Tuesday, January 22, 2013



Facebook: Advanced Advertising Analytics

Facebook, as a company, is really good at a lot of things. One of them is providing valuable advertising opportunities to businesses, and providing targeted ads to users.

At its inception, Facebook’s targeted advertising system was pretty straightforward. Advertisers could select certain criteria like gender, age group, or interests, and their ads would show up for users having those characteristics. That made sense, and because of Facebook’s massive user base, businesses used this system to advertise with decent results. However, in 2012, when Facebook became publicly owned and traded, this main source of income for the company needed enhancement.

Up to this point, advertisers were able define and track user’s self-prescribed characteristics. They could kind of judge how their targeting was, but a huge variable in the equation was that much of their data was simply what users said about themselves. They could only see if someone clicked their ad to get to their page, or clicked their ad to install their app. Up to this point, advertising on Facebook was a soft science. Enter the 2012 Facebook Ads Manager and Offers Analytics update, Page Insights. [3]


Now, advertisers are able to analyze their success by seeing, in detail, what users do after they are shown an ad. Do they mention it in a comment? Install the app from the advertisement? Spend app credit within that app? Share the company’s page with their friends, or tag it in a comment? Now, advertisers can optimize for all of that conversion data, and more. [2] It all comes down to getting an ad in front of the right person. For example, before, if ten people clicked an ad and liked a page, then the owners of that page would know that ten people like their page and could target their ads to those ten people. However, they don’t really know anything about those ten people, besides their interest (the “like”) and some demographic information. But with Page Insights, that advertiser might see “Hey, one of these users shares a ton of his “liked” pages on his wall, and constantly tags his friends at different businesses in the area. These other nine never do anything.” The advertiser could spend money on advertising to that one, valuable user, instead of him plus nine others, and likely achieve the same exposure.

When you run a service that relies on (read: completely depends on for continued existence) advertising, you better have a damn good system for providing it. Facebook, the brilliant company that it is, has exactly that.


------------------------------------------------------------------------------------------------------