I stumbled into an article the other day about a Twitter advertising company called Sling Digital. Sling promises to give Twitter advertisers more impact for their dollar.
What does this mean exactly? Sling Digital optimizes Twitter ad spend by offering trending, contextual equivalents. For example, let’s pretend you’re an iPhone case maker. If you’re going to advertise on Twitter (which is smart because it’s built around the interest graph), your first thought would be to buy ads that target a term like iPhone.
Chances are, the price of purchasing the conversation surrounding ‘iPhone’ is at a premium. This is where Sling steps in. They mine data in the social stream to figure out that people who tweet about their iPhones also tweet about apps, the iOS operating system, and solar panels. (I made those interest correlations up for the sake of demonstration.) Because the price of reaching your target audience by inserting yourself into conversations about apps, iOS, and solar panels is much cheaper, you save money (even after you pay Sling a percentage of your savings).
Bluefin Labs does something similar for TV (re: uncovering things like ‘this audience also likes’), as does the poor man’s approach to retargeting where website owners evaluate Quantcast statistics to figure out where to place display ads on websites. I’m sure there are a number of other examples, but it’s 2am on a Friday so I can’t think of them at the moment. If you can, I’d love to hear about your additions in the comments.
The point of this whole article though, is that we’re about to see this model explode across social sites. Users are creating more data than ever, and with user data comes the incredible power of discovery. I’m particularly excited for what’s next with natural language processing and pragmatics, as we’ve barely scratched the surface of what’s possible.