The Flaw Behind Facebook Graph Search
We had a class session today where we discussed SEO (Search Engine Optimization) and what is leading the charge for dominating your search results. Naturally, the conversation included the soon to be released Facebook Graph Search and how it will affect the way we look at reviews, search results, and our Facebook friends.
Graph Search is a new way to search Facebook’s massive amount of photos, people, and connections. It is important to note that this is not a web search, but instead is a reinvention of the Facebook search we’re all used to. In order to provide answers, Graph Search will essentially use an advanced tagging system, thereby allowing it to sort things like relationships, interests, and location to better refine search results.
Graph Search has been hailed by many to be a revitalization of search relevancy, and if I had to guess I’d say my professor would echo a similar sentiment to PC World:
Facebook’s Graph Search is the future of search…The hyper-personal search race has already been sparked and Facebook’s Graph Search ignites the revolution.”PC World
Quite frankly, I think Graph Search will be useful for finding people or photos, but when it comes it finding anything else it will fall miserably short. I do not believe Graph search can succeed because it relies on a tainted system: Likes
There used to be a time when pages were run by fans who purely had an interest in a brand, restaurant, or movie. But that has long since changed. Like any organization, Facebook is dependent upon profit (Especially now that it is a publicly traded on the stock market). Facebook understood that by allowing brands to own their unique pages, they could work with them to drive fans to their pages and to increase ad revenue. In return for brands getting fans to like their pages, Facebook began incorporating these likes into the News Feed and into the sidebar ads. So brands began to run campaigns where users had to “like to enter” a contest. Contests and sweepstakes paid out to relatively few people and in return got millions of “fans” to like their page. This meant that every fan who liked it was a new source of advertising on their friends’ News Feeds.
This brings us back to today. Graph Search relies upon likes to help determine what suggestions to make when you search for “Mexican restaurants nearby that John likes.” How accurate of a result are you getting if John liked Taco Bell’s page because they offered him free tacos for clicking a button?
The missing factor is intent. How are we to know if John actually likes Taco Bell and thinks you should too, or if he thought free tacos was just a great deal? Intent is what makes user reviews on sites like Yelp trustworthy. The intent of Yelp is to provide reviews and recommendations of local establishments; not to draw in advertising revenue.
For Graph Search to be truly effective, it will take a change in attitude by Facebook users, and how easily do you expect that to happen?