Chapter 4. Cross-Cultural Research – Cross-Cultural Design

Chapter 4. Cross-Cultural Research

In 2016, our agency partnered with a health tech company in need of a new digital home. Their business was unique: manufacturing anesthesia machines for use in hospitals in low-income countries, where repairing and maintaining medical technology can be challenging.

They managed this work through a network of technical specialists across seventeen countries. In our research and discovery phase, we identified this network as a key audience. The specialists needed access to repair documentation, updates, and other technical instructions from our client.

That insight got the creative juices flowing: “Let’s design a new portal for the technicians to log in to and access all their information. It’ll be great!” That is, until the client gently pointed out what the specialists really needed—the number to a WhatsApp group.

That was it. Their whole community already had a useful, intuitive system for sharing information and receiving updates. Rather than impose a new structure on these users, we designed a simple module for displaying contact info, including the WhatsApp group number (Fig 4.1).

Effective cross-cultural design means, in essence, creating experiences that flex and adapt with a global audience. To make clearheaded decisions about what audiences truly need, we have to conduct research. In this chapter, we’ll look at a number of ideas to help you do just that.

Fig 4.1: Instead of a complex community portal that would have required extensive product-design work and strategy, our client asked only for their WhatsApp community number to be displayed prominently.

Research Methods

One of the first questions you should ask in any cross-cultural design project is: “Who is my audience?” But how you ask (and answer) can be a bit tricky. I’m a huge fan of using creative research methods, ones that are flexible enough to account for cultural differences. I find the following tactics particularly interesting and useful. For each, I’ll describe the research technique, then offer suggestions for trying it yourself. Sound good? Let’s jump in.

Cultural probes

In 1999, researchers Bill Gaver, Tony Dunne, and Elena Pacenti began exploring “novel interaction techniques for increasing the presence of the elderly in their local communities” ( They gathered in Peccioli, a small Italian village outside Pisa, to meet with members of the community. After introductions, the researchers announced they had brought “a kind of gift” for the attendees.

They passed around blue plastic envelopes to each of the participants, who opened them up to find an assortment of booklets, maps, stickers, postcards, pictures, and even disposable cameras (yes, that used to be a thing). The mood of the room suddenly changed: tired, reluctant participants began to excitedly dig through their envelopes, chatting with their neighbors about the contents, and completing the prompted activities—like using stickers to mark visited countries on the included maps.

After finishing the activities at home, the participants were asked to fold up the packets and pop them in the mail, already stamped and addressed to go right back to the researchers. They called this technique a cultural probe—a way to engage audiences and provoke inspirational, open-ended, and multisensory feedback. Written, drawn, and photographed responses can shed light on users’ lives, thinking, and motivations, and reduce reliance on interviews, which can often fall short in cross-cultural situations.

There is something endearing about this low-tech, tactile method of learning from unfamiliar cultures and languages. As Gaver put it:

Understanding the local cultures was necessary so that our designs wouldn’t seem irrelevant or arrogant, but we didn’t want the groups to constrain our designs unduly by focusing on needs or desires they already understood. We wanted to lead a discussion with the groups toward unexpected ideas, but we didn’t want to dominate it.

Cultural probes can incorporate just about any task you can think of (Fig 4.2). Some researchers include individually wrapped activities with instructions to take photographs, fill out a diary, draw a picture, and so on. Others are more freeform or analog, like tear-off booklets and printed maps.

Fig 4.2: Cultural probes are an open-ended, multisensory method for gathering ethnographic feedback from your audiences (,

To conduct a cultural probe:

  • Identify the goals of the probe. What questions do you want to answer? Your goals will inform the tasks in the probe and the materials people will need to complete them. As with any workshop or other designed activity, you will need to be clear about the information you’re looking for.
  • Design the tasks. Write out instructions in clear, accessible language. Use the first languages of your participants. Consider listing out the materials that go with each task, in case they become separated.
  • Assemble the kits. This is likely to be a really fun experience for any creative team. Make a list of the things you will need to include (disposable cameras, maps, postcards, etc.). The more you can rely on light, recyclable materials, the better. Then set aside a few hours, get your team together, provide some food, and assemble away!
  • Distribute them. Doing this in person, as in the Peccioli example, is the best method. However, if you can’t be there physically, and your country has a reliable postal service, then send them off in the mail. To make it easier for participants to get back to you, give each task its own self-addressed, stamped envelope. That way the whole packet doesn’t sit around forever, waiting for one final task to get completed.
  • Synthesize the information. Once you get a critical mass of probes back in the mail, you can analyze the results. Use a spreadsheet or other UX research analysis tool like Dovetail ( to translate the responses into something quantitative, so you have numbers to inform your next phases of work.

Local facilitators

Traditional usability-testing advice tells us that facilitators must remain neutral at all times, speaking and interacting with participants as little as possible, to avoid skewing the test results. But this advice is based on a Western European model of dispassionate rationalism that does not take cultural issues into account.

What happens if we set that thinking aside, and instead use culture as the critical factor in how we conduct usability tests? I recommend using a local facilitator, someone with lived experience in the culture you are researching, in any kind of observational methodology.

Consider the results of a Danish study exploring this very premise ( Researchers recruited both Danish and Indian facilitators and participants in Denmark, ostensibly to test the usability of a clip art application. Participants were asked to use the application to create a Danish-themed birthday party invitation, while the facilitator encouraged them to think aloud about their image selection. The image set contained deliberately misleading images, such as reindeer (which definitely do not live in Denmark) and a Norwegian flag, to set up potential misunderstandings (Fig 4.3).

The results may surprise you! The researchers found that when the facilitator and participant shared a cultural background—both Indian or both Danish—they spoke aloud more, and were more likely to vocalize issues they found with the interface. They focused less on cultural differences in the imagery, and more on the task itself. In other words, local facilitators from the same culture as the participants ran more effective tests.

Fig 4.3: To test the effectiveness of local facilitators, researchers asked participants and facilitators to work with culturally specific images in a clip art application.

Not all cross-cultural research needs to be conducted by a local facilitator, but usability testing that requires speaking and interaction can particularly benefit from that shared cultural mindset. I also strongly recommend it in markets where participants use a lot of slang or speak in patois (such as Singlish, an English-based creole spoken colloquially in Singapore); having someone on hand who can understand and communicate in these linguistic forms will make your insights deeper and more culturally relevant.

To try this yourself:

  • Recruit locally. There are a number of places where independent UX consultants can be recruited, like LinkedIn. If you are planning to work with an agency or firm, get information on their international network before you begin the project.
  • Record the usability sessions. Keep note of when slang, euphemisms, or culturally specific knowledge is shared between the facilitator and participant. This may not be necessary to understand the results of the specific test, but you will start to build a bank of keywords and ideas to be used in site features, such as search optimization and microcopy.
  • Rework your time budgets to account for faster completion of the tests—you might be able to include more activities in each test, rather than signing off early.
  • Take note of the network. By setting up research systems with local facilitators at the start, you’re also laying the groundwork for future opportunities to collaborate with them. Keep contact information for everyone you talk to, even if you don’t end up working with them; you never know when they might be able to help out again.

The Bollywood technique

Sometimes you need to create a little drama to get results. When usability researcher Apala Chavan was testing a travel-booking website in India, she found it difficult to get honest feedback about the tool from her collectivism-minded participants ( Recall that this cultural dimension means that people tend to want to maintain harmony (not openly criticize a stranger) and defer to the group (not voice dissenting opinions). With a cultural framework like that, it’s no wonder Chavan couldn’t elicit feedback.

So she tried something different: she asked participants to imagine they were in a dramatic situation reminiscent of a Bollywood film:

The participant’s beautiful, young, and innocent niece is about to be married. But suddenly he gets news that the prospective groom is a member of the underground. He is a hit man! His whole life story is a sham, AND HE IS ALREADY MARRIED! The participant has the evidence and must book an airline ticket for himself and the groom’s current wife to Bangalore. Time is of the essence!!!
The participants willingly entered this fantasy and with great excitement began the ticket booking process. Even minor difficulties they encountered resulted in immediate and incisive commentary. The participants complained about the button naming and placement. They pointed out the number of extra steps in booking.

Because they were given permission to participate in the fantasy, they were also given permission to think outside their usual, collectivist social norms. Chavan called this the Bollywood technique.

To try this yourself:

  • Create a base narrative with broadly defined characters, roles, and resolutions. Many story lines are fairly universal, and mainstream movies use archetypal narratives that are easily recognized across the world (at least in places that watch TV and movies regularly). A Western analog to that dramatic Bollywood plot might be a heist movie or even a game show, where the participant needs to think and act quickly under pressure.
  • Create variations that match cultural dimensions. Using the cultural dimensions we looked at in Chapter 1, add variation to that base narrative. For instance, in a feminine-oriented culture, where people have more casual attitudes about sexuality, a narrative that highlights dating may be more resonant than one about a religious marriage ceremony.
  • Write out your testing scenarios and variations. Keeping documentation that details the stories you are creating, rather than winging it on the spot when the session starts going badly, will keep you cool and the project on track.
  • Use local facilitators. If you’ve never seen films from your chosen genre or market before or don’t know any of their famous actors, you’ll come across as insincere and untrustworthy, regardless of how compelling your narrative is. This is a good time to recruit expert help—as we saw earlier, participants who need this sort of scenario-driven prompt may feel more comfortable sharing their thoughts with a member of their own culture.

The Walking Havana method

In 2014, researchers Daniel G. Cabrero, Heike Winschiers-­Theophilus, and Hedvig Mendonca from the School of Computing and Informatics at the Polytechnic of Namibia (PoN) began researching unemployment in Havana, an informal settlement in Windhoek, Namibia (, PDF). One of numerous Black communities that were forcefully segregated by apartheid during the 1950s and ’60s, Havana is home to many marginalized young people and high rates of unemployment, as well as ongoing PoN projects to introduce technology and community centers that can support residents in employment, job training, and entrepreneurship.

Before anything could be designed, or even defined, the research team wanted to get a better understanding of the neighborhood of Havana and the challenges and opportunities for tech introduction. A common step for WEIRD designers would be to gather insights and quantitative data, head back to the office, and craft personas. Instead, these researchers took a different path—one they called the Walking Havana method.

Fig 4.4: Participants created sketches and collages showing proposed TV characters based on their own daily experiences.

The researchers gathered a small group of residents and asked them for help creating characters for an upcoming TV series based in Havana. Together, they walked through the neighborhood—ostensibly to scout locations for the show—while residents pointed out different issues and features of the community. This helped the researchers immerse themselves in the environment and the residents’ struggles with finding employment and economic opportunities.

Afterward, the participants proposed characters for the TV show. One group wrote a narrative called “Living Like Slaves—Havana Location” that featured nineteen-year-old Eddy, who had no access to electricity and cooked with firewood in his shanty. Another group created a collage called “Unemployed Youth,” made of newspaper clippings, drawings, and handwritten descriptions (Fig 4.4). It featured a woman named Tselestina, who struggled to find consistent work, and included references to her upbringing (the hero’s backstory), parents and other adults who counseled her (the supporting characters), and young people rioting because of a lack of jobs in the city (the extras). These characters were, in effect, personas created not by the researchers, but by the residents of Havana themselves, and they clearly embodied the experiences and economic challenges of living there.

A similar example comes from researcher Manako Tamura’s thesis project on cultural assimilation, code-switching, and race:

I hosted a co-creation workshop, where I invited users to share their pain points related to assimilation. The participants responded to my questions around cultural identity and race through creating collages of magazine cutouts. Using this tangible process enabled the participants (most of whom had never met each other) to open up about deeply personal experiences. (

To try this yourself:

  • Ask participants to name their work. In the Walking Havana example, participants gave their collages names that also functioned as TV show titles, in keeping with the researchers’ original framing of the exercise. Encouraging participants to “brand” their collages with descriptive names helps them build a narrative beyond some cutout pictures on a board.
  • Use appropriate images. If you ask participants to create personalized collages, they need usable source material. That may mean ordering magazines or picture books; don’t just grab a few issues of Vogue and expect the collages to feel authentic. (Unless, of course, your users are avid Vogue readers!)
  • Create space for sharing. If possible, gather your participants in a physical space. A community center, classroom, or even a large living room could all be places to conduct this kind of culturally relevant research. Make sure the ways you’re asking people to create and share are accessible for people with a range of physical needs.
  • Define what will happen to the work. Inevitably, someone will want to take their collage home. Define what the next steps are in the process, specifically which design artifacts or outputs participants can take with them. If logistics allow for it, take high-res photos of the collages for your records, then leave them in the community or with the participants.

Possession personas

In the northwestern part of Namibia, the ovaHimba are a group of people who lead a pastoral and seminomadic lifestyle that is increasingly under threat from urbanization and westernization. In 2015, they began working with a research group to create a system that would “collect, store, classify, and curate their Indigenous Knowledge (IK)” (

Their design partners wanted to create a database on tablets, so that the ovaHimba could record their lives and environments in a rich and engaging way. The responsibility for gathering data on the tablet fell to the community elder, who began to offer suggestions about a bag to carry it and keep it safe. Other participants jumped in to help, describing other decoration and ornamentation for the tablet bag.

While this may sound to some like a distraction from the task at hand, it wasn’t. By crafting physical fashion around the device he was responsible for, the man was embodying the user experience—beyond the direct usability of the tablet, he was demonstrating how it could be incorporated into his life.

Not every design project will call for this level of fashion detail, but there is a crucial lesson here. The things we surround ourselves with, our belongings and possessions, form a huge part of our lives. Just have a look around your room when you are at home—each one of the things on your desk, in your closet, in your bag, make up a part of who you are. Focusing on possessions in personas is a great way for cross-cultural design projects to incorporate a larger view of the user experience.

To try this yourself:

  • Ask questions about possessions. During research and inquiry, ask your participants about adaptations and ornamentation for their devices, possessions related to the project topic, and physical manifestations of their identities.
  • Discuss devices in personas. Make a clear connection between your users and the tools they use to access your content. Show devices—including screen dimensions, resolution, and representations of bandwidth—in your persona, so your team doesn’t forget what their view of the site will be like.
  • Show people with their things. The photographer and designer Jason Travis has a photo set called Persona, where he shows human beings with a neatly organized assortment of their things (Fig 4.5). Consider a similar approach that highlights the possessions you feel define the person.

    Fig 4.5: Jason Travis’s persona project shows how belongings can be paired with a portrait to give designers deep insight into the type of people they are designing for. Image courtesy of Jason Travis (

Bias, Brains, and Artifacts

Imagine your site user is a forty-two-year-old digital marketing manager with two kids. Even without additional details, your mind has already formed an image of what that user looks like, hasn’t it? You can’t really turn that off—it’s implicit bias, unconscious attitudes and stereotypes that affect your actions, perceptions, and decision-making. And it will trip you up every single time you conduct research or design work.

Implicit biases don’t have to be negative, but they are involuntary. They differ from explicit biases and stereotypes, things you know and act on consciously. Implicit bias creates ideas and feelings in your subconscious mind about people’s identities, like their sexual orientation, race, ethnicity, language, physical appearance, or age. The Kirwan Institute at the Ohio State University offers these definitions (

  • Implicit biases are pervasive. Everyone possesses them, even people with avowed commitments to impartiality such as judges.
  • Implicit and explicit biases are related but distinct mental constructs. They are not mutually exclusive and may even reinforce each other.
  • The implicit associations we hold do not necessarily align with our declared beliefs or even reflect stances we would explicitly endorse.
  • We generally tend to hold implicit biases that favor our own ingroup, though research has shown that we can still hold implicit biases against our ingroup.
  • Implicit biases are malleable. Our brains are incredibly complex, and the implicit associations that we have formed can be gradually unlearned through a variety of debiasing techniques.

Whatever design artifact you are working on, implicit biases about audience behaviors and motivations will creep in. It’s inevitable (see that first bullet). Any time you need to make decisions—about the language used in brand guidelines, the types of workshops that will engage your clients, or even the best code framework to recommend for an external developer—your mind will use those biases as a mental shortcut.

Clients, stakeholders, and team members bring their own biases to the project, too. Even as they’re working toward a shared vision of audience, the people tasked with using and referencing your design artifacts may interpret the same content differently, thanks to implicit bias.

Lene Nielsen, a human-computer interaction and business informatics researcher at the IT University of Copenhagen, investigated the ways different cultures interpret the exact same information in a user persona (, PDF). Nielsen wrote an ungendered persona description—all text, no image—about an online marketing manager, listing details about their job, personal life, education, media consumption habits, and proficiency with technology. Then she asked sixteen participants from India, China, and Denmark to:

  1. read the persona description,
  2. find a photo on the internet that resembled the description, and
  3. write a short explanation of why they had chosen that particular photo.

The participants all read and interpreted the same text and selected representative photos based on their own biases of what a “marketing manager” might look like (Fig 4.6). Many participants selected images of men; all the photos fit with a stereotypically Western concept of a businessperson.

Fig 4.6: A sample of the images that participants chose to embody the persona description (

When you make design artifacts like personas for your clients and teammates, be aware of your visual choices and how they are manipulated by your implicit biases. They show up in unexpected and unspoken ways, affecting the design work you are doing, and you’ll need to be diligent about sidestepping them. Let’s take a look at a way to do just that.

Avoid the lazy brain

The human brain is incredibly effective at forming stories, especially about the things happening right in front of our eyes. Most of our mental processing happens before we are even aware of our thoughts—and then the narrative is set, regardless of whether it’s true or not!

The world is a complex place; paying attention to so many details takes a lot of energy, so your brain finds ways to conserve that energy whenever possible. That means that when presented with a tiny bit of information, your brain will attach a narrative to help it keep things simple and understandable.

Let’s try a small thought experiment. Imagine you have just signed a contract with a client to create a tech solution for farmers in Bangladesh. Your client wants to explore ways of using data to manage farmers’ crops and market pricing options. Immediately, without much more prompting than that, I’m guessing an image has sprung into your mind, composed of bits and pieces of stories you have heard about Bangladesh. Already, that lazy brain of yours is forming a picture to help it with your project!

Based on that shortcut image, and with the addition of some perfunctory research (like looking at online stats about mobile tech adoption in Bangladesh), you could easily generate a persona of a “Fearful Farmer,” a woman uncomfortable with technology (Uncertainty Avoidance) and upset at what she sees as its time-wasting properties (Long-term Orientation). Think of all of the ways this mischaracterization will cause problems in developing technology to fit her needs!

Now let’s look at a real-world example that avoids this trap effectively. Paddy to Plate is a detailed report by Lauren Serota, a service designer and founder of Appropriate, about the rice ecosystem in Myanmar (Fig 4.7). The report is organized around archetypes like “The Common Smallholder.” Details about their economic situation, worldview, traditions, personal and professional relationships, and business practices help frame them as complex individuals operating in a complex cultural situation. Each archetype also includes a case study from a specific research participant, so you see real language and problems from a person on the ground, not just a synthesized persona (, PDF).

Fig 4.7: The archetypes in Paddy to Plate are authentic portraits of real people in rice farming and production in Myanmar. Each one is also paired with a detailed profile, complete with names, family descriptions, and quotes.

Culture lives and dies by the stories its members tell each other, whether or not they are true. Your lazy brain will happily ignore facts that don’t fit the narrative it’s already set up, so your challenge is to give it a new narrative, one that is not reliant on stereotypes and guesswork.

Deficit-framing vs. asset-framing

Regardless of the narrative the human brain spins up, it has a tendency to “other” people, focusing on their differences instead of similarities. Your brain subconsciously dismisses people you don’t identify as part of your in-group, seeing them as “not one of us.”

In digital projects, one result of this othering is deficit-framing, where we define an audience by a shortcoming. For example, a strategy statement like “We help poor farmers in climate-stressed countries use technology to grow what they need to survive” says nothing about the audience except that they have problems. That is their main reason for existing in reference to the design strategy you are laying out.

What can a designer do to redirect these conversations? Try asset-framing: defining the audience or strategy by a positive feature. It helps us short-circuit the brain’s tendency toward negative narratives, stay centered on our audience, and prioritize their success. A rewritten version of that strategy statement might look like this: “We help small landowners boost crop yields and build their communities’ climate resilience.”

Asset-framing is a term coined by Trabian Shorters, author and CEO of BMe Community, a US-based network of investors and fellowships for Black men and women (, video). Shorters explained that asset-framing is a powerful way for mission-driven organizations to tell forward-thinking, aspirational narratives about their work and the people they help. Asset-framing builds more multifaceted portraits of real people.

For cross-cultural design projects, try approaching your persona strategies like this:

  1. Identify positive truths of your audience. Asset-framing starts with descriptions and statements about your audience that don’t focus on negative things. You will, of course, need to explain specific challenges and issues your archetypes face, but start by framing them with positive qualities.
  2. Focus on your audience’s agency. Explain how they can overcome challenges by using their own ingenuity, community organizing, and power. Your service or digital platform then becomes another one of their tools, not their savior.

I have found this way of thinking extremely effective in my design work. By focusing my design strategy on aspirational stories about my users, I can craft artifacts, content, and interfaces that do not stigmatize or stereotype.

Culturally flexible artifacts

Because personas were originally developed to describe WEIRD audiences, they are sometimes too Western-focused to be usable in other parts of the world. We often don’t consider the political and social agendas that come with the traditional user persona, and in doing so, we make them much less effective.

Consider this sample persona for Spotify from the Interaction Design Foundation (Fig 4.8). Which of Hofstede’s cultural dimensions are visible here? The persona is named Rebecca, a popular given name in the West, and she has ready access to expensive technology. She is influenced most strongly by her own tastes, has only individual motivations, and uses “I” liberally. Not only is the character a clear example of the Individualism dimension, but the thinking that guided the creation of this artifact is deeply individualistic as well.

However, what about those households in other cultures where people are comfortable sharing smartphones or account logins with their family? If we were designing software and content for these family units, a persona structured like Rebecca would clearly miss the mark.

Personas don’t need to fall into that trap. With some cross-cultural forethought, we can use personas to creatively challenge narrow assumptions about our audiences. During the start of a client engagement for a legal nonprofit in New York, our agency put together a set of eight personas to guide our UX and visual design work. The artifacts were definitely on the traditional end—an image, text descriptions of the archetype, and some key challenges (not so different from the example in Fig 4.8). However, I saw them as a chance to challenge a few biases. I created personas for a wealthy donor couple, a husband and wife team who could contribute tens of thousands of dollars a year to the nonprofit—they were young, biracial, and had a Japanese last name (Fig 4.9). I didn’t need to justify these choices—the persona simply represented modern, global audiences. And if those choices help introduce more cross-cultural thinking into our business and design decisions? All the better.

Fig 4.8: In this sample persona from the Interaction Design Foundation, Western-focused cultural variables, especially Individualism, are in full evidence ( Would this be effective for a company serving a global consumer audience?

Fig 4.9: This couple would hardly be unusual in New York City, but they were a subtle opportunity to challenge traditional persona expectations.

Put People First

Defining our global audiences is not an easy task. We must examine our biases, understand our users’ cultures, and illuminate their lives and motivations. It takes some extra creativity to conduct cross-cultural research and build effective artifacts, but what are designers if not creative in approaching new challenges?

Enriched by our research, we can now make flexible, culturally responsive decisions about typography, color, imagery, and other elements of the digital experience. In the next chapter, we’ll look at how we can make design choices that put global audiences first, and keep them visiting.