Optimization for the search in the social graph on Facebook

on Friday, 25 January 2013
Since Facebook announced the Facebook Graph Search, ask many companies, as they can make sure to be found in the Facebook search. This is, in particular, from ORM view, a legitimate concern, I would like to dedicate this article.

As there is little experience with the new graph search, we have to rely on two things: Official Advice on Facebook, and the derivation of factors that are useful from an algorithmic perspective.

What is the Facebook Graph Search?

The new graph-search of Facebook could in the future, especially for local searches and recommendations (and thus also for companies with location-based services) are highly relevant.

Currently the graph search is enabled only for a limited number of Facebook users.

How do I sign up for the search in Social Graph?


 beta version of the social graph

Sign up for the waiting list for the beta version of the social graph search of Facebook

Under https://www.facebook.com/about/graphsearch You can leave you on the waiting list of the beta version.

Note: Currently, only users enabled to use the Facebook in the language "English (U.S.)".

What are the ranking factors?

The results of the search, according to Facebook based on a combination of several factors. If I put a query Facebook account primarily of the following things:

  1. What information does a company (always in the form of a corporate site) stated about himself?
  2. What information a company has geshared?
  3. How I am connected to the company through friends, likes, check-ins, friends of friends, friends of Likes, check-ins from friends, etc.?

Do I need a corporate website?

Even if a corporate website gives a huge advantage in the search graph, it is not absolutely necessary. It's enough already, when a company (or a subsidiary) was created by a user as a "place".

According to Facebook at least the following types of income are included in the graph search:

  • People
  • Pages (including corporate sites)
  • Applications (Apps)
  • Places
  • Groups

Even the current Facebook search results of mixed people, sites, places, groups, applications, events and music.

How will I look on Facebook?

Facebook wants to enable different types of searches. For example, "car repair shops that have visited my friends." In this case, obviously check-ins of my friends would be a fundamental ranking factor. Or "computer stores, like my friends"; dominant ranking factor: Likes my friends.

Of course, simpler queries are (as) possible!

FGSO concrete tips

At this time, there are few practical tips for Facebook Graph Search Optimization:

First Design of the corporate website

On the business side, you have full control and you should use them. Select the following items carefully:

  • Name
  • Category
  • Vanity URL
  • Content in the About section
  • Address (for local relevance)

Second Social interaction with the fans

So far, social interactions were challenged with fans to create virality and reach. In order to excel in the new graph search, it is important not only to reach many people, but the right.

Give your fans a reason to check in with you and your content to Liken.

Especially Facebook Offers may be a good way to strengthen the social networking a local company side. But coupon promotions in Groupon-style , which will in a short time attracted many customers to the store (which will hopefully lead to a couple of check-ins) can be worth it if in the ranking in the search graph is permanently improved.

Facebook search on Bing

The following screenshot is a search for trickmillion.com from Bing and shows how a Web search will look on Facebook.

An optimization for the search graph can thus be worth double when you get a universal search integration with Bing.

Future of SEO for Social Graph Search?

Matt McGee predicted that both SEO and social media marketers will fall on the social graph of Facebook search.

I think the statement is correct. One need only recall the early days on Pinterest or virtually any other social network, or in any search engine with a corresponding range: spam always works.