Some jobs are filled without a public posting, but "80% of jobs are never advertised" has no modern source. The figure traces to a 1966 article citing a pilot survey of 27 employers in Rochester, New York, and that survey counted newspaper ads. What is real today is narrower: referrals to jobs that are posted, postings that are up only briefly, and small employers that never post.
This guide is published by JobShifu, which crawls employer career pages directly and sells a resume tool. Outside sources are linked and dated in the Sources section, and JobShifu's own numbers are labelled as its own measurement.
Where does the 80% hidden job market statistic come from?
From career-advice books, not from a labour survey of the present day. Jesse Preston's 2023 trace of the claim, which works from the original publications, finds this chain [1]:
- 1960. Bernard Haldane's *How to Make a Habit of Success* splits jobs paying $600 a month into 80% filled by personal recommendation, 15% through newspaper ads and 5% through agencies. That is a claim about how jobs were filled, not whether they were advertised.
- August 1966. Haldane writes that 80 percent of existing job openings are hidden, citing a pilot by the National Industrial Conference Board, funded by the Ford Foundation, that surveyed 27 companies in Rochester, New York.
- What the pilot measured. Whether a job the companies filled had appeared in a Rochester newspaper ad. 25.1% had. Service jobs, the largest group, were mostly advertised. And the report notes that many of the larger employers belonged to a local council that asked its members not to advertise in the papers.
- 1970s onward. Richard Bolles's *What Color Is Your Parachute?* and other bestsellers repeated the "hidden market" as general advice, and the number travelled without its source.
So the 80% described one city's newspaper ads in the 1960s, in a sample where many employers had agreed not to advertise. It says nothing about career sites or online job boards, which did not exist.
A separate and much better known study is often cited alongside it. Mark Granovetter's *Getting a Job* (1974) asked how 282 professional, technical and managerial men in the United States found their jobs, and found that personal contacts mattered a great deal [2]. That measured how people heard about jobs, not whether the jobs were advertised. Both can be true at once: a role can be posted and still be filled by someone who heard about it from a friend.
Is the hidden job market real today?
Partly. Three things are real, and none of them is "80% of jobs are invisible".
Referrals to posted jobs. A posting is often the official record of a role that someone inside already knows a candidate for. The job is advertised; the referred applicant simply gets read first.
Jobs that are posted briefly. This is the part of the "hidden" market that a weekly search misses. JobShifu rechecks employers' own career pages daily, and between 8% and 17% of postings are gone within a week of first appearing [3]. A posting that is up for five days is advertised, but to someone who checks once a week it might as well not exist.
Employers with no careers page at all. Small businesses that hire by word of mouth or a sign in the window genuinely never post. No crawler sees them, including JobShifu's, which is also why no dataset can put a number on the unposted share. Anyone quoting a precise percentage for jobs that are never advertised is quoting something nobody can measure.
What can be measured is how much is posted. On 18 October 2026 JobShifu was watching 1,258,928 open roles at 15,074 employers, read from the employers' own applicant-tracking systems rather than from job boards [4].
How to find jobs in the hidden job market
- Search employers' career pages, not only job boards. The employer's own board carries the original listing, and it is the copy that comes down when the role closes, while a job board can keep showing it. See postings as employers publish them.
- Check new postings every day or two. With up to 1 in 6 postings gone inside a week, cadence matters more than reach. A short list of employers checked often beats a long list checked weekly.
- Ask for a referral to a specific posting. Find the posting first, then ask someone at the company to pass your name to the hiring team for that requisition. A referral attached to a real opening is a concrete request; "let me know if anything comes up" is not.
- Contact small employers directly. The one part of the market that is truly unposted is employers without an applicant-tracking system. For them, a short, specific note to the owner or manager is the application.
- Confirm a posting is live before you spend an evening on it. Paste its URL into the free checker to see when it went up, whether it is still listed and whether it was posted before.
Common questions
Is it true that 80% of jobs are never advertised?
There is no modern evidence for it. The figure traces to a 1966 article by Bernard Haldane citing a pilot survey of 27 companies in Rochester, New York, which measured how many filled jobs had appeared in newspaper ads, in a city where many large employers had been asked not to advertise. No current dataset measures the share of jobs that are never posted, because jobs that are never posted cannot be counted from postings.
What is the hidden job market?
The phrase covers jobs filled without a public ad: roles filled by referral or internal move, and roles at employers that never post. In practice the largest part a job seeker can do anything about is jobs that are posted only briefly. Between 8% and 17% of postings leave the employer's own board within a week of first appearing, per JobShifu's daily recheck of employer career pages.
How do you access the hidden job market?
Watch employers' own career pages rather than only job boards, check new postings every day or two, ask for referrals to specific postings rather than in general, and contact small employers without a careers page directly. Each targets a part of the market that a weekly job-board search misses.
Is networking better than applying online?
They are not alternatives. Granovetter's 1974 study of 282 professional, technical and managerial men found personal contacts played a large part in how they found their jobs, but a contact usually points you to a role that exists, and that role is usually posted. The strongest version of both is a referral attached to a specific posting.
How this was measured
- The 80% trace relies on Jesse Preston's 2023 reading of the 1960 and 1966 Haldane texts and the 1966 National Industrial Conference Board pilot report. JobShifu has not read those primary documents itself; the citation is to Preston's account.
- Postings gone within a week comes from JobShifu's survival estimate on employers' own career pages, rechecked daily, for postings first seen 1 July to 15 August 2026: 83% to 92% were still listed at day 7, so 8% to 17% were gone. The range spans two groups of sources; method and caveats are on the lifespan report.
- Open roles watched is from JobShifu's relisting report, edition of 18 October 2026: roles first listed on or after 8 August 2026 on employer applicant-tracking systems.
- What this cannot measure. A crawler sees only what is posted. The share of jobs that are never posted is outside any posting dataset, this one included.
Sources
- Jesse Preston, "Hidden Origins of the Hidden Job Market Pt. 2", Learning Curves: learningcurves.org/hidden-origins-of-the-hidden-job-market-pt-2. Published 1 July 2023, read 23 October 2026.
- Mark Granovetter, *Getting a Job: A Study of Contacts and Careers*, 2nd edition, University of Chicago Press: press.uchicago.edu. First published 1974; read 23 October 2026.
- JobShifu, "How long do job postings stay open?": /research/how-long-jobs-stay-open, measured to 15 September 2026, CC BY 4.0.
- JobShifu, "How often are job postings relisted?": /research/relisting, edition of 18 October 2026, CC BY 4.0.