Doing user research in emerging markets
I learned to do user research in well-lit meeting rooms in London. I learned by watching the experienced researchers of GDS do their thing, talking to a selection of the British public and getting them to use our new website. Since I moved to WorldRemit, some of the user research I’ve done has taken a very similar format.
But — as an international money transfer company for migrants — only half of our service takes place in the UK (or the US, or Europe). So some of my work has involved getting on planes to Africa or Southeast Asia, and setting up research in a totally different environment to what I’m used to. I just got back from one of these research trips to Ghana, where we were talking to people who receive remittances from abroad and testing out a couple of prototypes.
With a few of these trips under my belt, I’ve learned a lot about what works and doesn’t work when doing research in emerging markets. I thought I’d share a few of those lessons.
1. Find a local partner
Other people will tell you that the best way to do work in emerging markets is to engage a global agency, like GfK, who will organise all the local contact for you. It is undoubtedly true that this would be easier, but it only works on a big budget (when I started, WorldRemit was a small startup) and it puts another layer between you and the local experience.
When I sat down to organise my first ever research trip abroad, I searched LinkedIn with the words ‘Philippines’ and ‘User research’. Only one name came back, Angela Obias. A few months later she was showing me round Manila.
Angela organised and led an insightful week of discovery research among Filipino recipients of remittances. Before I flew out we had a few calls refining the aims of the project, creating the target profile for recipients and writing the interview guides. While I was there she organised all of our logistics, conducted the interviews herself and arranged a translator so that I’d understand what was going on. Afterwards, she wrote up the research, infusing it with her local understanding of the culture.
It went even better than I had hoped. I was definitely lucky to find a great researcher on my first try. On other trips I haven’t always been able to track down someone local who is familiar with the kind of methods I want to use. Even so, it is better to use a local research partner. But if you find yourself in this situation, remember to make a greater investment of time up front in explaining exactly what you’re hoping to learn, what kind of people you want to talk to, how you want the timetable to run etc.
2. Assume nothing
When I do user research with people who live in London, I have a huge amount of shared context with each participant. I already have a good understanding of what their life is like, and frameworks for understanding their habits, beliefs and skills.
This framework just doesn’t exist when you sit down with your first participant in a new country. And that means you’re liable to make more mistakes in interpretation. So I recommend changing your interview style to start with questions that build a foundational understanding of the person and their lives.
I begin each interview by asking the person to introduce themselves however they want — where they live, their family situation, their work — and tell me about a typical day in their life. I’ll also ask them questions about their access to technology and connectivity, from what phone they have to how much data they buy a month. Sometimes this means we spend a lot of time on topics that aren’t the main focus of our research, but I’ve found that if you skip this step, you’ll find it much harder to draw the right conclusions about your chosen topic.
3. Have a local interpreter (even when you don’t need a translator)
If you don’t speak the local language, you’re bound to need local support in interviews. For example, in the Philippines, the local language is Tagalog.
In Ghana, almost everyone spoke English. But successful communication is about more than speaking the same language. Sometimes there were confusions of pronunciation and idiom. Or there were concepts or terminology that didn’t translate.
For example, our Ghanaian interpreter Walter was on hand to explain ‘Susu’, a traditional banking substitute a bit like microfinance that we’d never heard of. Even more valuable were his interventions when we thought we had grasped something, but he knew there had been a misunderstanding.
4. Split your time between contextual interviews and hosted ones
On that first trip to the Philippines, we only did contextual interviews. By this I mean we went to visit people at their homes and workplaces, and conducted the interviews there. It’s an effective way of accelerating your understanding of someone by seeing first-hand how they live, and gives you a rich set of information to observe in addition what they say.
But it’s time consuming. If you’re screening for uncommon characteristics in participants (e.g. receives money from abroad from a family member), it’s unlikely you’ll be able to arrange lots of interviews in a small area. Many cities in the developing world also have serious traffic problems, which will make journeys unpredictable and lengthy.
Nowadays, I always aim to have a combination of contextual interviews and ones that are hosted at a central location. When you’re travelling to their location, you’d be lucky to conduct more than three interviews in a day. When participants are coming to you, you can comfortably fit in six interviews, with time in between for writing up notes as well. Doing some of each gives you the best of both worlds, with rich understanding from the contextual ones and the chance to meet lots of people in a short space of time from the hosted ones.
5. Record everything as you go
After a full day of talking to participants, I am always exhausted. But I’ve learned that if we don’t synthesise our notes and talk over our experiences at the end of the day, the resulting insights will be much poorer, and we’ll struggle away with frustrating half-memories. ‘Didn’t he say his friend had been the victim of a scam? Or was that her? What was the scam, again?’
Jan Chipchase said
“Data, like milk, is best consumed fresh; the longer we take to analyze it, the more likely we are to lose the thread that connects it to its original meaning”
As well as collating your notes and insights, make sure to photograph everything. Take photos of people, roads, signs, forms, ads, shops, markets, phones, clothes. It will help your own memory, but it will also be a powerful way to tell the story to people back at home.
6. Adapt as you learn
If you are not doing purely discovery research and you have a prototype or product to test, plan in the ability to adapt what you’re testing as you go. When I test wireframes or websites in London, I always have the luxury of bringing in more participants the very next week if I want. But if your research trip is to a different country, the chances are that you won’t be going back for a while.
Ghana was the first trip on which I’ve been accompanied by a designer. The first few days of interviews confirmed some of our original assumptions and busted others. They gave us ideas for ways we might be able to build something better. It was really satisfying to refine our prototypes based on what we learned halfway through the week, and keep pushing our understanding forwards.
As people familiar with user research know, testing with 5 users will uncover about 80% of the insights. It would be a missed opportunity to travel to a different hemisphere to meet just 5 users, so make the most of your time by improving your prototype or trying a different approach after each set of 5. To adapt a phrase from Leisa Reichelt, if you haven’t learned some things you were wrong about, you haven’t been bold enough.
Conclusions
Working on a product that spans international boundaries is one of the things I love most about my job. It was initially daunting, but now I think it’s a huge privilege to get to travel around the world and meet people, learning about their lives.
I don’t have everything figured out yet. Indeed, every research project and every country has its own opportunities and challenges, meaning only so much generalisation is possible. And wherever you are, some things are bound to go wrong — for example the time I was in a minor car accident driving to an interview and had to spend all morning in the police station (as per the local law) rather than meeting participants.
But I have learned some important ways to make user research in emerging markets a success, and I hope they’re helpful to others who might be planning similar trips. Feel free to get in touch (@alicenewtonrex) if you’re in that situation and want to talk further.
