Menu Icon
Menu Icon


Facebook Announces Graph Search Preview

Intouch Team

Posted by

Facebook recently announced a new, third pillar of core functionality, called Graph Search, to help users make new connections. This new pillar complements the other two pillars in Facebook: NewsFeed and Timeline.


Graph Search is an enhanced contextual search functionality that will expand on the current Facebook search bar. Originally, the search bar was used to find basic results for people, places and companies without any association to other content. Graph Search is positioned as a social search engine and will not replace the existing search relationship with Rather, it will complement it. If Graph Search is unable to find results inside Facebook, it will automatically search to provide any relevant results from the Internet.

The initial set of available Graph Search data includes:

  • People
  • Photos
  • Place
  • Interests

Graph Search is still in beta, and the rollout will initially be limited to several hundred thousand Facebook users. It is only available to individual users, works only in English and does not work on mobile devices. At their press conference, Facebook did discuss the desire to continue to expand Graph Search’s capabilities, including a future API. Future mobile integration is planned once they completely rewrite the iOS and Android applications. At this time, Facebook mentioned they have no support for building out voice search capabilities.

Advertisers are not currently permitted to advertise to users based on users’ Graph Search queries. However, Facebook has the option to make this functionality available in the future. While Facebook did not provide specific details about empowering business page owners with the opportunity to utilize Graph Search’s capabilities, it’s worth noting that they did not rule it out.

How Does Graph Search Work?

Graph Search is different from Facebook’s original search functionality and traditional search engines in three ways:

  1. Local Search: Recommendations vs. results. As user contributions grow, recommendations can appear based on the combination of friends, Facebook check-ins, “Likes” of local businesses and other data. For example: “people who live in Miami” or “people who like Intouch Solutions.”
  2. Multi-Dimensional Search: Graph Search utilizes layers of search refinement. Individual queries can be combined to look for multiple items in one search, such as: “people who are not my friends and like diabetes.” [Please note: While it’s unlikely that someone might “Like” a disease state, it’s important to remember that to maximize Graph Search effectiveness, it helps to think in terms of Facebook vernacular as opposed to traditional English linguistic patterns.]
  3. Personalization: Connections vs. keywords. Graph Search personalizes the social context to return lists of answers — not links — of the most query-relevant Facebook entities (people, places, photos and interests). Because Graph Search uses natural language queries, users can ask questions and, in return, receive socially relevant results not driven by keyword algorithms.

In the query example ”people who are not my friends and like diabetes,” the search phrase is comprised of three data sets:

  1. People who are not listed as a friend
  2. People who have indicated an interest in diabetes because they clicked the Like Button on a business or brand page(s) that pertain to diabetes
  3. People who have previously mentioned the term “diabetes” in a status update and/or comment on Facebook

Graph Search has streamlined the search for photos, too. Users can modify the search to look for photos of disease states instead of people, e.g.: “pictures of diabetes.”

What Does Graph Search Look Like?

Once Graph Search is applied to a user’s profile, it will appear as a larger search bar at the top of the page. As a user types in their text, the search bar will attempt to auto-fill the remaining content. The user can then narrow search results by selecting from a list of categories below the search box, and the search results will display on a unique page. Facebook will automatically save each search query and store it privately within the users Activity Log for future reference.

Example of the new Graph Search query process

Facebook Graph Search

Example of Graph Search results page

Part of the goal of Graph Search is to connect people with similar life experiences to each other through Facebook. Within the pharma space, this technology provides an opportunity to enhance online community interaction for users with similar or matching disease states within Facebook. Users can select from the search results or further refine search via the dropdown menus on the right side of the screen.

For example, a user could refine the results by “Current City.” This opens the door to build the user’s local network and possibly positively impact someone struggling with their disease state in their community. Search results could be refined even further by clicking the “Add” link next to the “Likes” filter to look for businesses and/or brand pages.

Privacy Settings Updates

Recent enhancements to the privacy controls panel impact how and what user content is available within the Graph Search results. Users will have to decide what Timeline content they wish to hide from Graph Search and then configure those settings through the enhanced privacy panel located next to the user’s mini-profile picture at the top right of the page.

Facebook stated that previously hidden Timeline comments will appear again in Graph Search results unless the user applies the latest privacy settings. Any content previously set to “private” will have to be reset to private again. Removing tags, descriptions and geographic data will help keep pictures from appearing in Graph Search results.

Graph Search will only display content that a user’s friends have already shared with each other or content that is set to “Public” status by default such as profile pictures, Likes, employers, locations and schools.


While Graph Search is still in beta, now is the time to begin thinking about how to maximize brand page content for inclusion in future Graph Search inquiries. The following practices can help drive Graph Search results in the future.

  • Continue to generate relevant posts that matter to the fan base of the brand page(s).
    • To help boost brand exposure within Graph Search, include words and phrases people would likely use in future inquiries.
      • For example, a user may search for: “Which drugs have my friends taken for a cold?”
  • Tag each picture with meaningful information.
    • Include location-based data to boost geographic association
    • Include a detailed description with contextually relevant search words
      • For example, if a user searches for “pictures of psoriasis,” Graph Search would search photo descriptions for that term.
  • Utilize other forms of paid placement like Sponsored Stories to promote brand messages and boost fan engagement.


Since Facebook has only begun to publicly introduce Graph Search, many more opportunities for this technology have yet to materialize. Invariably, features will change and evolve as part of the continued rollout and expansion of the enhanced functionality. We will continue to monitor future developments as they are released and provide updates as warranted.

Content Resources:
Facebook Marketing: Graph Search:

Facebook Technical: Graph Search (Beta):


* All fields are required.

By on

You may also like