Publication note. Replaces earlier commentary at this URL. Broad prevalence estimates and unsupported claims about employer intent have been removed. The revised piece explains what each measure can establish.
Start with the event being counted
An online ad is a record of recruiting activity. A BLS job opening is a position meeting the survey’s criteria, including available work, the ability to start within the specified period and active recruiting outside the establishment. A hire is an addition to payroll. The measures answer different questions.
An ad can persist after a role changes. One ad may cover multiple positions, while the same role may appear in several places. A count of ads therefore needs a stated unit and rules for duplicates before it can support a claim about demand.
Sources: U.S. Bureau of Labor Statistics
Intent is the difficult part
The term ‘ghost job’ is used for several behaviors: deliberately advertising without an intention to hire, leaving a filled position online, maintaining an evergreen recruitment pool, or carrying a stale copy on another site. Some are administrative failures; some could be deceptive. The worker’s wasted time can be real in each case, but the causal claim differs.
A long posting duration alone cannot establish deception. The same observation can fit an employer struggling to recruit, a paused requisition, repeated hiring or a neglected feed. An allegation about deliberate conduct needs records, consistent corroboration and an opportunity for the subject to respond.
Fraud is a separate category
FTC reporting on task scams documents schemes that imitate paid work and then seek payments from the supposed worker. These reports establish a serious consumer harm, but complaint data do not measure the share of all vacancies that are fake. Reporting and sampling methods belong beside any total derived from those records.
Sources: Federal Trade Commission
A useful public measure
A defensible tracker could report whether an observed listing remains accessible, when it was first and last seen, and whether an authoritative employer page still carries it. It should state what cannot be observed, especially whether the position was filled or whether the employer intended to recruit.
Before publishing a rate, the researcher should test duplicate handling, distinguish observation date from advertised posting date, and assess missing data. A missing salary field, for example, can mean that the advertisement omitted pay or that the extraction failed. The public finding must not silently choose the more dramatic explanation.
These distinctions strengthen the worker’s case. They identify which failures can be documented and which questions require records or reporting. A narrower, reproducible measure is more useful than a sweeping prevalence claim that cannot survive scrutiny.
Scope and limits
This guide offers measurement principles. It does not estimate a national ghost-job rate, validate a particular employer, or conclude that an old advertisement is fraudulent.
Sources
- U.S. Bureau of Labor Statistics. JOLTS data definitions. Accessed September 9, 2026.
- Federal Trade Commission. Paying to get paid: gamified job scams drive record losses. December 12, 2024.
Cite this article
Opening Research. “What a job posting actually measures” September 9, 2026. https://opening.works/research/ghost-jobs-the-listings-that-were-never-real. Include your access date. Cite the original study when using its estimates.