WeChat Data Report Reveals Unreliable Nature of Chinese Tourism Data

WeChat recently released a brief data report, unveiling user behavior and trends on its services in 2016. Sporting some 846 million monthly active users and a range of features ranging from social networking to consumption, WeChat may possess the best dataset in the world on Chinese consumer habits—justifiably making it a relevant data source for everything Chinese consumers. However, its travel data is so far off the mark that it prompts the question if any statistics on Chinese travel can be trusted. If WeChat can’t figure out where Chinese tourists are going, who can?

While clearly more of a promotional tool than a resource for academic research, WeChat proudly presented its travel data during a “pro workshop” in China to later distribute it to partners and other stakeholders through its official, English-language, channels. As can be expected when a platform of WeChat’s magnitude releases user insights, the story was quickly picked up by both domestic news outlets, overseas media such as Business Insider, CCTV International, as well as various more niche outlets covering topics such as technology and travel.

The elephant in the room is that the data is, without question, completely misguiding—at least insofar insights on Chinese travel are concerned.

WeChat’s 2016 data insights into Chinese travel preferences.

There are more reliable numbers than others. The gold standard, used by organizations such as the World Bank, the United Nations, and the European Union, comes directly from official statistical offices around the world, such as the U.S. Department of Commerce. Depending on reporting practices, such data can be more or less in-depth, but the bare minimum is generally to report the number of arrivals of different nationalities, nights spent in the country, as well as seasonality with monthly intervals. While certainly not in-depth enough to serve as a minute-to-minute dashboard that covers all types of data about international travelers, it acts as an indispensable guide for both domestic and international stakeholders and authorities.

When it comes to China, things become a bit more involved. A fundamental, yet often unanswered, question is the definition of China. Does it include the Special Administrative Regions (SARs) of Hong Kong and Macau? How about Taiwan? With a few exceptions, the consensus is to treat mainland China as a separate market—a practice that makes sense for historical, economical, and practical reasons. After all, mainland Chinese people hold a different passport and face different visa regulations than what for instance Hong Kong travelers do. While China has many political reasons to highlight the “Chineseness” of Hong Kong and Macau, it also treats these places as overseas destinations in its reporting. The official number of Chinese outbound tourists in 2015, the latest number to be released by Chinese tourism authorities, stood at 120 million and included journeys to Hong Kong and Macau, number one and two on the list of top destinations respectively.

Now, since WeChat doesn’t provide any actual arrival numbers for the destinations it lists as the top destinations in 2017—it instead provides some sort of indication with differently-sized bars—we can’t really say anything about for what destinations the numbers add up, and for what destinations they don’t. What we can do, however, is to put WeChat’s top destinations in comparison to the numbers reported by the destinations themselves.

The number of arrivals of Chinese nationals between January and October 2016 according to the official statistical offices of each country.

Assuming that WeChat considered Hong Kong and Macau “domestic destinations,” we’re left with the United States, Taiwan (which it labels as Chinese), and Japan rounding up the top three destinations in 2016. This would indeed contradict official data by a substantial margin—especially when taking the height of the bars into account. According to WeChat, the United States received more than double the number of Chinese visitors than the closest runner-up, and some 400 percent more visitors than Thailand which ranks as the fifth most popular destination.

The real story is that Thailand ranked the most popular “non-Chinese” destination in 2016, receiving approximately three times the number of Chinese tourists than the United States did in the same period. Given the recent crackdown on Chinese “forced shopping” tours in Thailand, many industry voices have questioned the reliability of the government’s figures altogether, but even if statistical foul play was involved, the sheer number of flights connecting China and Thailand dwarfs the number of connections between the United States and China, making such a dramatic shift an extremely unlikely event. No, the United States did not receive hundreds of percent more Chinese tourists in 2016 than it did in 2015, and no, Thailand’s tourism crackdown in the fall of 2016 didn’t cause Chinese tourism to Thailand to drop to zero. Even if not a single Chinese tourist visited Thailand after the crackdown, the number of Chinese tourists it had received by the start of the crackdown would still outnumber the full-year arrival number in the United States.

Similar things can be said of WeChat’s second top destination, Taiwan. While the severity of China’s imposed limits on tourism to Taiwan has been widely overstated, it still falls far behind other East Asian destinations such as Japan and South Korea—not to mention Hong Kong and Macau. However, Taiwan performing better than Japan and South Korea would require Chinese arrivals to double more than twofold, a highly unlikely event given the political fallout between the Chinese and Taiwanese governments.

Without picking WeChat’s top destinations apart any further, how could it go so wrong?

The Chinese government doesn’t publish any destination-specific data, making a “top list” of destinations a politically less controversial topic than for instance contradicting the total number of Chinese tourists, which is a number the Chinese government usually publishes on a yearly basis. All-in-all, political factors seem unlikely to have influenced WeChat’s ranking, with the United States a clear winner, despite outrage over everything from Trump’s anti-Chinese rhetoric to freedom of navigation patrols in the South China Sea. Likewise, stories coming out of China about limiting tourism to both Taiwan and South Korea for a multitude of reasons have been many throughout the year.

It could be the case that WeChat leverages the number of social media posts made by Chinese nationals in each country to gain an idea of the number of tourists visiting each country. Long-haul destinations such as the United States, where tourists spend longer, would likely generate more posts per visit, while popular weekend getaways such as the SARs, Japan, and South Korea would see the opposite result if using such a methodology. If overseas Chinese are also taken into account, the vast number of Chinese students at U.S. universities could also skew the numbers in the United States’ favor.

Methodology, as it happens, is key to getting an accurate picture of the state of Chinese tourism, and it would appear that it is in this area where WeChat failed miserably with their report.

The reasons such studies, despite their flaws, gain significant traction among stakeholders and media around the world are however easy to explain. They’re free, easily digestible, clearly presented, and come from a famous Chinese company that can claim user numbers in the hundreds of millions. The data should be an almost perfect snapshot of Chinese tourism, yet it isn’t.

The most reliable data, however, is unexciting and usually hidden in some government spreadsheet. Generally limited to arrival numbers, it doesn’t necessarily tell stakeholders too much about their potential Chinese visitors either. For marketers, more interesting data such as demographic insights and up-to-date trends are best extracted from focused surveys, perhaps in a particular city or for a particular market segment. To get something representative for the whole Chinese population, the survey would have to reach so many people that the cost is simply unlikely to motivate the benefits, especially considering how trend-sensitive (and politically sensitive) Chinese travel still is. Other estimates and surveys, no matter how interesting they may seem, usually tend to rely on a very limited number of survey respondents, making it difficult to draw reliable conclusions, or are pure guesses based on historical growth rates combined with assumed impact of events reported in the media. As the cases of, for instance, South Korea, Taiwan, and Thailand in 2016 would prove, media tends to have a distorted picture of reality—also, ironically, because of the lack of good data. If the methodology seems questionable or if the data is completely out of line with official estimates, the results of relying on such data are probably questionable as well.

Until Chinese companies like WeChat, UnionPay, and Ctrip up their ante with Chinese tourism data, tourism stakeholders seem to be stuck with quite bland, but highly reliable, data. Bland data, however, is better than unreliable data, even if the latter is the data going viral on social media platforms such as WeChat.

WeChat did not respond to requests for comment on its 2016 data report.


Market Analysis, Tech