In a recent Social Media Today post, Alexandra Cojocaru asks, How Predictive Can Social Media Analytics Get? She goes on to offer a few areas where she thinks predictive analytics can be applied to forecast the future using data collected from Facebook, Twitter and other sites.

Alexandra mentions a project from HP Labs that analyzes social media data to predict the success of newly relased movies. Likewise, I recently came across this article that says analysts from Indiana University are using Twitter data to predict future stock prices.

If those examples don't convince you of the predictive capabilities of all those tweets, consider the work of Jason Harper, an economist at Organic, a digital marketing agency. Using the basic concepts of velocity and acceleration from calculus, Jason measures how quickly an idea is spreading.

Jason has applied his model to online marketing campaigns and can successfully predict within a few days of launching the campaign whether it's on track to meet the sales goals associated with the campaign. He has worked with big consumer brands, including Jeep and Kotex, and - in one particular campaign - the client even used Jason's data to make immediate changes to the campaign that resulted in instant improvements.

Here's more about Jason's work, excerpted from an article in MIT Technology Review:

The model came about during his work for Chrysler. Harper homed in on how Jeep's TV commercials were driving traffic to the "Jeep Experience" website as well as the rate at which the website was triggering signup's to Jeep's fan page on Facebook. Then he tried to see if the social media activity had any effect on the number of test drives. Using statistical analysis software from SAS Institute, Harper came up with a correlation: that consumers who engaged with one of Jeep's online touch points were about twice as likely to schedule a test drive at a dealership. Since the auto industry is so focused around increasing test drives as way to reliably boost cars sales, this was a promising start.

The Agency Spy blog also wrote about Organic's work in the post, The hard math of social media.

Predictive modeling, of course, is not for the faint of heart. Bob Lucas from the SAS training department offers some tips in a post today that discusses the importance of the time dimension in predictive analytics. Likewise, author Mike Gilliland often discusses mistakes made in forecasting on his blog, The Business Forecasting Deal. Heed their advice, but don't let it stop you from starting a predictive initiative in your organization. There's a lot to be learned from these huge new data sources on the social web.

Photo by: Kraetzsche Photo