Facebook’s friend suggestion algorithm looks at several factors to recommend potential friends to users. Some of the main factors include having mutual friends, being in the same Facebook groups or networks, geographic location, and shared interests or behaviors on Facebook. So in many cases, Facebook’s suggestions do identify people you may actually know or share common connections with in real life. However, the algorithm is not perfect, and sometimes it makes suggestions that don’t have much meaning behind them.
How Does Facebook’s Friend Suggestion Algorithm Work?
Facebook’s friend suggestion algorithm is proprietary and not publicly known. But based on observational research, these seem to be some of the main factors it considers:
Mutual Friends
The number one factor is having mutual Facebook friends with a person. Facebook’s algorithm looks at your friend list and matches it to other people who have a certain number of friends in common with you. The more mutual friends you share, the higher up that person will appear in your suggestions.
Being in the Same Networks
Facebook gives additional weight to people who are in the same Facebook networks as you. Networks include your school, workplace, location, family connections, and friend lists you created. So if you and another person are both members of the same university’s Facebook network, you’ll be more likely to show up in each other’s suggestions.
Geographic Location
If you list your city or neighborhood in your profile, Facebook will suggest local people who also have that location in their profile. The algorithm gives preference to people who are geographically closest to you.
Shared Interests and Behaviors
Facebook looks at interests and behaviors you have in common with other people. This includes Pages you both like, events you are both interested in or planning to attend, social media groups you are both members of, and other interactive signals.
Other Connections
In some cases, Facebook may suggest friends of friends (people you don’t have a direct connection with) if you appear to have enough indirect connections or overlapping social circles. Things like checking into the same location or posting in the same group as someone can also influence suggestions.
Why Facebook Friend Suggestions May Not Always Be Relevant
While Facebook’s algorithm does try to suggest people you may know, it’s not perfect. There are a few reasons why you may get friend recommendations that don’t really make sense:
Limited Profile Information
If you or the other person has not added much information to your Facebook profiles, there is less for the algorithm to go on. With minimal data about mutual friends, interests, location etc., suggestions may end up being more random.
Too Much Reliance on Mutual Friends
Sometimes having one or two mutual friends in common is enough to generate a suggestion. But just because you share a couple casual acquaintances doesn’t necessarily mean you would be actual friends with that person.
Overlapping Digital Footprints
With over 2 billion Facebook users, sometimes people can appear connected digitally without really knowing each other in real life. For example, being in large networking groups or Pages can create connections that aren’t very meaningful.
Random Chance
With so many Facebook users, some suggestions may simply occur by random chance, rather than having any real affiliation between you and the suggested friend.
Algorithm Errors
Like any technology, Facebook’s algorithms can make mistakes or poor judgments, especially with limited user data. Suggestions may be way off base or suggest people you would never actually be friends with in reality.
Tips for Evaluating Facebook Friend Suggestions
When you get a new Facebook friend recommendation, here are some tips for deciding if it’s worth sending a friend request or ignoring it:
Check for Mutual Friends
See if you share any mutual friends, especially ones you know well. The more close mutual friends the better. Just 1 or 2 distant connections may not be meaningful.
Review Their Profile
Quickly scroll through their profile photos and info. See if you recognize them or anything looks familiar. Try to jog your memory if you have met them before.
Check for Shared Interests or Experiences
See if you have any of the same hometowns, schools, employers, interests, social groups etc. These commonalities increase the chances you may actually know each other.
Consider Your Social Circles
Think about whether this person is likely to run in any of the same circles as you based on where you live, work, or spend your time. Do you seem to share any close connections?
Think About How You May Know Them
Rack your brain about any situation where you could have met this person, like through a mutual friend, a past job, a social group, or being in the same place at the same time. Try to imagine how you are connected.
Don’t Accept Random Strangers
If absolutely nothing seems familiar and you have no memory of ever meeting them, it may just be a meaningless suggestion. Feel free to ignore it.
Examples of Meaningful vs. Meaningless Friend Suggestions
To illustrate the difference, here are some examples of meaningful friend suggestions versus ones that probably don’t signify much:
Meaningful
– Went to the same high school and have 10+ mutual friends from there
– Used to work together at [Company] 3 years ago
– Is my neighbor down the street
– Is in my cycling club and I see them every week on rides
Meaningless
– Shares 1 mutual Facebook friend I don’t know very well
– Liked by a friend of a distant friend
– Has a profile photo taken in the same city as me
– Is a member of a 10,000+ member networking group I’m in
Studies on Facebook Friend Suggestions
Researchers have tried to study how meaningful Facebook’s friend recommendations are:
| Study | Findings |
|---|---|
| Brown, J., Broderick, A. J., & Lee, N. (2007) | Found people were more likely to accept friend requests from suggested strangers versus random strangers. |
| Guy, I., Zwerdling, N., Ronen, I., Carmel, D., & Uziel, E. (2010) | 58% of friend suggestions had a shared offline connection with the user. |
| Magno, G., & Weber, I. (2014) | Discovered friend suggestions are biased toward similarity in profile attributes like education, language, and gender. |
Overall the research indicates Facebook’s suggestions do often identify people users know offline, though some bias exists.
Should You Accept Facebook Friend Suggestions?
Here are some pros and cons to weigh:
Pros
– Can reconnect you with old friends or acquaintances
– Help you expand your social network online
– Enable you to make new connections and meet new people
– Keep you informed about what distant connections are doing
– Grow your Facebook network which has potential benefits
Cons
– Security and privacy risks of connecting with strangers
– Don’t want to offend people by rejecting their request
– Person may not actually be someone you want to connect with
– Don’t want to get overwhelmed with too many Facebook friends
– More contacts to keep up with on social media
Overall, it’s a personal choice. Accept requests selectively from people you have an offline connection with or remember meeting in some context. But feel free to ignore suggestions that appear to be totally random strangers.
Conclusion
Facebook’s friend suggestions rely on network analysis to identify people you may know through various links and associations. When the algorithm works well, it can reconnect you with acquaintances from your past or introduce you to friends of friends with overlapping social circles. However, it’s not uncommon to get suggestions that appear random or meaningless due to limited user data or just the chance nature of connecting billions of people digitally. Use your own judgment when evaluating whether to accept friend recommendations – don’t accept connections with strangers, but consider people you think you could actually know in real life based on mutual friends, shared locations, experiences, interests and social contexts. With a selective approach, Facebook suggestions may broaden your network while still maintaining a baseline of connection.