◆TDW

Ghost Jobs: How to Spot a Listing Nobody Will Answer

About one job posting in five is not being actively filled. What the evidence says about how common ghost jobs are, why companies post them, which warning signs actually have data behind them, and the new laws in Ontario, New York and the UK.

A ghost job is a listing for a role the employer is not actively trying to fill. You apply, nothing happens, and there was never a realistic chance that it would. This article covers how common they are, why they exist, and what you can check before spending an hour on an application. It also separates the claims with evidence behind them from the advice that is simply repeated.

Three different problems

Ghost jobs are often lumped together with two other problems that call for a different response:

What it isRisk to you
Stale listingA real role that has been filled, frozen or cancelled while the ad stays up, often as a copy on another siteA wasted application
Ghost jobA real employer advertising a role it is not trying to fill soonA wasted application, and a distorted picture of the market
ScamA fake employer, or someone impersonating a real oneLost money or stolen identity

Scams are the only one of the three with official figures: US reported losses to job scams rose from $90 million in 2020 to $501 million in 2024, according to the FTC. They need different checks, covered in how to verify a listing, a recruiter and an offer. The rest of this article is about the first two.

How common are ghost jobs?

There are no official statistics in the US, UK or EU, and the US Congressional Research Service notes that the widely cited estimates come from companies that sell hiring services. The figures differ mostly because they measure different things:

  • Up to 21% of job ads, from Hunter Ng's 2024 study, which used a language model to classify Glassdoor interview reviews describing a process that went nowhere. It is a preprint, not peer reviewed.
  • 18–22% of postings in any quarter, from Greenhouse's own data on its hiring platform. Greenhouse sells hiring software and has not published how it classifies a ghost job.
  • 39–40% of companies posted a fake job in the past year, from a Resume Builder survey of 649 hiring managers in May 2024. That counts companies, not listings, and it is self-reported.
  • 27.4% of US LinkedIn listings, from a résumé company that simply counted every post older than 30 days as a ghost job.
  • 30% of US openings "did not result in a hire", from government openings and hires data. That gap says nothing about whether the employer intended to hire.

The defensible figure is about one listing in five. Two widely shared numbers need a warning. The UK's "34.4% of listings are ghost jobs" comes from two scrapes of Indeed in May and June 2023 that counted ads tagged "30+ days ago"; it has since been re-dated as 2025 and 2026 data in several reports. And the much-quoted Clarify Capital survey was carried out in August and September 2022, not 2023 or later.

There is no reliable figure for tech specifically. US government data for March 2026 shows openings in the information sector down 33% year on year, which suggests employers are cutting speculative postings rather than adding them.

Why companies post them

Hiring managers' own answers point to four kinds of motive. In Resume Builder's 2024 survey, where respondents could pick several reasons, 67% said they wanted to appear open to outside talent and 66% to look like the company was growing. 63% wanted current staff to believe help was coming, and 62% wanted them to feel replaceable. 59% were collecting résumés for later. A 2024 recruiter survey by MyPerfectResume adds quieter reasons: staying visible on job boards when not hiring, testing a job description, and building a talent pool.

Most ghost listings are probably not deliberate, though no source has measured each cause on its own:

  • Hiring freezes turn a real role into a dead one after it is posted.
  • Evergreen postings stay up for roles a company hires for continuously. Recruiters argue these are not ghost jobs when a real, recurring role sits behind them.
  • Internal candidates chosen in advance, where policy still requires a public posting.
  • Staffing agencies posting generic roles to build a pool of CVs.
  • Copies of a filled role that stay up on job boards that scraped the original.

US tech has one legally required kind. To sponsor a green card, an employer must advertise the role through the PERM labour certification process — a state job order, two Sunday newspaper ads and three further recruitment steps — even when the sponsored worker already does the job. In 2025 ProPublica followed up Sunday ads for software developers and systems analysts and found mail drops and empty offices; one manager said simply, "This is a PERM ad." Facebook (about $14.25 million, 2021) and Apple (up to $25 million, 2023) settled US Department of Justice claims that they kept these roles hard for outside applicants to find, without admitting liability. These ads appear mainly in print and on state job banks, but they explain a class of oddly specific US listings that seem written for one exact person.

Signals with evidence behind them

No public study has tested any listing feature — age, reposts, missing salary, vague text, applicant count — against whether someone was actually hired. So the honest approach is to rank the signals by how much evidence supports them.

Posting age, read against the platform the job is hosted on. This is the strongest check. In a September 2026 snapshot of 64,250 postings by SeekerScore (a vendor), more than 99% of postings on the Ashby, Lever and SmartRecruiters hiring systems had closed by day 60, against 85% on Greenhouse and only 30% on Workday. So an old posting means something very different depending on where the "apply" link leads.

For a benchmark, tech roles take about 62–75 days to fill: Ashby puts the median time from opening to first hire at 75 days for technical roles, and Workable at 62 days for engineering. Specialised roles legitimately take longer. As a rule of thumb:

  • under 30 days: normal;
  • 30 to about 75 days: inside a typical tech hiring cycle;
  • more than about 90 days on Ashby, Lever or Greenhouse, or the same role reposted three or more times over several months: unusual, and worth checking before you invest time.

On Workday, age alone tells you little. A repost also resets the visible age, so a "posted 3 days ago" label may hide a role that has been open for months.

Interview reviews describing a process that went nowhere. Ng's 21% estimate is built from exactly this kind of text on Glassdoor, so reading a company's recent interview reviews for the same role is a reasonable check.

Signals that don't tell you much

  • Missing salary. About 77% of LinkedIn postings had no salary in one 2026 analysis, so it does not separate real jobs from ghost ones. It means more where the law requires a range: a role covered by a US state transparency law that still shows no pay is worth a second look.
  • "Reposted" labels. A repost can mean an expired free post, a system refresh, a failed hire or an evergreen role. LinkedIn does not show how many times a job was reposted or its original date.
  • Verification badges. LinkedIn's verification confirms details about the employer or the person posting, which is a scam check, not a check that the role is open. Greenhouse Verified rates how the hiring process is run. Indeed's Responsive Employer badge measures how often an employer replies, and automatic rejection emails count.
  • A huge applicant count, or recent layoffs. No dataset links either to ghost jobs.
  • Ghost-job browser extensions. None publishes how well its score predicts anything.

Checks worth a few minutes

These are recommended by career services and recruiters. None has been tested against outcomes, but each is cheap:

  1. Open the actual application link. Confirm the role is live in the employer's own system or careers page. If it only exists on a job board, it may be a copy of a filled role.
  2. Read the age against the platform, as above.
  3. Search the same title at the same company for older postings of the same role.
  4. Skim recent interview reviews for that team or role.
  5. Ask before investing heavily. For a role that needs a long application or a take-home task, a short message to the recruiter asking whether the role is open and when they expect someone to start costs little.

Then treat what's left as a cost to manage. With about 3% of cold applications reaching an interview even for real roles (see what happens after you apply on each job board), silence is the normal outcome of most applications, and ghost jobs are only part of the reason. Volume is the larger one: applications per hire roughly tripled between 2021 and 2024.

The laws that arrived in 2026

  • Ontario (in force since 1 January 2026): employers with 25 or more staff must state whether a public posting is for an existing vacancy, disclose AI used in screening, include a pay range, and tell interviewed candidates within 45 days whether a decision has been made.
  • New York: bill S8877 passed both houses in June 2026. It would require postings to say whether the role is a current vacancy with an intended fill date within 90 days, require filled roles to come down within two weeks, and fine employers and job platforms from $2,500 per ad. At the time of writing it was awaiting the governor's signature. Similar bills are pending in California, New Jersey, Kentucky and Pennsylvania.
  • UK: no ghost-job law, but two existing rules already point the same way. The advertising code (CAP Code rule 20.2) says employment advertising "must relate to genuine vacancies", and the 2003 employment agency regulations bar agencies from advertising roles without the hiring company's authority. No enforcement against a ghost job has been found, and a 2025 parliamentary petition to ban them closed with 329 signatures.
  • EU: the Pay Transparency Directive covers pay information for applicants, not whether a vacancy is genuine, and most member states missed its June 2026 deadline.

What it costs

The cost is time multiplied by silence. In a 2026 UK survey by the recruiter Robert Walters, 18% of professionals said they spend two hours or more on a single application, and 44% said ghost listings had stalled their search. In an Indeed UK survey from 2024, 55% of job seekers said they don't hear back from more than half of their applications, and 52% said the silence makes them question whether to keep looking at all. In a 2026 US survey by Resume Builder, 58% of job seekers said they had interviewed for a role they suspected was a ghost job. These are vendor surveys, and most of them measure silence in general rather than ghost jobs specifically, but they show where the time goes.

How The Dream Work handles it

The 2026 laws regulate the listing itself: say whether the vacancy exists, take it down when filled. Some of that can be built into a board directly, and The Dream Work does it by design:

  • Every listing expires after 30 days unless the company renews it, and the job page shows both the date it was posted and the date it expires.
  • Every listing shows a salary range. A company cannot publish without a minimum and a maximum.
  • Posting costs something. Companies pay a flat fee per 30-day listing (the first one is free), which makes leaving a dead listing up less attractive than on boards where posting is free.
  • Every application gets an answer or a close within five days. If the company doesn't ask to connect or pass in time, the application is closed and you get an email saying so, so a role nobody is reviewing can't leave you waiting.
  • You can see whether the company opened your application in your pipeline.

None of this makes a ghost listing impossible, but it limits how long one can waste your time. You can browse tech jobs with the posting and expiry dates shown, or start with where and how to find the best remote tech jobs.

Limits of the evidence

Almost every figure on ghost jobs comes from a company that sells hiring software, résumé tools or job-search products, and none of their surveys is a random sample. No study has validated a warning sign against actual hiring outcomes. No figure exists for tech specifically, or for Germany, the Netherlands or France. The status of New York's bill should be checked before relying on it.

Sources

For candidates

See your match score on every open role.

Build your profile in 5 minutes and get scored against every active role.

Create free account →