The Trouble With Social Media Monitoring in Asia

Anyone who has worked in digital communications in the West, particularly the U.S. but also Europe knows that social media monitoring (or listening) is really the first step on the journey for brands moving into the social media space.

To monitor online conversation in the U.S. or the U.K., where English is the predominant language used online is really very simple. Select one of many listening platform tools from a variety of vendors, input your keywords, optimize the conversations returned, and set about deriving insights and actions from the resulting data.

It’s also surprisingly cost effective. A good tool is going to cost you in the region of US$500 – $1,000 per month. Add in analysis time and costs and for less than say US$30,000 per annum you’ve got a highly valuable market research data-set that can inform your social media strategy and content and provide a means of measuring the effectiveness of your efforts.

To undertake the same research and monitoring in Asia is significantly more difficult and costly however. I’ve seen many multinational companies make the [mistaken] assumption that what works in the U.S. (and to an extent in Europe) will work as well in Asia and set budgets accordingly. It doesn’t and there are three key reasons why:

1. Languages and character sets: Asia (like Europe) is a maelstrom of multiple individual countries and languages and (unlike Europe) also has the added complication of different character sets that differ from the Latin characters used in English. All this combines to make the conversation extremely fragmented, requiring specialist tools and native language analysts to be able to derive insight from the data, especially when it comes to understanding the sentiment of the conversations. This adds significant diseconomies of scale to Asia-wide social media monitoring, requiring a country-by-country, language-by-language approach.

2. Data collection: In the West a lot of the key platforms where social conversations occur offer up an API (application programming interface) to allow social monitoring tools to pull data directly from the back-end of the platform. This allows the tools to access a lot of key data easily (and cheaply). But in Asia this is not always the case. A lot of Chinese platforms for instance provide no such access and therefore the tools we use in Asia have to adopt a “screen scraping” approach of literally picking up conversations from the actual user interface of the social platform by periodically visiting the page and dumping the contents into their databases. This makes accurate data collection far more cumbersome, time consuming, and expensive. In addition, there are so many more platforms where conversations are likely to be occurring in Asia. Forums and BBS still host a large amount of conversation here in Asia as well as lots of home-grown market-specific platforms. Whereas you can largely rely on monitoring Twitter, Facebook, key online media sites, and blogs (found via search) in the West, identifying where the conversation is happening in Asia is far more difficult.

3. Sample sizes: Because of the issue involved in data collection mentioned above, sample sizes that are used to provide the reports in social media monitoring tools in Asia need to be closely monitored. This is a big problem I have found when using Western monitoring tools in Asia (even constraining the analysis to just English language). Because they are not able to collect enough data from a wide enough variety of sources, you can get very skewed samples giving you a completely warped view of what’s going on out there.

So what to do? My advice is to take social media monitoring on a market-by-market, language-by-language basis and accept it’s going to cost a lot more than you would expect to pay in the West. To adopt a region-wide approach is a false economy because the data you’ll get is likely to be unrepresentative of the actual conversation occurring and will thus lead you to the wrong decisions on how to communicate your brand. For multinationals requiring analysis of the conversation across the region, it’s going to get prohibitively expensive to do this market-by-market analysis in all markets, so prioritize key markets for ongoing monitoring and periodically “dip” into other markets as budgets and time allow.

Image on home page via Shutterstock.

Related reading

Vector clipart of Donald Trump giving two thumbs up
A screenshot of two tweets by Donald Trump about the United States' supposed 60 billion dollar trade deficit with Mexico.