3 Common Pitfalls of International Market Research (and How to Avoid Them)

Zach Brooke
Marketing News
Current average rating    
Key Takeaways

​​What? Quality international research is critical to doing business in foreign markets.

So what? Much of international research is fraught with the potential for misinterpretation if not translated and administered properly.

Now what? Use native researchers and translators whenever possible, make sure the survey can be administered on the most popular platform in your target market, and adjust for the response styles of the local culture.  

Oct. 1, 2017

The world is smaller and more interconnected than ever before. But don’t mistake that degree of closeness for homogeneity

There remain stark differences among humanity’s social groups, many of which can pose a problem for researchers seeking to gather data from several distinct cultures. 

No one wants to stereotype any culture, but it is a hard truth that significant differences exist between societies, which will lead to inaccurate and ultimately useless market analysis if left unchecked. To accurately compare brand performance across countries, it’s necessary to understand and incorporate cultural nuances into research methodology. Or, as Michael Czinkota, a professor of marketing and international business strategy at Georgetown, writes on his personal website:

“The culture of the region being researched will have an impact on how marketers conduct the research, what is asked and the length or form of the information received. The willingness and ability of respondents to spend time on the process and provide a free-form response are influenced by factors that include culture and education, the market conditions and the segments being studied. Cultural and individual preferences, which vary from country to country, also have an impact on research techniques.”

Any researchers wanting to gather samples across cultures need to level the playing field before conducting their questionnaires to ensure the responses are representative of actual consumer sentiment and predictive of future brand performance. To achieve that, Maarten Lagae, senior manager of insights and analytics at Landor, helps identify the most common issues that crop up in international research and presents common troubleshooting techniques. 

1. Translation

The most basic stumbling block to international surveys is often translating research questions into the language of its subjects. 

“Sometimes words don’t exist in other languages,” says Lagae, recalling a time he was hired to create a study for a tobacco company that focused on new types of product packaging designed to keep the products fresher. His employers wanted to know if these package changes would be well-received in Russia. His team went to work, only to learn that Russians use many variants of the word fresh, each with a subtle difference that is lost in verbatim translation.

“It took us a while before we discovered that we were asking consumers questions around taste, as in a mint cigarette,” he says.

Mannerisms and norms also can become garbled across cultures. Here in the U.S., the phrase “thank you” is employed as a matter of routine, a polite way to complete a personal interaction. Not so in the rest of the world. 

“In India, for instance, there is no such phrase as ‘thank you.’ If you use those words, it’s most often [seen] ironically or even sarcastically.” Lagae says. “If you want to express your gratitude in Indian culture, you won’t just simply acknowledge the fact that you’re grateful. Instead you’ll very explicitly ask for an opportunity to return the favor.” 

The best solution to overcoming unnecessary obstacles posed by language conversion is simply to hire local researchers to convert ask and answer data.  

2. Data Collection

There was a time when all research was done in-person, by paper or pencil, or, later, via telephone. Now, digital surveys are the primary means of conducting international market research. It may sound standardized at a distance, but there are great differences worldwide in the adoption of digital technologies. Platform variance among respondents places pressure on researchers to develop universal questions and visual materials.

At Landor, Lagae says, much of the testing infrastructure related to visual identities or marketing collateral is designed to be run on desktop or laptop. However, this poses a problem in areas of the world which have bypassed an entire generation of technological hardware and are encountering the internet for the first time largely from mobile devices. 

“Mobile is becoming the gateway to consumers, and in many of the emerging markets, if you think about Africa and India, marketers will talk about leapfrogging. Those markets skipped a phase where desktop and laptops are the main devices to connect to the internet,” says Lagae.

Lagae’s recommendation for this problem is to design the survey specifically for mobile. “Use less text in those markets for mobile devices,” he advises. However, the strategy runs into problems as soon as the researchers head north to Europe, where most survey questions will inevitably be translated into German, the language of the continent’s largest economy.

“Interestingly, if you translate a question to German, the German language has a tendency to use very long words, and it can easily add a couple of pages to your questionnaire and also [increase] the efforts that research participants will need to go through to read all of your questions and answer options.”

Given these idiosyncrasies, it may be best to conclude no survey is universally translatable between formats, and different versions of digital surveys need to be developed for specific regions. 


  Market Research Capabilities Assessment
READ MORE:
Market Research Capabilities Assessment
This self-assessment tool will audit your market research capabilities.



3. Response Styles 

Even if language and platform are accounted for, cultural upbringing can influence how people respond to what they are asked. Within certain identities, respondents are more prone to “extreme answering,” meaning they tend to use the strongest indicators of approval or disapproval at their disposal. A separate response pattern is called the agreement bias, or yea-saying, where respondents answer affirmatively to every yes-no question presented to them, regardless of their personal feelings. 

“That is linked to very specific cultures,” Maarten says. “When you do research in collectivist cultures—think China and India for instance—there’s a much greater emphasis on maintaining interpersonal harmony. The group is bigger than the individual, so it’s more important that the group is in harmony than it is to express your personal opinion. That leads people in those cultures to agree to whatever you’re asking them.”

Such places can be particularly vexing when it comes time to test new innovations or determine why a certain product is not performing well within the internal marketplace. When consumers in these countries are asked about a new product, they tend to tell researchers it sounds like a great idea, only to promptly ignore it once it is released within the culture.  

Fortunately, this problem has been known to market researchers for decades and many methodological advances have been developed to counter the yea-saying effect. One simple way is to benchmark new product research against historic scores of past releases. 

“If I’ve tested 1,000 products in China, I can compare any new product to how those products have performed in the past. If I have data from surveys on 1,000 products before they were launched, and I can check with sales data after they were launched, I can also correct how much of a discrepancy there is between what people will say they will do and what they actually do,” Lagae says.   

For marketers lacking the large databases necessary to perform such a test, it’s possible to unlock the truth about a product’s future success or failure with detailed statistical analysis. Taking all the generous, positive responses from a collectivist society, it is possible to normalize distribution to reflect that higher ratings are less compelling in China than they are in the U.S. or Western Europe. 

“You reconfigure the distribution of your data in a country, and instead of looking at the absolute differences going from a four to five, you look at the relative differences. Going from a four to five in China may be a much bigger deal than going from a one to two in the Netherlands,” says Lagae.


Recommended For You:
AMA PCM Digital Marketing Exam 3 Ways Marketers Can Navigate an Ad-free World Does Market Research Need a Makeover?

AMA PCM Digital Marketing Exam

3 Ways Marketers Can Navigate an Ad-free World

Does Market Research Need a Makeover?


 

 Sign Up For Marketing News Weekly

 
Get the best marketing thought leadership delivered directly to your inbox!



Author Bio:

https://auth.ama.org/publishingimages/zack_bio.jpg
Zach Brooke
Zach Brooke is a staff writer for the AMA’s magazines and e-newsletters. He can be reached at zbrooke@ama.org or on Twitter at @Zach_Brooke.
Add A Comment :
 

Become a Member
Access our innovative members-only resources and tools to further your marketing practice.