Do Weak Ties Find You Your Next Job
Granovetter's claim that acquaintances carry job news has become standard networking advice. A large LinkedIn experiment supports it, with a peak at moderately weak ties, and a Facebook study points the other way on a single close tie.
Answer to the question
Do weak ties find you your next job?
Often, with the best causal evidence from a 2022 LinkedIn experiment on digital industries. Moderately weak ties, with around ten mutual connections, raised job transmission most. A Facebook study found one stronger tie worth more at the margin, and in less digital industries stronger ties did more.
Key takeaways
- Granovetter (1973): of 54 people who found a job through a contact, 16.7 per cent saw the contact often, 55.6 per cent occasionally and 27.8 per cent rarely.
- Gee, Jones and Burke (2017), about 6 million Facebook users: most people were helped through one of their many weak ties, but a single stronger tie was significantly more valuable at the margin.
- Rajkumar and colleagues (2022), over 20 million LinkedIn members in randomised experiments: job transmission peaked at around ten mutual connections.
- In the same LinkedIn data, the weakest ties by messaging raised job transmission most, reversing the correlational pattern in which stronger ties looked better.
- Weak ties produced more job applications in digital sectors and strong ties in less digital ones (Rajkumar and colleagues, 2022).
Mark Granovetter argued that every bridge is a weak tie, and a long line of networking advice has grown from that claim. For finding out about jobs, the weak ties idea has held up better than most, with causal support from a large LinkedIn experiment, though not as neatly as the advice implies. The claim has led two lives, as an argument about information flow in networks and as advice about whom to call for a job. Granovetter's sample was small, and later researchers have tested the argument at a scale he could not. The result depends on how tie strength is measured, and on whether the job sits in a digital industry or an older one.
What Did Granovetter Claim, and What Did Gladwell Add?
Granovetter's paper, The Strength of Weak Ties, appeared in the American Journal of Sociology. He described the strength of a tie as a combination of the time people spend together, the emotional intensity, the intimacy and the services they exchange. A bridge, in his usage, is a tie that provides the only path between two points in a network. His claim was that every bridge of that kind is a weak tie.
The reasoning is about information. People who are close tend to know each other's friends and hear the same news, so a close friend mostly repeats what you already know. An acquaintance moves in other circles and can carry in something new.
A link that bridges two separate circles is always a weak one.
Granovetter's job data fits that picture. Among the people in his small sample who found work through a contact, about a sixth saw the contact often, a little over half saw the contact occasionally and a little over a quarter rarely. Most of the people who passed on a job lead were not close.
He also granted the obvious objection. People with whom you have strong ties are more motivated to help you, and he said so in the paper. His answer was that structure counts for more than motivation, because a motivated friend with no new information has little to offer.
Malcolm Gladwell's The Tipping Point uses weak ties for a different job. The book describes Connectors, people with an unusual gift for linking worlds, who know many people across many circles and treat acquaintances as worth keeping. Roger Horchow keeps a roster of sixteen hundred names, with a note on how he met each one. Lois Weisberg belonged to eight or ten separate worlds. Gladwell presents Granovetter's job-finding result as a general rule that such people exploit.
Granovetter's claim concerns the information reaching one job seeker. Gladwell's concerns the role of a particular kind of person in spreading ideas through a population. The book states the idea more firmly than the later record supports, and we did not recheck the percentages it prints from the original study. The Connector examples are anecdotes, and the book's surname test for spotting a Connector is Gladwell's own informal exercise, not a validated measure. For what the idea says about jobs, the later studies are the better guide.
A Facebook Study Against the Prediction
Laura Gee, Jason Jones and Moira Burke tested the idea on about six million Facebook users and published the result in the Journal of Labor Economics. Job help is hard to observe directly, so they used a proxy: people who eventually worked with a pre-existing Facebook friend. They linked that outcome to objective measures of how strong each tie was.
Most people, they found, were helped through one of their many weak ties, which fits Granovetter's count. Per tie, though, the picture changed. A single stronger tie was significantly more valuable at the margin. The authors describe this as evidence opposite to Granovetter's prediction.
The design shapes what that means. Gee and colleagues observed what people did without randomising who met whom, and working later with a friend is a stand-in for hearing about a job. Read that way, the two findings are less opposed than they sound: weak ties win on numbers because everyone has far more of them, while a close friend is worth more per contact.
Randomising Twenty Million LinkedIn Members
The experiment came in 2022. Karthik Rajkumar, Guillaume Saint-Jacques, Iavor Bojinov, Erik Brynjolfsson and Sinan Aral published it in Science. They ran randomised experiments on LinkedIn's People You May Know suggestions between 2015 and 2019, covering over 20 million members. In the 2019 wave alone, over 16 million members were involved and the experiment recorded over 70 million job applications. That wave tracked applications, not completed hires, and LinkedIn is a professional network that may not behave like friendships elsewhere. The paper discloses that two authors are or were LinkedIn employees.
Because LinkedIn varied which ties it suggested, the authors could see what the new ties did. Moderately weak ties worked best. Adding ties with more mutual connections raised the chance of a job transmission up to a point, and the peak fell at around ten mutual connections. Beyond that, ties with more mutual connections did less. Measured by how much two people messaged each other instead, the weakest ties raised job transmission the most and the strongest raised it the least.
That second result reverses the correlational pattern in the same data, where stronger ties were associated with more transmission. It shows how observational work such as the Facebook study can mislead on this question: people who already talk a lot may also be the people who tell each other about jobs.
Industry changed the answer. Weak ties produced more applications in digital sectors, which the authors defined by IT intensity, software use, suitability for machine learning and remote work. Strong ties produced more applications in less digital, less automated sectors.
Three Studies, Three Yardsticks
The three studies use three yardsticks for the same word.
| Study | Data | How strength was measured | What it found |
|---|---|---|---|
| Granovetter (1973) | 54 people who found a job through a contact | How often they saw the contact | Most contacts were seen occasionally or rarely |
| Gee, Jones and Burke (2017) | About 6 million Facebook users, observational | Objective measures of tie strength | Weak ties help most people; one stronger tie is worth more |
| Rajkumar and colleagues (2022) | Over 20 million LinkedIn members, randomised | Mutual connections; messaging | Moderately weak ties peak; weakest by messaging helped most; varies by industry |
Frequency of contact, number of shared friends and amount of messaging are related, and they are not the same quantity. Two people with ten mutual connections may talk daily or once a year. A result stated for one yardstick need not hold for another, and the LinkedIn paper itself reports that the measure matters.
Three things survive. Acquaintances do reach information that a close circle lacks, which was Granovetter's mechanism, and the randomised LinkedIn evidence supports it for digital industries. Where work is less digital, stronger ties did more. And the advice to cultivate more acquaintances is easier to defend than the advice to avoid close friends, since Gee and colleagues found a single strong tie worth more at the margin.
Neighbouring questions sit in other articles. How Many Friends You Can Actually Keep covers how many close ties a person can sustain, The Office That Stopped People Talking looks at ties inside a workplace, and loneliness is a different question, which Which Is Worse, Being Alone or Feeling Lonely takes up.
Gladwell's Connector is a person with a large and varied roster of acquaintances. The strongest experimental result concerns a single tie, one with around ten mutual connections, and it says nothing about people who hold the largest address books.
Common questions
Do weak ties find you your next job?
Often, and with the best causal evidence in digital industries. A 2022 randomised study on LinkedIn found moderately weak ties, with around ten mutual connections, raised job transmission the most. In less digital industries, stronger ties did more.
Did Granovetter prove weak ties beat strong ties?
No. His 1973 paper argued the mechanism and used a small sample of 54 job changers. Later work supports it in some settings and, on Facebook, found a single stronger tie worth more at the margin.
Is a Connector the same as a weak tie?
No. A weak tie is a relationship, measured in time, intensity or shared contacts. Gladwell's Connector is a person with many varied acquaintances. The experimental evidence is about ties, and says little about the people who hold the most of them.
Sources
Gladwell, M. (2000). The Tipping Point: How Little Things Can Make a Big Difference. Boston: Little, Brown. Seed book; vault hub Books/The Tipping Point.
Granovetter, M. S. (1973). “The Strength of Weak Ties.” American Journal of Sociology, 78(6), 1360-1380. doi:10.1086/225469 https://doi.org/10.1086/225469
Gee, L. K., Jones, J., & Burke, M. (2017). “Social Networks and Labor Markets: How Strong Ties Relate to Job Finding on Facebook's Social Network.” Journal of Labor Economics, 35(2), 485-518. doi:10.1086/686225 https://doi.org/10.1086/686225
Rajkumar, K., Saint-Jacques, G., Bojinov, I., Brynjolfsson, E., & Aral, S. (2022). “A causal test of the strength of weak ties.” Science, 377(6612), 1304-1310. doi:10.1126/science.abl4476 https://doi.org/10.1126/science.abl4476