Social advertisers are like goldfish, not elephants. Because social data is largely real time, it forces marketers to rely on their short-term memory. That’s fine for serving a Taco Bell banner to someone who just tweeted that they’re hungry, but valuable social signals aren’t only those communicated right now.
For example, when Target rolls out its Black Friday ads, the retailer may want to target consumers who tweeted about visiting Walmart on Black Friday last year. Social ad firm LocalResponse is making that possible.
Three months ago LocalResponse began testing the ability to turn back the clock on the social data it used to target ads, and on Wednesday it officially rolled out historical intent targeting. “Before we could only target based on relatively real-time data, [going back] a few weeks. Now it’s years,” said LocalResponse CEO Nihal Mehta.
LocalResponse has partnered with social data platform Datasift—one of three Twitter-certified data providers—to access historical tweets, which LocalResponse then mines for signals of intent to be used to target desktop and mobile display ads.
As with LocalResponse’s intent-targeting tool launched earlier this year, the company only digs through social data that users have made publicly available, which are usually tweets. If someone has shared a Foursquare check-in or Pinterest pin to Twitter, then LocalResponse can use that to target ads, Mehta said.
While Datasift only retrieves tweets from 2009, its competitor Gnip recently made available the full Twitter archive dating back to 2006. Mehta downplayed the three-year gap, saying that the volume of tweets sent from 2006 to 2008 is equivalent to one month’s worth of tweets in 2009.
Sony Pictures piloted the capability with a recent movie campaign. Mehta declined to share campaign results but emphasized how film studios could use historical targeting. For example, a studio could promote a film’s DVD release to users who had tweeted about the movie when it was in theaters.
LocalResponse’s historical intent targeting could help advertisers keep their targeting segments fresh. For example, the company could see that someone had tweeted years ago about being pregnant and derive that that person’s child is now a toddler and promote Gap Kids instead of babyGap to them. Or vice versa, Gap could set date ranges for the targetable data so that it only runs babyGap ads to anyone who tweeted about being pregnant nine months to a year ago.