Careers

How to Spot a Ghost Job Posting Before It Eats Your Week

By Ubaid Rehman · Updated 17 September 2026

A ghost job costs you forty minutes and gives you nothing back. Not a rejection, not a recruiter screen — nothing. Here is what the research actually says about how common they are, the seven tells that hold up under scrutiny, and a triage you can run on any listing in about ninety seconds.

The worst part of a ghost job is not the wasted time. It is what the silence does to your read of the market. Apply to thirty roles, hear back from two, and the obvious conclusion is that you are the problem — your resume, your experience, your years. Sometimes that is true. Often a meaningful slice of those thirty were never going to answer anyone, because nobody at the company was assigned to fill them.

That distinction matters more than any resume tweak. So let us get precise about it.

What a "ghost job" actually is

The term gets used for four different things, and conflating them is why the statistics look so wild. Worth separating:

• The evergreen requisition. A permanently open listing — "Software Engineer, All Levels" — kept live to build a pipeline. There is no specific seat. There may be one in March.
• The frozen req. A real role whose budget got pulled after the posting went up. Nobody took the listing down, because taking listings down is nobody's job.
• The internal-candidate formality. The role is filled. Someone in HR is posting it publicly to satisfy a policy or a compliance requirement. Your application is filed, never read.
• The pipeline farm. A listing posted with no intent to hire at all, to harvest resumes, gauge the talent market, or signal growth to customers and investors.

Only the last one is cynical. The first three are mostly entropy — and entropy is the more common cause by a wide margin. That is oddly good news, because entropy leaves fingerprints you can read.

How common are they, honestly?

You will see "1 in 3 jobs are fake" headlines. You will also see 1 in 7. Both are cited. Both come from real analyses. They disagree because they are measuring different things.

A Clarify Capital analysis of more than 175,000 US listings, reported in April 2026, put the figure at roughly one in seven — about 14%. That is a posting-level count with a fairly strict definition.

Analyses that define a ghost job more loosely — including evergreen reqs and stale listings — land between 18% and 33%. A September 2025 ResumeUp.AI analysis of US LinkedIn listings came in at 27.4%. A widely quoted "1 in 3" figure sits at the top of that range.

Two more numbers are worth holding onto. In a Clarify Capital survey from January 2025, 81% of recruiters acknowledged posting roles they were not actively trying to fill. And a Resume Genius survey of 1,000 active US job seekers found 67% believed they had hit a fake, misleading, or unfilled posting more than once. The perception gap between job seekers and the market is not paranoia.

One caveat that almost nobody prints: the Bureau of Labor Statistics does not track this. There is no official number. Every figure above is a private estimate with a private definition, and you should treat the precise decimal points accordingly. The range is the finding. Somewhere between one in seven and one in three listings will not produce a hire, and the rate is worse in some sectors than others — a Columbia Law Review analysis of JOLTS data found government, education and health, and information/tech carrying the heaviest concentrations.

The seven tells, ranked by how much they actually mean

Not all red flags carry equal weight. In rough order of signal strength:

1. Age on the board. The strongest single indicator. A posting live past 30 days with no edits is suspicious; past 45 days it is strongly associated with ghost behaviour. Real hiring processes move or die. They rarely sit.

2. The repost cycle. LinkedIn's "Reposted" badge is the tell most people scroll past. A role reposted every two weeks for four months is a pipeline farm, not a search. Check the badge before you check the job description.

3. Missing pay where pay is required. In jurisdictions with pay-transparency laws — Colorado, California, New York, Washington, Illinois and others — a compliant employer posts a range. A missing or absurdly wide range ("$70,000–$260,000") on a role that should be covered suggests the listing was not reviewed by anyone accountable for it.

4. No named human. No recruiter on the post, nobody from the team sharing it, no hiring manager anywhere. Real searches generate noise: people want their req filled and they tell their network.

5. Requirements that describe two jobs. A "generalist" listing that wants senior backend, data engineering, and stakeholder management is often a placeholder written from a template, or a wish list assembled by committee with no seat behind it.

6. Timing against layoffs. A posting that appeared within weeks of a publicised layoff at the same company is worth a hard look. Sometimes it is genuine backfill in another division. Sometimes it is optics.

7. Board-only presence. If the role is on Indeed and LinkedIn but not on the company's own careers page, the listing may be an aggregator scrape of something already closed — or bait from a third party.

Any one of these on its own proves nothing. Three stacked together is a strong signal. That is the whole logic of triage: you are not looking for proof, you are looking for a score.

The 90-second triage

Before you open a document, run this. It costs a minute and a half and saves hours over a month of applying.

• Find the true post date. Boards display "2 days ago" for reposts. Open the company's own careers page and find the same req — the date there is the honest one.
• Look for a req ID. A job with an ATS requisition number (Greenhouse, Lever, Workday) attached to a live pipeline is more likely real than a free-text listing.
• Search the title on LinkedIn. Filter to that company. If the same title appears three times in six months with no new hires visible in the team, you have your answer.
• Check the team page. Did someone just leave the seat? A departure that matches the listing is the single best confirmation a role is real.
• Score it. Green: apply properly, tailor everything. Amber: apply fast with a light touch, and move on. Red: skip, or spend your energy on a direct message to someone on the team instead.

The amber category is where most people go wrong. They either apply with full effort to everything or give up entirely. A two-minute amber application is a rational bet; a two-hour one is not.

Let the filter run the triage for you

Running that checklist by hand across forty listings is a chore, and chores get skipped on the days you most need them. The Ghost Job Filter does the pass for you: paste a listing or its URL and it scores the posting against the signals above — posting age and repost pattern, salary disclosure against the relevant jurisdiction, requirement coherence, named-contact presence, careers-page mirroring — and hands back a green / amber / red call with the specific reasons behind it.

What makes the output useful is not the verdict, it is the reasoning. "Red — live 61 days, reposted three times, no range on a Colorado-based role" tells you something you can act on. It also tells you when the tool is wrong, which matters, because it sometimes is.

Where a listing comes back green, that is your cue to spend real effort: a tailored, machine-readable resume from the Resume Rewriter and a cover letter written to that specific req. Effort is a finite resource. Triage is how you aim it.

What this cannot tell you

Honesty about limits, because the alternative is a tool you stop trusting:

No filter can see inside an ATS. A posting can look immaculate — fresh, salaried, named recruiter — and still be an internal-candidate formality. Nothing visible from the outside distinguishes that case, and anyone claiming otherwise is selling something.

Age is also a noisy signal in specific contexts. Senior and highly specialised roles legitimately sit open for months. Government and academic postings run on statutory timelines that make a 60-day listing completely normal. A filter tuned for tech hiring will over-flag a tenure-track posting, and you should override it when you know the sector.

And a green score is not a prediction that you will hear back. It only means the role appears to exist. Those are different claims, and conflating them is how people end up angry at the wrong thing.

What to do with the time you get back

The point of filtering is redirection, not subtraction. The hours you reclaim are worth more spent on the channels ghost jobs cannot contaminate.

Referrals bypass the problem entirely — a person who confirms the seat exists is better evidence than any signal on this page. A LinkedIn profile built to be found flips the direction of the search, and if yours has not been touched since 2023, the 2026 optimisation guide is the place to start. If the filtering is turning up a market that has genuinely moved past your current stack, the Skill Coach maps the gap, and the Salary Replacement planner is for the case where the answer is not another employer at all.

Ghost jobs are a tax on job seekers that nobody voted for. You cannot repeal it. You can stop paying it in forty-minute instalments.

Frequently asked questions