Social discovery is the sibling to social search. By definition, it is the act of a website presenting content that is predicted to be desirable or relevant to a person based on that person’s social preferences. I like to say it’s an algorithm that tells you what you want before you want it.
At the center of social discovery is Facebook. By encouraging people to “like” people, brands, and content, Facebook effectively controls an army of unwitting spokespeople. Your grandma Sue’s innocent mouse-click, indicating a fondness for The Joy Of Cooking, finds its way into an ad that reads: “Sue Smith likes The Joy Of Cooking,” with a picture of your venerable grandmother along with an image of the cookbook and a link to its page on Facebook. Without realizing it, Grandma has become an endorser, not unlike Michael Jordan sporting a pair of Nikes (but less tall).
There are two reasons social discovery exists. The first is that it is part of the natural evolution of the Internet, which has presented us information first in human-edited directories, then in a multitude of websites and blogs, then on social bookmarking sites like Digg and de.licio.us, and most recently on social networks. Now that we have so much data on our likes, interests, and connections, both social and work-related, we are ready for the next phase of information delivery, one that is even more convenient: the delivery of personalized results through search and discovery.
The second reason social discovery exists is that it is certain to be extremely profitable for the companies that get it right. Facebook in particular needs social discovery in order to succeed as a megabusiness. Whereas Google search is overtly linked to commerce because it is the tool people use to research anything they want to buy, Facebook is a utility people use to connect with friends, and there are far fewer commercial associations with friendship. There are, however, some, and Facebook has explored a number of the natural avenues. Gaming is one of the most obvious ones, because people love to play games with their friends. Serving ads for group-buying sites like Groupon has been another good revenue driver because people love to get deals with their friends.
Transitioning Facebook from a purely personal space to a business tool, which occurred in late 2007 with the introduction of brand pages and ads, was an essential move for Facebook. As I mentioned in the previous chapter, Facebook’s ad-targeting capabilities give it a true differentiator in the advertising world. However, Facebook’s ultimate success will depend on the public’s embrace of buying products and services they weren’t looking for, that they serendipitously discovered and fell in love with.
If we break social discovery down into its most basic form, it is essentially a massive assembly of labels—or, in Internet speak, tags. Tags are notes that help us expand our knowledge of objects on the Web, especially objects that a computer can’t fully analyze. A picture, for instance, might be tagged with a location, resolution, camera type, and a description about what is happening in the picture. Any modern camera could automatically generate all those tags except for the last one. Subjective descriptions are still well within the realm of human-only knowledge, which is why tagging exists.
In an ideal world, we would tag every picture, video, and experience we have online so social sites like Facebook could have a full description of every person, place, and thing we care about. It could then easily cross-reference people’s experiences algorithmically to suggest great new content for them. Because people aren’t willing to do all that work for the benefit of social networks, these sites have had to invent clever ways to get people’s opinions on things. Hence the invention of likes, +1s, upvotes, diggs, and other one-click indicators of preference. The point is to make it so effortless to supply data to websites that plenty of people end up doing it.
By substituting general positive/negative votes for the detailed descriptions they would ideally like, social networks miss out on a ton of social data that would help them advance in social search and discovery; but it seems they have concluded that if they asked the public to do anything more complicated than clicking a Like button, they’d get far less data. I think that remains to be seen. After people see the new world that social data can open up for them, I believe they’ll be more willing to share information. That is why sites like Quora (a high quality question and answer site), which are social but require a huge output of information from its users, work: there is a feeling of community, that by putting out good information, you will get good information in return.
- “After people see the new world that social data can open up for them, I believe they’ll be more willing to share information.”
Indeed, Facebook is gearing up to ask more of its users. At the same time it consistently pushes users to accept a less and less private environment, it requests more data from them. If you think back a bit, you’ll see what I mean. A few years ago, Facebook just required you to fill out a profile and add photos. Then suddenly it asked you to “tag” your friends, which felt like a bit of work at first. Next, it asked you to indicate all your interests using the Like button. Soon Facebook will be asking you to tag again, but this time it will want you to tag products. In May 2011, the first glimpse of this next phase of knowledge transfer occurred when Facebook announced the capability to tag your pictures with the names of brand and celebrity pages.
I believe that Facebook will eventually decide that it has habituated its users to sharing data well enough that it can begin getting more information from the start. While writing this book, I predicted that Facebook would start asking you to categorize your social connections the moment you add them as friends. In September 2011, they began doing so, offering you just “close friend” and “acquaintance” as options. I predict these categories will become more granular in time: perhaps “family member,” “close friend,” “acquaintance,” and “business connection.” I believe most people will go ahead and categorize their social connections. After all, doing so feels like filling out a personal questionnaire, which is fun. Figure 3.1 shows an example of what this might look like.
Figure 3.1 A possible way Facebook will collect more information about the affinity of each of your friends so it can better deliver personalized data.
Facebook may even go a step further and ask you to rate how much you value each new friend’s opinion. If we aren’t comfortable sharing that information today, we will be a year or two from now. This is how Facebook will eventually learn who tastemakers are and enable social discovery to be more useful. In the coming months and years, expect to see continued movement toward social discovery and the commercialization of your social life.