Data is all the rage in marketing and media. It would be quite difficult to find a single marketing or media-related conference this year that doesn’t offer at least one panel in its program on data, and most of them feature two or more. Deterministic data (i.e., data that can be matched to a unique and identifiable individual, such as a name or an email address) is being touted as an untapped precious resource. But in this discussion regarding data, many marketers and media companies are missing the opportunities offered by social media data.

Social media data tends to be ignored for a couple of reasons. The first is that its deterministic dimensions are hidden, largely, behind the privacy barriers of the major social media platforms. As a consequence of these user privacy protections, the bulk of the social media data available for mining is probabilistic—that is to say, it offers the probability of being matchable to a unique individual. The most common forms of probabilistic data in use today are browser cookies and mobile device IDs. Generally speaking, data scientists will model these identifiers into segments using variables such as the types of websites associated with each of the identifiers respectively. These segments can then be deployed by media buyers and/or sellers in the execution of programmatic media sales and/or buys. 

This brings us to the second reason that social media data has received less attention. Most of the social media data available can’t be used for platform-agnostic programmatic media buying. But there are other use cases for data beyond programmatic media buying and selling, and social media data happens to provide very effective fuel for these use cases. Of course, effectively harnessing this unstructured consumer chatter can be a daunting task. Fortunately, advances in natural language processing—many of which involve machine learning and artificial intelligence—have made it possible to automate the analysis of this data to a large extent and thus derive zero latency and accurate and actionable insights. Once one has enabled this capability, there are a number of powerful use cases to pursue. Here are a few of them.

Audience discovery—Marketers often miss the key psychographic variables that distinguish different audiences among their existing and potential consumers by relying too heavily on traditional demographic variables. Organic social media conversations can provide clues about the different tastes, mindsets, and opinions that define distinct groups among consumers. These audience insights can then be used to tailor marketing, programming, or sales content.

Trendspotting—Exploiting trends is similar to surfing. One has to spot them, in advance, like the approaching swells of an ocean, before they actually become trends, just as one must select the right swell before it begins to break. Once trends have hit their peak, it becomes almost impossible to effectively capitalize on them without getting lost in the clutter. The key to exploiting trends is to prepare to appropriate them before they’ve been triggered—as how, in surfing, one must start paddling vigorously ahead of the wave before it begins to break. Monitoring social media conversations in real time with the assistance of an automated text analytics tool enables one to see trends before they hit their peak and ride the wave back to the shore.

Content optimization—Mining and analyzing social media data in real time provides marketers and content creators with immediate insight into how content is resonating with consumers and how consumers are responding. Content programmers can gain valuable insights into what consumers like and/or dislike and even discover new ideas and ways to modify and optimize content.

Companies should leave no stone unturned as they seek to monetize their data assets. Additionally, while they certainly should strive to capitalize on opportunities in programmatic media, they should also explore other use cases for their data—of which there are many. Social media data enables companies looking to develop robust data strategies to take advantage of many of these opportunities and should not be overlooked.