The Job You Have Not Heard About Yet

A great deal of job search energy goes into perfecting a resume for a stack of postings on a job board, and comparatively little goes into the channel that, by a wide margin, actually produces most hires: the people already in a person's own extended network, many of whom the job seeker has not spoken to in years.

Sociologist Mark Granovetter first documented this systematically in the 1970s, in a survey of two hundred and eighty two men in the United States about how they had actually found their current jobs. Most had gotten their position through a personal contact rather than a formal application, and the striking part of the finding was which contacts mattered most. It was not close friends and family, the people seen often. It was casual acquaintances, seen rarely, who turned out to be disproportionately responsible for the leads that actually led somewhere.

Granovetter called this the strength of weak ties, and the mechanism behind it is fairly intuitive once it is explained. Close friends tend to travel in the same circles a person already moves in, which means they mostly already know what that person knows. A casual acquaintance, someone from a previous job, an old classmate, a neighbor from several years back, sits in an entirely different information pool, with access to openings and leads that never would have reached the immediate circle at all.

This was not just an old finding from a pre internet era. A massive study published in Science in 2022, run by researchers from MIT, Stanford, and Harvard using five years of controlled experiments on LinkedIn involving twenty million users, tested this causally by adjusting which kinds of connections the platform's algorithm recommended to different groups. The result confirmed Granovetter's original theory directly: weaker social connections produced significantly more job mobility than stronger ones, holding up half a century after the original study and across an entirely different technological landscape.

This was not just an old finding from a pre internet era.

The 2022 study added an important refinement to the original theory as well. The benefit of a weak tie was not perfectly linear, growing stronger the weaker and weaker a connection became without limit. The most valuable connections tended to be moderately weak, someone with a modest handful of mutual connections rather than either a close friend or a total stranger, since a connection that weak still needs some shared context to translate into an actual introduction or lead.

Reaching out to a loose acquaintance from three jobs ago can feel considerably more awkward than applying to a posted opening, since it requires initiating contact rather than passively submitting a form, and for a lot of men, that kind of direct ask runs against an instinct to handle problems independently rather than lean on other people. But this framing misreads what the interaction actually is. Reaching out to someone you have not spoken to in years is not a request for a favor. It is an exchange of information, catching up, mentioning what you are looking for, and asking what they have heard, which is a normal, low stakes conversation rather than the awkward plea it can feel like in anticipation.

In practice, this means deliberately working through a list of dormant contacts rather than waiting for the right moment to reach out to feel natural, since it rarely does. Old coworkers, former clients, people met once at a conference, alumni from the same program: each is a low pressure message away from potentially surfacing an opening that was never going to appear on a job board at all, because plenty of positions get filled through exactly this kind of informal referral before they are ever posted publicly.

This also means a job search benefits from casting a wider net of contacts rather than a deeper one aimed at a small handful of close connections. A list of fifty loose acquaintances, each contacted briefly, is likely to surface more genuine leads than repeated, more intensive requests aimed at the same five people a person already knows best.

The resume and the formal application still matter, and neither Granovetter's research nor the studies that followed it suggest abandoning them. But treating networking as a peripheral, slightly uncomfortable extra, rather than the primary channel the data consistently shows it actually is, means overinvesting in the smaller lever and underinvesting in the one most likely to actually work.