Social data is information that social media users publicly share, which includes metadata such as the user’s location, language spoken, biographical data, and/or shared links. Social data is valuable to marketers looking for customer insights that may increase sales or, in the case of a political campaign, win votes. There are many types of social data, including tweets from Twitter, posts on Facebook, pins on Pinterest, posts on Tumblr, and check-ins on Foursquare and Yelp. Facebook for Business and Twitter Ads are two programs that help advertisers use social data to market to targeted users who are likely to be interested in their ads.
Breaking Down Social Data
Users voluntarily make much of their social data public, allowing companies free and easy access to it. If a company that sells tickets to athletic events sees that a user follows several sports teams, that company could target ads to that user to try to entice her to buy tickets to see her favorite team play. Another way a company can use social data is to provide timely ads based on recent posts, such as appliance ads for someone who has shared that they are shopping for a home.
With high-quality social data that is aggregated and analyzed correctly, companies can target ads to the people who are most likely to buy their products or services. Social data can also help companies determine the most effective places to advertise. Companies can refine their advertising further by narrowing their target audience by gender, language spoken, electronic device used, age, interests, location, and other factors. Social data not only helps companies acquire new customers, but it also helps them further engage with existing customers.
Analyzing Social Data
There are normally two steps to analyzing social data. The first is collecting the data generated by users on networking sites and then to analyze that data. The process of analyzing typically takes place in real-time — and that is then used to determine influence, reach, relevance, and other considerations. Businesses that use this type of data analysis have to keep several things in mind, including how to distinguish between social data and sentiment, time relevance (what's relevant today may not be tomorrow), quality (how impactful certain messages and comments are by specific people), and how viral activity starts and spreads.
Limitations of Social Data
Social data is imperfect for several reasons. It is limited to the information that users decide to share about themselves. For example, some users may not share their location or their gender, giving advertisers an incomplete profile to work with. Another problem is that many users on social media are not real users but fake robot, or bot, accounts. Even with real users, attempting to gauge their feelings about a brand or political candidate (called “sentiment analysis”) based on the comments they make is not always possible because many of their comments are neutral and algorithms can incorrectly classify comments as positive when they are negative and vice versa. Further, many positive and negative comments that are available are extremes, making it difficult to accurately evaluate how consumers overall feel about a product, service, brand, or political candidate.