What is Graph Search?A common trend among the major search engines (Google and Bing) is the provision of more personalized search results, or results that the user will be more interested in/are more relevant to them. The basic idea of Graph Search is that it can and will provide Facebook users with personalized search results when they look for something using the Search Bar. Facebook is just taking this idea to extremes by integrating results related to your social circle and your social circle’s circle. Basically, Facebook has made searching not only more social but more personalized and relevant to the user. According to the Facebook boffins, Graph Search will help users find what they are looking for through their various friends and connections.
How does it work?So, we know that Graph Search will help users find relevant things based on what their friends and social circle (on Facebook) have liked and recommended, but the question many users are asking is how? The older version of Facebook’s search relied on keywords and search filters to provide search results. For example look at the Facebook search (Image 1 below) for ‘Pizza’ conducted using the older version of Facebook (pre-Graph Search). Image 1 – Old Facebook search resultsFacebook, and most search engines, would read this as I am searching for a page, friend, group, etc. that contains the keyword Pizza and will return results with that word. The problem with keyword centric search is that we don’t think in keywords. If someone enters Pizza, they could in fact be looking for a nearby pizza place. Graph Search relies on more natural search queries to return results, meaning you type searches using the same sort of language as you would talk to a person. For example, ‘Pizza places nearby’ (results in Image 2 below) will tell Facebook to search for the nearest pizza restaurant to your current location. Image 2 – Facebook Graph SearchWhen a user searches for something, Graph Search, and the algorithm that powers it, will look at every word typed and determine what you’re looking for. This might sound familiar, as Google and Microsoft both use the same idea for their local searches. What separates these search results from say a local search on Google is that that Graph Search uses your social network to determine the results. What the Facebook engineers have done is essentially create a massive web that visualizes the relationships between you, your friends, pages you have liked, your photos, posts, and groups. It is these relationships that influence the results. Take a look at Image 2 above, Graph Search was able to figure out that I was looking for a nearby pizza place and so displayed the results and listed them by relevance. In this case, a friend has liked a place that’s fairly close to my current location which causes that restaurant to be displayed first. And this personal association is important. Because I have seen that a friend has been to a pizza place kind of near me, and liked it, my first reaction was, “I’m gonna have to check it out.”
What does this mean for my business?In truth, Graph Search is really aimed at public users. If you take a look at the Graph Search website, you can see a rotating banner that highlights the three main benefits of the tool:
- Find people who share your interests
- Explore your world through photos
- Discover restaurants, music and more