Challenging assumptions and designing across cultures
A conversation with designer Senongo Akpem
Senongo Akpem is the author of Cross-cultural design, a book that shares how to design for the needs, perspectives, and expectations of multicultural audiences. He spoke to the Shopify UX team earlier this month about the principles he shares in the book. We took the opportunity to ask him a few questions about the lessons and best practices of cross-cultural design.
You can learn more about Senongo and his work at senongo.net. This conversation has been edited for clarity and length.
What are some misconceptions that people have about cross-cultural design?
The first thing that people struggle with is why cross-cultural design is important in the first place. I think that there can be — especially in the West, but in any place — a sense of cultural superiority. People think “The way that we do it is objectively better, so why should we then make ‘concessions’ to others?” That can be very dangerous thinking, but that’s the first hurdle that a lot of people need to jump over. Speaking as a Nigerian, many Nigerians might say, “Our way of doing X is better. So why should we make concessions?” That attitude is common, regardless of the country or part of the world.
The second misconception is that all you need to do is switch some language or just translate your web pages and you’ll be okay. Of course that never does the job, since localization is a much more expensive process than people would expect.
If people want to improve how their website or product works across cultures, where are some good places for them to start?
I think the first place is in your own head. That sounds somewhat trite, but changing your frame of thinking to something which acknowledges your assumptions, acknowledges your bias, and then works to actively counteract that in the research and the way that you set up your projects. Something that I always advocate is based on an essay called Non-assumptive design that Dorothy Deasy wrote, which is non-assumptive design. It basically says that when you’re in your kickoff meetings, and when you’re talking with your team about what you’re going to do with the design project, and somebody says, “Well, we know that the users do this, or we know that our audience likes that,” immediately you can write that down as an assumption. Then you flip that around into a question. So instead of, “We know that our users like to go to their login page first,” you ask, “Where do you like to go first?” Now it becomes a research question that you’re going to have to follow up on and find answers to. There could be more than one answer.
The same thing will be true cross-culturally, where you’re asking people, “Where do you want this item to be grouped logically,” rather than, “We know that users always find toasters in the home goods section of our website.” So I think it starts there with more of a framework for challenging those assumptions. And after that, then you’re able to do more concrete research, research that identifies cultural interesting points and functions and peculiarities of language and so on and so forth. But you have to clear your palate first.
Are there techniques you’ve used that help people think more broadly about cultural influences?
Oh wow, there’s a million. If we want to focus on research techniques, there’s one that I always love, which is called a cultural probe.
In the ’90s, the researchers Bill Gaver, Tony Dunne, and Elena Pacenti wanted to find out how to better integrate senior citizens into the cultural life of a town in Italy. Instead of the usual approach of “Here’s a survey, go fill it out, let’s do a few interviews and then we’ll write our report,” they gathered the seniors together and gave them these little packets of information. They were maps of their town with little colored stickies and disposable cameras, little questionnaires, and other kind of ephemera which they could fill out very easily. And because it was all self-addressed stamped envelopes, they could then pop it back in the mail when they were done.
Essentially the questions in cultural probes are things like, “What do you like/dislike about your town? Please fill out a postcard about it, or please take a picture of all the things that you really like to do when you’re out and about.” Using this cultural probe, they were able to build up a first-person perspective of how these senior citizens were interacting with their city and the things that they do in their daily life. So I think there are some research techniques and tools like that which are more interactive and clever in the way that they approach how we do that culturally responsive research.
One of the things we think about in our design practice at Shopify is “tripwires” — how do we know when we’re drifting off course in our approach? Are there cross-cultural design tripwires that people can use for approaching problems the right way?
Non-assumptive research is a big one, making sure that things are written down and documented. Another, very interestingly, isn’t to do with UX research at all, but about the practice of design, the way that you approach type, color, layout, and so on. Designers have certain wells that we go to quite a bit for design inspiration. It’s not a bad thing. You have to have something to look at. There’s very few designers that can consistently create out of thin air, and it’s okay to gather your ideas and your thoughts and see what other people are doing. But if you’re looking at Dribbble or Behance consistently for your UI inspiration, note the names of the people and the locations of who’s posting the stuff that you’re looking at and you start to see patterns that are maybe counterproductive. So I think that sort of thing can help.
I’m the Design Director at Constructive, and years ago we worked with a healthcare company on a project. They sell anesthesia machines in, for lack of a better word, developing nations. Many of the medical devices that get donated to hospitals in developing nations don’t come with, for example, the correct power plugs. They don’t come with any software updates. They break easily in more rugged conditions and there’s no replacement parts. So this organization’s anesthesia machine is very easy to fix and they have a network of fix-it people all across sub-Saharan Africa who can go and repair them when they break.
We thought it would be a great idea to build a community for all those repair people where they can log on and talk. But when we approached the client about it, they said, “Well, there’s a WhatsApp community that the repair people all use.” So all we had to do was add the number for the WhatsApp group and that was enough for the fix-it people. Not knowing what your assumptions are— like “we should build yet another tool” — can actually work against you.
One thing that really struck me in your talk was the idea of designing your products to flex with people’s mental models. Can you explain what you mean by that?
One of the things I was fascinated by when doing research for my book was how much critical research has been done by human computer interaction researchers and writers, and how so much of that information hasn’t filtered out into the wider web design community. There’s a lot of people working in universities who take a much more academic approach to this question of how we interact with machines. I love the idea that computers are electric rocks, essentially. We’ve managed to electrify silicon and that’s what we’re interacting with.
Anyway, one set of researchers looked at information architecture in Taiwan versus the U.S., specifically with an e-commerce site. They wanted to see if people’s mental models were consistent across those two cultures.
They tested a thematic architecture and a functional architecture. Each was slightly different. Thematic architecture being the items for the home which are grouped according to a theme, whereas with functional architecture, the items for the home work grouped according to what they actually did and what their function was.
So for something like a toilet brush for your home, in Taiwan, users had better recall with the thematic IA, where things were grouped according to their location in the house. In a different culture, such as the Americans in the study, the functional organization (where things were grouped according to their purpose) was more effective instead.
That’s what I was getting at when you have to talk about more flexible mental models. How do you then build an information architecture which allows both of those structures to be true? When my locale is Taiwan and/or I set my language to Chinese, is my e-commerce architecture going to refigure itself in order to match that? That may mean more sales, but it also could mean a lot more dev work on the backend to get that even working correctly. Is it worth it? And so those are the types of questions about mental models and culture that I find to be really interesting.
When you’re thinking about designing global e-commerce experiences, are there things that people should take into account when designing their sites and their products to meet those different mental models?
It’s all of the additional information around the name and the picture of the product itself — the price, the date that something is posted, or when it will become available, the currency exchange rates in some countries that may not have an economy that allows people to purchase online in their local currency. They may be spending with dollars or euros. There’s all of these little pieces of information that, when presented incorrectly, can confuse your customers.
From a more practical design perspective, making sure that people focus on type. There are a lot of things that you can do to be aware of what happens when you use a non-Western, non-Latin typeface, and finding ways to choose typefaces that are similar across your brand is a really fun but difficult exercise. In the interest of non-assumptive cross-cultural design, it’s probably also a spot where you would want to pay an expert to help you.
If you’re doing something with Arabic typography, for example, you’re not only going to have to think RTL (right to left) for almost your entire UI, but you’re also going to have to think about which typeface itself is representative of your brand. And that may mean hiring an Arabic typographer as a consultant for a few hours just to give you some structural advice. When choosing icons, making sure that they’re consistent. I think all of those things are practical design explorations that you can do when you want to get into cross-cultural design. And they’re interesting too!
