• by Laurie Sullivan , Staff Writer @lauriesullivan, June 24, 2013

Social media posts and search queries follow similar patterns during real-time events, but activities connected to the former tend to lead the latter by as much as 4.3 hours, providing enough time for search engines to index and process relevant content, according to Microsoft Research and Stanford University researchers.

The two groups released a paper detailing support for search over trending social media events after analyzing a series of query logs and social media data associated with hundreds of trending events. User interest in trending events diverges during peak activity, as the overlap between content searched and shared rises. The researchers use the findings to explore real-time social media content with traditional search results.

Search marketers should consider the findings a roadmap to create online content connected to physical events, whether related to weather or breaking news. It turns out that lag in time benefits search marketers because it gives people the opportunity to process the event before taking action or interacting with the related content. The findings show user interest diverges during the peak of activity for a trending event. It's reflected by a spike in searches and sharing of related content.

Studying how people interact with trending events in search and social media, the researchers collected trending events from Twitter Trends and Bing's Popular Now queries for six weeks, starting July 19, 2012. For each trending event, the group collected a dataset of matching queries and tweets from users within the United States.

The paper also describes how the group removed "stop" words from the Trends and Trending Queries shown to users, and matched them against all queries issued via the search engine homepage and all public tweets for one week before the trend appeared and continuing one week after. When all tokens appeared in a query or tweet the group considered it a match. For instance, "'Toyota Recall' matched the query 'Toyota Camry recall,' but not the query 'toyota recall.'"

Search marketers should consider the lag time an invite to participate in real-time content marketing, which shows how social media might better serve search marketers.