Twitter is known for real-time data. There are too many examples to list that have surfaced over time that have proven the value in the here-and-now style crowd-sourcing information over the lagging traditional media and news outlets.
One example is the twitter account that was setup for the recent Waldo Canyon wildfire in Colorado Springs: http://www.gazette.com/articles/fire-141309-media-twitter.html.
Unfortunately the best data available is often the kind that is aggregated well after the fact. This is typically the type of data of which maps are made.
Here is an example where six days of tweets were filtered and aggregated to produce a Beer Vs. Church map by Chris Crum. This information serves a great purpose, but it is still only beginning to touch upon the type of tools to come.
Taking this type of data (real-time, streaming, feeds, etc.) and transforming it into information that can be visualized for better real-time decisions. This is what separates the good decisions from the best decisions. Here is an example where Bern Szukalski uses Yahoo! Pipes to transform web feeds into real-time map layers.
I envision a business intelligence (BI) software that has social media platform tools built in that will instantly map streaming data to reveal the patterns and allow intuitive visualization as they are happening!