What Facebook Graph Search Means For Both Businesses & You
It’s been exactly a week since Facebook unveiled Graph Search to the world, its way of organising everything associated on the site and making it easily accessible for everyone. The term ‘social search’ has been thrown around quite a lot, but by looking beyond keywords and instead focusing on the individual, Facebook’s attempt will be looked at rather closely as it tries to graph everything in and around its site.
We had the opportunity to try out the new feature and after a bit of testing and experimentation, here is our guide to Graph Search and what it means for you.
From the outset, the only cosmetic change you can see is the redesigned toolbar at the top of the page. All status updates, messages and friend requests are now located to the right-hand side, along with privacy and profile settings. Ultimately, it’s neater and it makes the inclusion of search more noticeable.
First of all, it’s pointless comparing it to Google’s search engine. If you’re going to compare, compare it to Bing and that’s a debate that’s been well covered. Instead, look at it as an experiment and one that many other sites will be looking at with interest. True, Google brought in “Search Plus Your World” into the mix almost a year ago, but it had difficultly meshing both social and organic together in a coherent way. Graph Search keeps the two separate – Bing serving the function of the traditional search engine – which keeps things straightforward.If you had to compare it to Google, you could say that while Google looks at what you said, Graph Search looks at how you say it. Or at least that’s the theory behind it, by finding results that are more relevant to you and your friends, their quality increases. A better analogy is to say that it works a lot like Advanced Search, as you’re presented with a number of criteria that can help narrow your search.
The major purpose of Graph Search is that instead of typing in keywords, Graph Search requires you to put in sentences and phrases. So phrases like “Friends who like social media,” and “Restaurants in Dublin, Ireland” are order of the day. At first, it’s a bit strange since we’re so used to typing in one or two words into a search bar, but you quickly get used to it. Failing that there are extra search options that you can access, which we’ll get to later.
If there’s one word to describe Graph Search, it’s ‘context.’ Everything you search for will be relevant to your own likes, your friends’ likes and your friends of friends. As social search is built around the concept of people being more likely to try something out if their friends already have, it’s a powerful tool to use that will only grow and blossom once Facebook graphs the entire site.
The main criteria that you enter in this are friends’ names, gender, relationship, employer, current city, hometown, school and friendship. There is the option for a more advanced search which goes into considerably more detail covering topics like work & education, likes & interests, photos & videos, life history and relationships & family. It’s this advanced search that tells you what kind of plans Facebook has for this and how Graph Search is really at the early stages of life.Beyond that, you can get contextual search results relating to those people or pages that you’re viewing. Seeing what they like, what their interests are and photos of them opens an entire mine of information that you can access. Granted, you have to search within the constraints of Facebook itself, but even at this early stage, there is a lot to work with especially if you consider how far-reaching the term “friends of friends” is.
However, there is still much room for improvement. For now, it seems that the only thing taken into consideration is the main name of the page or app. In all the searches we tried out, at no point did keywords in the actual page feature. It’s unlikely that Facebook will include these in future searches since the core of its search experience is personalisation.
Also, search results are limited to content you or your friends have liked. This is fine for now as Facebook graphs the entire site, but once it completes this, any further refinement will have to look beyond this metric. Likes on their own doesn’t give you the entire picture, so the ideal aim would be incorporating likes, engagement, social relevancy and keywords into search. It’s a very ambitious request, but high aims are necessary if it wants to be regarded as a viable search engine.
The purpose of Facebook’s graph search is to find search results that are relevant to you or your friends. Therefore, any of the searches you undertake has to reference what you’re looking for and whether it relates to you or your friends. When there aren’t any relevant results, the search engine resorts to using Bing, which is a good idea while Graph Search is in the early stages. However, at the moment, it feels like it’s a section that has been tacked on, especially when you consider how it’s presented.More importantly, since this is a new type of search engine, Facebook’s recommended search suggestions is teaching you how to use it. While a traditional search engine like Google would let you enter in keywords, Graph Search presents search ideas to help get you started, not bogging you down, but training you for a different way of search.
You can search by entering one or two words, but to get the most out of it, you will need to search for phrases and provide context to your search for it to be effective. It also allows you to come up with the most convoluted search imaginable (e.g.: friends who live in Dublin, Ireland, who like Manchester United, who visited Starbucks, who like Coke-Cola, etc.), but you would imagine that most searches will be kept down to three terms or less.
It also answers the question of when Facebook would develop a search engine for its app store. Incorporating it into the actual search engine was a logical choice, and will make discovery easier for users, but it also needs to find the middle ground between recommendations and presenting entirely new content.
After it refines the service, the next move Facebook will want to make is to move this search bar to smartphones and tablets. It’s the obvious next step in the refinement process, and for good reason. It’s a continuously growing sector and one that Facebook is placing greater importance on, but mobile search will bring its own set of problems. For one, smartphone users won’t want to type in all the details that’s required already, so Facebook will have to change or modify how users interact with Graph Search on these devices.
This is a move which will take place further down the line, and it’s a possibility that Facebook is redesigning its app (a rumour that had been circulating before it announced Graph Search) to incorporate this, but the company may be reluctant to release it for mobile until more of the teething problems have been fixed.
(Note: The above image is a mockup of what Graph Search could look like on mobile).
Since the only type of keywords that are relevant in this search bar are names, businesses and brands would be encouraged to fill out their profiles as much as they can so they have a better chance of being found. The key to Graph Search is context: If brands don’t provide that or just the bare minimum, then the chances of them appearing in any search that goes into specifics is very slim. Ultimately, what brands put into it is what they’ll get out of it so it’s up to Facebook to convince brands that this is what’s best for them.
While encouraging users to do this will be difficult, brands should be happy to do this since they want to be found by as many people as possible. Even just looking at search, you can see that it’s designed to give as much context as possible behind your search, even if likes are the main metric it judges them by.The second place that may be of interest for brands is advertising. Considering that a significant portion of Graph Search is dedicated to pages and places, there’s a lot of room for Facebook to monetise this if it wanted to. Of course, its first priority is to grow the service and make sure it’s ticking along properly before it attempts this.
Currently, there are no ads found on it, but the places it could look at are as follows:
1) Sponsored Search: The most obvious idea, and an idea that has already been covered before, these would be placed first or second in the search results, similar to how sponsored stories are presented in the news feed. Alternatively, it could do something similar to Google and place ads that match the general keyword – restaurants in Dublin might bring up a similar business that your friends like or have recently checked into. If you’re on your smartphone and you’re searching for nearby cafés that your friends liked, then seeing these types of results will help your decision
2) Sponsored Images: An avenue that Facebook is unlikely to head down, but it is an interesting thought. Considering that images get the most engagement in the news feed and the fact that Instagram is going down the route of monetisation, it could be an area that Facebook might consider. Think about it, you’re searching a friend’s album where they were in Paris. Alongside their photos, you will see other photos from brands, maybe one or two choice shots from an airline.
3) Contextualised Services: This is something that can’t be explored until Graph Search improves significantly, but the potential treasure trove of data awaiting Facebook when it does can be mined for other purposes. Things like job hunting and dating would be the obvious ones, but what about travel, where potential destinations could be suggested to you based on your likes and previous destinations? This could work in the form of open graph apps which take in the data from search and make your results more relevant.
Will it threaten the likes of LinkedIn, Yelp, Foursquare and other social media speciality networks? The answer to that is at the moment, no. Unlike these sites, Facebook’s search data is very much dependent upon what you put into it and that is both its greatest strength and weakness. While you’re always going to like content and post photos, you don’t really have a reason to share what job you have, where you checked in and what your relationship status is.
At the moment, Facebook hasn’t given users a good enough reason to enter in this type of data and will spend the next few months trying to convince you to do this or more likely, find other ways to derive this information from you. A way it could get past this is to team up with some of these social networks and sites to derive more information from them. Since many people use Facebook to sign into other sites, the scope for cross-platform data is already there if they don’t already have an open graph app.
Instead, what Graph Search will provide are clues in improving the user experience through contextualised search results. It will be a while before Facebook can competently mine that data to show job opportunities, restaurant suggestions, dating, etc., but there’s nothing to say this mightn’t be possible in the future.
Graph Search clearly has a long way to go before it begins to resemble the finished product, but like all things, it will take time. Google didn’t become the search juggernaut it is overnight, it took years of refinement and improvement to get to where it is today and even then, it’s still improving and tweaking the service to ensure the best possible search results. Facebook’s two main problems is extending its results beyond the ‘Like’ button and encouraging regular users to share more, but even without that, it still has a significant amount of data to work with before these problems become an issue.
Search is not a destination, but a journey and as users and rivals look at how Facebook and co refine social search, it will be a very exciting one for the industry.