4 things you thought you knew about social media listening

In the age of a social media explosion, more and more marketers are beginning to learn and embrace the usage of social analytics to better effectively track, assess and optimise their marketing efforts.

However, the learning curve is a relatively steep one, and knowledge gaps and misconceptions still exist among brand marketers regarding the application of social analytics and social media research.

Photo credit: quiklit.com

Here are some common misconceptions encountered.

“The older folks, being my target audience, are usually not on social media. Hence, I don’t see the need to invest much on social media spend.”

According to the IDA Survey 2014, findings indicated that more and more of the older generation is going online via smartphones. In fact, the 50 to 59 age group, who said they used a smartphone to access the Internet in the last three months, increased by 30 percentage points to 76 percent last year. Hence, just because the older audience are not actively commenting online, it does not mean they are not consuming your brand’s social content.

“Social media monitoring just means that we monitor social networks.”

Aside from just the major social networks (Facebook, Twitter, Instagram, YouTube etc), most social media monitoring tools can crawl all sorts of online websites, including forums, blogs, news sites, review sites, and others.

What sets a tool apart from the competition depends on the coverage offered not just locally, but at a regional level as well. Take note that certain sites will have technical restrictions in how it can be crawled by a social listening tool.

“I can rely on social social listening tools to get an accurate sentiment sensing of the topics monitored online.”

Natural Language Processing (NLP) and Bayesian Classification are some of the most common methods of automated sentiment analysis utilised by most social listening tools. However, the human language is a complex one. It is an uphill and difficult process of training machines to analyse and recognise local slangs or sarcasm, various grammatical nuances and culture variations.

Hence, the accuracy level at which social listening tools are able to accurately pinpoint or determine the prevailing sentiment of a discussed topic is probably likely to be only at 30 percent to 50 percent. This is why it is crucial to have an added layer of a human intervention of sentiment checking. While it may be a more manual process, the accuracy of sentiment checks will be elevated, especially when it is done by a local analyst with the relevant contextual background.

“Social media research seems faster and more unbiased. We can rely on it alone for research and tracking purposes.”

Social media research cannot completely replace traditional marketing research as there still exists certain limitations of social data.

For instance, social media analytics may not include specific demographic information, such as geographic regions or professions or education levels — and some people may make up or hide their profile information.