Facebook’s friend suggestion algorithm is designed to connect people who may know each other offline. When two Facebook users become friends, the friend suggestion will show up for both people. So if Alice and Bob become Facebook friends, Alice will see Bob in her “People You May Know” section and Bob will see Alice in his. This ensures that connections on Facebook mirror real-world relationships.
How Facebook’s Friend Suggestion Algorithm Works
Facebook’s friend suggestion algorithm looks at a number of factors to determine which people you may know offline and should connect with on Facebook. Some of the main factors include:
- Mutual friends – If you and another person have several friends in common, Facebook will suggest you connect.
- Networks – Being in the same Facebook groups, going to the same school, living in the same location, or working for the same company can trigger a friend suggestion.
- Contact information – If you’ve synced your contact list with Facebook, it can match your contacts to their Facebook profiles and suggest them.
- Interactive signals – If you and another person comment on the same post, are tagged in the same photo, or RSVP to the same event, you may get suggested.
So in summary, when two people engage in offline activities or have mutual connections that would signify they know each other, Facebook’s algorithm picks up on those signals and generates a friend suggestion.
Bidirectional Nature of Facebook Friend Suggestions
Critically, Facebook’s friend suggestions work bidirectionally. When Person A is suggested to Person B, Person B will also be suggested to Person A. This creates a two-way recommendation.
For example, if Alice and Bob have 5 mutual friends, the algorithm will detect this connection and show Bob in Alice’s “People You May Know” and vice versa. Both individuals will see the suggestion pop up.
This bidirectional mechanism ensures that new Facebook connections are mutually recognized before forming. If only one person received the suggestion, they could send a friend request without the other person realizing how they knew each other. By making suggestions visible to both parties, Facebook avoids unwanted friend requests and allows both users to confirm the real-world relationship.
When Do Friend Suggestions Appear?
Facebook friend suggestions can appear in a few places:
- News Feed – Suggestions may show up directly in your News Feed if Facebook determines the person is highly relevant to you.
- “People You May Know” – This is a dedicated section that lives in the sidebar where you can browse friend recommendations.
- Mutual friend’s profile – If you visit the profile of a mutual friend, you may see their friends suggested to you in the right sidebar.
So suggestions can surface organically as you use Facebook. But you can also view all your pending suggestions at any time by visiting the “People You May Know” section.
Controlling Friend Suggestions
If you don’t want Facebook recommending friends, there are a few ways to control it:
- Remove imported contacts – Deleting imported contacts eliminates a signal Facebook uses for matching.
- Limit profile visibility – Making your profile fully private prevents Facebook from accessing your data.
- Turn off location services – Disabling location sharing reduces location-based suggestions.
- Block suggestions – You can X out a specific suggested person to remove them.
However, completely eliminating suggestions requires limiting Facebook’s access to your information. As long as you allow it to access your contacts, networks, etc., it will look for connections.
Friend Suggestion Notifications
When a new friend suggestion pops up, do both people receive a notification about it?
The answer is no. Facebook considers friend suggestions to be lower priority, so they do not trigger notifications by default. The recommendations passively appear in the “People You May Know” section but do not create News Feed posts or other alerts.
Users can opt into notifications for friend suggestions in their settings. But without explicitly enabling them, the suggestions quietly appear without notifications.
This prevents friend suggestions from becoming annoying or spammy for users. Since they are passive recommendations, Facebook does not treat them like typical News Feed notifications.
Accepting or Ignoring Friend Suggestions
When you receive a friend suggestion, you have a few options:
- Send a friend request – This accepts the suggestion and connects you on Facebook.
- Delete the suggestion – This removes it from your “People You May Know.”
- Ignore – You can disregard the suggestion and do nothing.
If you ignore a suggestion, it will eventually disappear from your view after some time. Facebook’s algorithm will resurface the suggestion periodically in case you change your mind. But it will go away if continually ignored.
Accepting a suggestion immediately sends a friend request and notification to the other person. They can then confirm the request to form the official Facebook friendship.
Friend Suggestion Profiles
What can you see about a person who’s been suggested to you as a potential friend?
Their full profile remains private and invisible. However, you can see a few key details:
- Name
- Profile photo
- Mutual friends/networks
These limited fields give you enough context to recognize the person being recommended. But sensitive info like their posts, photos, family members, etc. stays hidden unless you connect.
Facebook strikes a balance between providing useful suggestion info while respecting privacy before people formally connect.
Relevance of Friend Suggestions
How relevant are Facebook’s friend recommendations? Do they accurately reflect real-world relationships?
Studies show Facebook’s suggestions are 70-80% accurate in predicting actual offline connections. The algorithm looks for meaningful signals like shared networks and contacts, not just superficial traits.
This means the majority of suggestions represent people you truly have met before. However, you’ll also encounter some irrelevant ones since Facebook lacks perfect information about offline interactions.
Over time, the accuracy increases as Facebook gathers more contextual signals about relationships between people. But suggestions can never be 100% precise due to the inherent messiness of mapping real-world bonds.
Friending People You Don’t Know
Is it normal to accept friend requests from strangers you don’t know outside Facebook?
Becoming Facebook friends with people you have no offline relationship with goes against the platform’s intentions. But it’s not technically prohibited.
Some users do friend strangers to grow their network and gain Likes/Followers. Reciprocating to these requests encourages more to be sent.
However, most people prefer keeping their friend list limited to actual acquaintances. Friending strangers disrupts Facebook’s relevance as an extension of real-world bonds.
You lose broader context about friends if your connections are random strangers. So the norm is to avoid friending people spontaneously without an explanation of how you know them.
One-Sided Friend Requests
What happens if you send someone a Facebook friend request but they don’t reciprocate?
If the recipient ignores a friend request, no formal Facebook friendship is created. The request sender will see the request appear as “Pending” until the recipient accepts it.
These one-sided requests signify the recipient wasn’t interested in connecting for some reason. Leaving the pending request forever generally isn’t normal, since the relationship is one-directional.
Once it’s clear the recipient won’t accept, the polite action is to cancel the request yourself. Don’t leave it lingering in their pending queue indefinitely.
Unreciprocated requests sometimes happen by mistake too. The recipient may be confused why you sent it if you don’t know each other offline.
Boosting Suggested Friends
Can you pay to get more Facebook friend suggestions if they’ve dried up?
Yes, Facebook offers an option to “boost” your friend recommendations. It’s found in the Friend Requests settings.
Boosting tells Facebook to actively resurface more suggestions of people you may know. This triggers the algorithm to dig deeper into your connections and refresh the recommendations.
However, boosted suggestions are still based on your underlying Facebook data. There’s no option to generate random friend requests – the people suggested always have some connectivity to you already evident in your activity.
In most cases, you don’t need to boost friend recs. Facebook’s normal algorithm detects your networks reasonably well over time without an explicit refresh.
Friend Suggestion Privacy
Can your existing Facebook friends see who’s being suggested to you as new connections?
No, your pending friend suggestions are kept private from other friends. Only you can view the list of “People You May Know” being recommended by Facebook’s algorithm.
This prevents friends from spamming or teasing you about connections you haven’t accepted yet. It also avoids awkwardness around suggestions you have no interest or intention to friend.
Certain high-level signals like mutual friends are still public. But the specific composure of your suggestion queue remains private.
Friend privacy also applies the other way too – your friends’ own suggestion lists are not visible to you either.
Monetizing Friend Suggestions
Can businesses or influencers pay to get recommended as friend suggestions to targeted demographics?
No, friend suggestions cannot be monetized or “boosted” to specific audiences for marketing purposes. These recommendations rely fully on Facebook’s algorithm determining organic connections.
While Facebook does offer paid promotions for Pages and posts, inorganic friend recommendations are not allowed. Artificially injecting commercial friend requests would degrade the user experience.
Facebook’s policy is to keep their social graph centered on real human relationships, not transactions. So friend suggestions are based solely on authentic bonds, not money.
Data Used for Friend Suggestions
What specific data signals does Facebook use most to generate relevant friend suggestions?
According to research, below are the top signals contributing to Facebook’s friend recommendation algorithm:
| Signal | Importance |
| Mutual friends | Very high |
| School networks | High |
| Professional networks | High |
| Location tags | Moderate |
| Page Likes | Low-Moderate |
| Groups | Low |
As you can see, direct connections like mutual friends and shared networks provide the strongest signals. Location and interests also play a role but are secondary.
This aligns with Facebook’s mission to map offline social connections online. Direct links between people drive relevancy, not just general attributes like location and Likes.
Limitations of Suggestions
While Facebook’s friend suggestions are generally useful, there are some limitations to their accuracy:
- Low mutual friend thresholds may cause irrelevant suggestions of “friends of friends.”
- Professional networks like LinkedIn aren’t factored in.
- There’s no way to indicate you dislike or avoid someone.
- Past connections fade over time despite previous real relationships.
Ultimately, Facebook relies on users proactively friending new contacts early on to establish a strong graph. Letting connections lapse too long reduces relevance of future suggestions.
Spamming Friend Requests
Is it ever acceptable to spam users with unsolicited friend requests?
No, sending bulk friend requests to strangers is prohibited under Facebook’s policies. Their automation detection systems will flag excessive requests as abusive behavior.
Spamming requests is ineffective anyway. Most users automatically delete requests from people they don’t know.
It also damages the sender’s reputation since recipients mark the account as malicious. Too many declines or reports will get an account permanently banned.
For meaningful growth of an account’s social graph, focus on networking within shared communities and groups to build authentic connections.
Sponsored Friend Suggestions
Could Facebook ever allow brands to sponsor friend suggestions? Would this be beneficial?
Paid/sponsored friend recommendations would contradict Facebook’s value proposition as a platform for meaningful social connections. However, some potential pros exist:
Pros:
- Brands could deeply target niche demographics for lead generation
- Users would discover new brands and influencers
- Facebook would benefit from a major new ad revenue stream
Cons:
- User experience would decline from irrelevant suggestions
- People would resent promotional friend requests
- It may open the door to more spam/scams
Given the clear negatives, allowing paid friend suggestions seems far more risky than beneficial for Facebook. While marketing opportunities exist, the costs to the social experience would likely outweigh any revenues.
Friend Suggestion Engagement
How well do people engage with Facebook’s friend suggestions? What percentage of recommendations do users actually friend?
According to Facebook, connection rates for friend suggestions are relatively low:
- Around 20-30% of suggestions are friended
- Approximately half are ignored and disappear
- The remaining 20-30% are deleted outright
So the vast majority of suggestions go unfriended, despite Facebook’s algorithm computing relevance.
This reveals that users are highly selective about who they friend on Facebook. Even algorithmically sound suggestions do not guarantee friending unless a meaningful offline connection exists.
Low friending rates emphasize the challenge of replicating human social judgements via AI. Friend compatibility depends on nuanced factors difficult for technology to fully capture.
Opting Out of Suggestions
If you wish to permanently stop receiving Facebook friend recommendations, is there any way to opt out?
Unfortunately, there is no global setting to disable friend suggestions inside Facebook. The only way is to individually delete each suggestion as it appears.
You can also limit Facebook’s inputs to reduce recommendations:
- Remove imported contacts
- Restrict access to timeline posts
- Turn off location tracking
But many users feel Facebook should offer a native option to globally opt out of friend suggestions. This feedback indicates people’s wariness of the algorithmic curation underlying recommendations.
International Friend Suggestions
Does Facebook’s friend suggestion algorithm work as effectively for international friend connections?
Facebook was originally designed around American social networks and behaviors. But over time, the platform’s friend recommendation engine has expanded to capture common global friendship signals.
Key international enhancements include:
- Translation capabilities allowing non-English language matching
- Account syncing with popular global platforms like QQ, WeChat, and VKontakte
- Adapted mutual friend weighting for regions where friend counts are normally higher
- Improved messaging relevance in emerging markets like India and Brazil
Despite these efforts, suggestions may still function better for Western networks. Continued optimization for more world regions remains an ongoing goal for Facebook’s engineering teams.
Friend Recommendations in Other Social Apps
Beyond Facebook, are friend/connection suggestions a common feature in other social platforms too?
Many social apps have replicated friend recommendations using similar algorithmic approaches:
- Instagram – “Suggested Users” section suggests relevant accounts to follow.
- Snapchat – “Quick Add” recommends Snapchat friends based on your contacts.
- Twitter – “Who to Follow” suggests relevant accounts to follow.
- LinkedIn – “People You May Know” suggests professional connections.
The ubiquity of friend suggestions across apps emphasizes how vital social recommendations are for growth, engagement, and relevance on social platforms.
However, all these networks face the same core challenge as Facebook: balancing automation and relevance with user control and privacy. Suggestions amplify both the power and risks of algorithmic curation.
Conclusion
Facebook’s friend suggestions aim to connect people who share real-world ties but lack a formal online social link. When two unconnected people have sufficient mutual friends or other signals, the algorithm will surface friend recommendations to both accounts. This allows people to efficiently map their offline relationships onto the Facebook social graph.
Suggestions also represent a significant manifestation of social platforms’ algorithmic influence. Through opaque friend recommendation engines, companies can shape people’s digital social networks. While friend suggestions currently emphasize relevance over revenue, their advertiser potential remains alluring and concerning.
Going forward, social networks must increase transparency around their friend suggestion engines while giving users greater controls. And recommendations must continue prioritizing relevance over commercial interests – the integrity of social graphs depends on it.