a look at what Hofstede’s cultural dimensions are, how they relate to design work, and what we should watch out for. Later, we can ask ourselves if this kind of taxonomic framing is beneficial or not, but first, let’s lay all of the dimensions out:
- Power Distance
- Individualism versus Collectivism
- Femininity versus Masculinity
- Uncertainty Avoidance
- Long- versus Short-term Orientation
- Indulgence versus Restraint
The first dimension is Power Distance (PD), which refers to how less powerful members of a society both accept and expect that power is distributed unequally. Lower PD means that members of the society can openly question authority and the distribution of power. Higher degrees on the index mean that there is a clearly established hierarchy that is not openly questioned.
Societies with low PD see relationships between leaders and subordinates as much more equal. Businesses and organizations have less of a difference in salaries and worker status, and students are on a more friendly basis with teachers. Parents and children may not be equals, but the lines of respect go both ways. Ideas of equality are welcomed and expected.
In contrast, societies with high PD are much more likely to have extreme hierarchies in business and government, and with that, large discrepancies between salary and status. Parents and teachers demand respect, and children and students almost automatically put their elders on a pedestal. Inequality is expected and even promoted.
Let’s take a look at a real-world example. The University of Colombo in Sri Lanka displays high PD on their homepage by prominently displaying university leaders (Fig 1.12). A page titled People (under the About section) displays links to the chancellor, vice-chancellors, and other members of leadership. Almost half the news images on the homepage are of people receiving awards or commendations from senior staff members. In both imagery and architecture, Power Distance is highlighted and reinforced.
Questions to ask yourself as you design:
- Who gets access to this information? We often say that information on the web wants to be free. However, cultures with high PD will restrict access to certain kinds of content, or will need to organize information by credentials, degree, or status.
- What symbols of hierarchy or power will my audience accept and welcome? These can be moral, national, religious, or other widely accepted markers of authority, such as a Christian cross, a flag, a national animal, or even a political figure.
- What visual and contextual markers of expertise and authority should I use? Do those in power have their degrees, certifications, or signs of authority prominently listed? In the United States, it is almost a given that when visiting a doctor’s office, their diplomas and certifications will be hung on the wall behind their desk; how would this translate to a digital experience?
- How does the perception of authority affect the delivery of information? One of the trends you see in the West is AI-powered customer service. However, the medical advice of a chatbot on a healthcare site may not be trusted by members of a society that values high Power Distance, because the bot lacks credentials.
- What mental models and taxonomies will my audience recognize and accept? For instance, for the faculty section of a university site, a shallow and wide taxonomy (Home > Lecturers > All) may work better for cultures with low PD. For a high-PD culture, however, a deeper page structure that prioritizes leadership or executive officers (Home > Authorities > Department Heads > All) is a stronger model of hierarchy and seniority.
The second dimension is Individualism versus Collectivism (IC). In individualistic societies, people have loose social ties to one another. At work, they value personal opinions, challenges, and material rewards. Their relationships with others are centered around self-respect and honesty, and using guilt is the best way to affect personal change. On a governmental level, they believe in a strong right to privacy, limited use of state power, a strong free press, individual interests over group interests, and self-determination. Sound familiar?
Now over to collectivist societies. They are defined by close-knit webs of friends, extended families, and other social groups; everyone belongs, in some way, to a larger group. At work, people value training, and believe that work (and the mastery of skills) has intrinsic value. Avoiding shame, saving face, and maintaining harmony within the group are the best ways to affect personal change. Governments have tighter control of the press and the economy, and prioritize consensus and social harmony over personal freedoms.
In the IC dimension, Iran is a more collectivist culture. You can see those values expressed on the Iran Tourism Board website (Fig 1.13), through the headline “Guests are friends of God,” messaging that emphasizes hospitality, and a request to tourists to “spread the word back home.” The copy repeatedly uses plurals like “they” and the photos emphasize groups and relationships.
Questions to ask yourself as you design:
- How do I show evidence of success in society? Collectivist societies will respond better to group harmony and the achievement of sociopolitical goals, while Individualistic societies will be drawn to displays of personal success and material wealth. This is true in imagery as well: collectivist societies may prefer seeing people in groups, while individualistic societies may prefer images of people working or living alone.
- How do I handle personalization concerns? Are people comfortable providing personal information to websites and digital communities? For instance, while users from a more individualistic society may want to highlight personal achievements on a résumé or portfolio site (“I won silver in the 400m”), those from a collectivist society may instead want to highlight their group or team achievements (“My team medaled a number of races, including silver in the 400m”). Ensure that your UI and database don’t require people to enter information that over-personalizes their profiles.
- What voice and tone are appropriate for my audience? Is it party-line slogans and official messages, or assertive, even argumentative language and imagery that demonstrate personal viewpoints?
- How do my users make decisions? Are they motivated by external rewards, such as money or personal recognition, or by internal ones, such as the mastery of a skill? Do they want to avoid guilt, as in individualistic societies, or do they want to avoid social shame, as in collectivist societies? Consider how content elements like instructions, calls to action, and tutorials might change to reflect these values.
Masculinity versus Femininity (MAS) is the third dimension in Hofstede’s system. Keep in mind that the terms “masculinity” and “femininity” are based on traditional gender normativity, and are not the best labels for the qualities Hofstede was trying to describe.
The Masculinity dimension defines a society that prefers assertiveness, achievement, heroism, and toughness. Gender roles are strictly maintained, and there is little sympathy for those who are deemed weak or overly caring. At work, people value financial and social rewards for achievement or career advancement. The political sphere is usually dominated by men, and there is little public support for people who don’t conform to traditional gender roles and expressions.
By contrast, Feminine societies prefer cooperation, modesty, and quality of life. The vulnerable are cared for, and there is less competition for resources and rewards. Gender roles are not rigidly enforced. At work, people value good working conditions, collaborative and balanced relationships with their coworkers, and a promising and secure career.
In the case of parenting website BabyCenter, this dimension shows up as an automated part of the user experience (http://bkaprt.com/ccd/01-32/). When a user posts a message with an acronym (“DD”), the system automatically adds the meaning in parentheses (“dear daughter”) (Fig 1.14). This functionality helps clarify in-group language for new users, demonstrating care and attention; the language itself reflects a culture that is comfortable with shorthand and publicly expressing terms of endearment.
Questions to ask yourself as you design:
- How are opinions highlighted? For example, in a site that examines and explores a complex or controversial political or social issue, does the design portray politicians and advocates as strong and assertive—such as large blockquotes attributed to one person—or as modest and cooperative—like a series of nuanced Q&As, and avoidance of personal attacks?
- What voice and tone are appropriate for my audience? In a feminine-oriented society, where caring for others is a priority, the language may be more helpful and cooperative. For instance, links to support might be labeled “How Can We Help?” instead of a more competitive phrase like “Find Answers.”
- Does my imagery focus on cooperation and empathy, or achievement and toughness? For instance, a site for a new architectural development might feature photos of men and women in the act of playing with children, attending to a neighbor with special needs, or helping keep their public areas clean. Conversely, in masculine-oriented cultures, illustrations of residents surrounded by evidence of their wealth, having “won” at the game of life, may play better.
- Are competition and comparison a core part of the user experience? Shefit, a custom sports-bra ecommerce site, cultivates a brand identity around physical fitness and healthy competition. These “masculine” qualities show up in their imagery as well as in the gamification of their loyalty program (“Earn crowns for spending”) (Fig 1.15).
The fourth dimension in Hofstede’s system is Uncertainty Avoidance (UA), which measures a society’s tolerance for ambiguity and the unexpected. A high degree of uncertainty avoidance means a preference for rules, formality, structure, and absolute truths. At work, managers prioritize tactics over strategy, and there are more rigid behavioral expectations. People expect clarity in communication, and so are more direct and active in getting their point across. Differences are seen as threats.
On the other hand, weak UA cultures are more accepting of differences, and show less anxiety around new or unexpected things. People don’t tend to use aggression or strong emotion as a way to communicate. At work, people are more easygoing, and the business focuses on strategy for the long term over daily operations and tactics.
Designer Jenny Shen encountered UA challenges in an interface for travel deals app TravelBird (http://bkaprt.com/ccd/01-33/). The search screen displayed trips within a certain number of hours from a given location, but Shen learned that German users were not comfortable with this metric (Fig 1.16). Because driving times in Germany can vary greatly depending on the road taken, German drivers preferred travel offers to specify a precise number of kilometers. Shen’s team updated the UI to reduce those users’ uncertainty.
Questions to ask yourself as you design:
- How do I treat interface errors and navigational dead ends? While it is great practice to make your errors and microcopy as plain and clear as possible, blunt language may be seen as more reassuring in strong UA societies. Conversely, in weak UA societies, less explicit instructions may be acceptable in things like form validation.
- Does my IA encourage browsing and wandering, or targeted search and drill-downs? Cultures that are comfortable with ambiguity may appreciate an information architecture that encourages exploration (such as deeper, meandering paths and frequent links to related information). Stronger UA cultures may prefer shorter navigation paths and clear context, such as additional microcopy on links, or hover functionality that previews the linked content.
- What kind of control does my audience have over their browser and interactive experience? For example, do links open in new tabs so your audience can maintain their original position? For audiences that are comfortable with some uncertainty, this might be one way to give them more control over their browsing experience. On airbnb.com, navigating to an individual location or experience page means opening up a new link; for people uncertain about their next vacation, this can be a quick way to compare different locations without losing their original search parameters. However you approach it, make sure to user-test your assumptions, and weigh them against any accessibility or business concerns.
- How visible should I make complex tasks and information? Being presented with complex information and incomprehensible explanations can cause anxiety for anyone. For example, when viewing a financial account dashboard, people from cultures that see uncertainty as a threat may be overwhelmed by an initial view that displays all their investments for the past year; they may be comfortable, instead, with less information upfront and an option to view more.
- How are visual metaphors such as icons and color used? Users with low UA may not always need icons to be paired with text if they can keep navigating and exploring. On the opposite end, for high-UA audiences, visual metaphors should have redundancy and backup; pair icons with text to make the action extremely clear.
The fifth dimension from Hofstede’s research is Long-Term Orientation versus Short-Term Orientation (LTO/STO). This dimension was added more recently after research and feedback by Michael Bond, one of Hofstede’s contemporaries.
LTO societies see more value in looking toward the future. Members of these societies are pragmatic; they appreciate adaptability and commonsense solutions. They look to family and friends as sources of information and credibility. At work, they focus on gaining useful skills over time, saving for the future, and patiently adapting to market and cultural changes.
STO cultures see more value in looking to the past and present. People in these societies prefer problem-solving for immediate results, and cultural norms remain largely fixed over time. They see rules and traditions as their primary sources of information. At work, they focus more on quarterly results and near-term gains.
Banking websites are a classic example of long-term versus short-term orientation. The very concept of banking seems focused on the long term, but in reality, different cultures treat money differently. Compare the Citibank and Cathay United Bank websites (Fig 1.17). Citibank’s homepage, from the short-term-oriented society of the US, has marketing language about living in the moment, using credit cards for everyday purchases, and enjoying immediate rewards. Taiwan’s Cathay United site, in contrast, promotes saving, long-term funds for the next generation, and favorable exchange rates.
Questions to ask yourself as you design:
- Does my content prioritize short- or long-term engagement? For users in an LTO culture, a strategy that rewards return visits, tracks progress, and builds on past activity will fit well in their cultural model. By contrast, users in an STO culture may respond better to pressure to complete tasks or unlock achievements.
- How are cultural traditions and rules respected or adapted? STO cultures work to maintain tradition and “the way things were.” This trait can be clearly seen in the campaign logos from the 2018 election for United States Congress, collected by the Center for American Politics and Design (Fig 1.18). Note how many contain stars, stripes, or colors from the American flag; Americans have come to expect these traditional references in political branding. When introducing visual elements to a design system for an STO culture, you might begin to include these types of markers of tradition.
- How does my content and visual identity treat family? In societies with a long-term orientation, family is important as a source of information and a way to maintain social status over time. The photos or illustrations you use to demo products may be more appealing when they show people consulting with their family members. On the flip side, photos showing family members enjoying leisure time together may work better for short-term-oriented societies.
The final dimension is that of Indulgence versus Restraint (IND). This dimension refers to how much freedom people in a given society have to act on and fulfill their human desires. An indulgent society is one that allows its members relative freedom to enjoy life, participate in recreation and leisure, and pursue individual satisfaction. Work culture emphasizes employee comfort through perks and interpersonal relationships. On a societal level, people feel free to voice opinions, and free speech is protected as a core human right.
Conversely, restrained cultures control how their members satisfy their needs and wants. There are strict social, sexual, and disciplinary rules. Money is saved, not spent. The right to speak freely is deprioritized, and the maintenance of social order takes precedence.
Finance, once again, provides a strong example of this dimension in action. Nigeria ranks as a highly indulgent society, and PiggyVest, a personal-savings startup, is there to capitalize on those values. Its marketing messages focus on personal happiness, control over one’s destiny, and the ease of savings. The site showcases personal stories of savings success through customer testimonials (Fig 1.19).
Questions to ask yourself as you design:
- Is happiness part of your digital experience? For all the talk of surprise and delight as a core design strategy over the past few years, this may be exactly the wrong idea for restraint-oriented societies. When developing a digital strategy, consider how elements such as animations, humorous microcopy, and appeals to joy may fall flat with some audiences.
- Is it okay to smile? A persistent cliché in the West is companies choosing stock photos with lots of people smiling at nothing (women smiling at salads, elderly patients smiling at doctors, scientists smiling at their microscopes, and so on). That may be a default option when design teams have limited budgets and time, but consider how uncomfortable people in restraint-oriented societies might feel with images of smiling strangers on every page of your website.
- Are vices hinted at, ignored, or commented on openly? In restrained societies, behaviors considered vices (such as drinking, drug use, or sex) may be acknowledged only through euphemism or metaphor. In indulgent societies, they may be treated as a normal aspect of social life, or an individual’s private business. Any language, imagery, or brand direction should explicitly (sorry, pun intended) account for that. The landing page for Muzzle—a Mac app that automatically silences notifications during screensharing—displays inappropriate and sexually explicit sample notifications to demonstrate the discomfort that would occur if they were to pop up while you were on a work call (Fig 1.20). While the notifications make me giggle with secondhand embarrassment, the language and marketing—even as examples!—would be totally out of place in a restraint-oriented society.
- How are leisure activities marketed? A 2017 research paper showed that this dimension affects attitudes toward travel and vacation-planning: “Tourists from indulgence cultures may engage in more extensive information collection, analysis, be more astute in holiday and leisure related decision-making and have more expectations from the service business. This in turn may require a more intricate and careful design of tourism products and destinations targeting customers from cultures with high indulgence scores” (http://bkaprt.com/ccd/01-35/). Service-industry sites targeting indulgent societies therefore need to include high-quality images, testimonials, and video. Let’s be honest—it’s all about how good the vacation will look on Instagram.
Limitations of the framework
Hofstede is one of the most cited researchers in cross-cultural human-computer interaction literature. Much of his initial work was done with IBM, a large multinational corporation. At the time Hofstede first came up with his theory in the 1980s, the world was just starting to globalize, and businesses and governments were very interested in the social impact of cultural differences. However, the world has moved on, and I want to briefly note three criticisms that his work faces today.
First, Hofstede focused overwhelmingly on nations, not individuals. It can be extremely misleading to assume everyone from a certain country shares exactly the same outlook. There are multiple cultures and subcultures within every nation, and a wide variety of personal and social variance within every culture. Huatong Sun, the University of Washington Tacoma professor, explained that in order to prevent stereotyping in cross-cultural design, we need a design methodology that is more rigorous and human-centered (http://bkaprt.com/ccd/01-02/). Sun recommended that we get out and conduct fieldwork and ethnographic research, even if the findings are messy and don’t fit neatly into a system like Hofstede’s. Seeing our audience interact with our interfaces in a natural context is a clear way to understand their true needs (and to avoid perpetuating stereotypes).
Second, Hofstede’s dimensions make it easy to blame our bad design decisions on users’ cultural orientations. If an open mention of sex in a company’s branding is criticized by someone from a restraint-oriented culture, we might be tempted to blame that cultural outlook, instead of examining our decision to incorporate a subject that is taboo for them. We should think of culture as a lens through which our users see the world, rather than as a source of “faults” or user errors (http://bkaprt.com/ccd/01-36/, PDF).
Third, Hofstede’s dimensions struggle to offer practical guidance for modern digital experiences. Culture can be heavily fragmented by the digital medium, especially when interfaces must accommodate many different people from many different cultures. What happens when the experience breaks? For instance, in 2014, Facebook suspended the account of Shane Creepingbear, an assistant director of admission at Antioch College and a member of the Kiowa Tribe of Oklahoma, because his name “violates our name standards” (http://bkaprt.com/ccd/01-37/). My own cousin, whose given name is Senator, often deals with this same issue. When digital platforms enforce rigid cultural templates, they make it clear that not all users are welcome.
In a study comparing the geographic information systems of German and American users, Francis Harvey, a professor at the University of Minnesota, identified a similar flaw in Hofstede’s system: “Hofstede’s dimensions of national culture are a good basis for understanding the influence of national culture on organizations’ self-representation, but miss the actual practice of social activities” (http://bkaprt.com/ccd/01-38/). For instance, a culture’s views on children may be an important factor in how a daycare markets itself, but an overworked parent trying to sign up for emergency care is only worried about the website’s frustrating contact form and how much it’s going to cost them.
Hofstede’s work does have a purpose. If it didn’t, it would not be one of the most mentioned theories in cross-cultural research today. However, culture is a series of negotiations over influence, meaning, and language, and those happen when people actually get together and talk. No one sits down to compare their cultural dimension indexes—they’re trying to solve their technical, design, and content problems.
Digital design is the mediation of those interactions on the web. Cultural dimensions are great for comparisons, but they can lead us down too many blind alleys if we treat them as sacrosanct. Always test your work with real people!
Understanding the Global Web
Talking about people and culture means talking about complexity, about the ways that society affects, and is reflected in, the design choices we make. But it’s important to know who our global audiences are, and what motivates them. We’ve looked at who is online on the modern internet, how to understand their complex identities, and how a framework like Hofstede’s cultural dimensions, however flawed, allows us to gain a better understanding of motivations, worldviews, and potential preferences.
In the next chapter, we are going to look at how to conduct honest, culturally appropriate research with our cross-cultural audiences to get at how they understand and use the digital experiences we design.