Ghost Jobs: What the Data Can and Cannot Tell You
We hold a corpus of 257,534 live job postings and re-check them continuously. We still cannot tell you that any specific posting is a ghost job, and neither can anyone else working from listings alone. This guide is about what is measurable, because that turns out to be more useful than the statistic you were looking for.
Key Takeaways
- • A posting staying visible proves visibility, not that an employer declined to hire. Those are indistinguishable from outside.
- • Age is not a tell. 46.1% of live postings are over 30 days old.
- • “Reposted 50 times” claims are usually an artefact. Most feeds mint a new ID when a description is edited.
- • Long-lived postings are an employer trait: some companies run 30%+ past 90 days, others literally 0%.
- • The real, measurable problem is staleness on our side and every board's: 10.5% of what we serve has not been re-checked in over a week.
Why we will not give you a ghost-job percentage
A ghost job is usually defined as a listing the employer has no intention of filling. To measure that you need to know the employer's intention. From listing data you can observe that a posting is still up, that it was up last week, and that it is up again today. None of those observations distinguishes “not hiring” from “still hiring, slowly.”
This is why published ghost-job rates vary so wildly. They are not measuring the same thing, and mostly they are measuring listing behaviour and labelling it intent. We would rather hand you the listing behaviour and let you draw the conclusion.
The honest version of the claim
“This posting has been continuously visible for 94 days” is a measurement. “This is a ghost job” is an inference about intent that the measurement does not support. Everything below stays on the first side of that line.
Three popular tells that do not survive checking
It has been up for ages, so it is fake
46.1% of live postings are more than 30 days old and 4.8% are past 90 days on the employer's own date. If age were the tell, roughly half the market would be fake. Age is normal, not suspicious.
They keep reposting it
This one is mostly a measurement artefact. Many feeds identify a posting by hashing its content, so editing a description mints what looks like a brand new listing. We found 28,816 groups of same company, title and location where no two rows share an identifier. You cannot separate a genuine repost from a typo fix.
One company posted the same job 171 times
We checked exactly this and it collapsed. The biggest apparent case was one posting per branch location, which is a bank hiring in 171 towns. The largest genuine same-location repeat in the entire corpus is 11, and only 30 groups reach 5.
Skip the guessing about which listings are real
Seeker ranks live postings by how well they actually fit your skills, so you spend your applications where you are competitive.
New here? Learn how the resume match score works →
What is genuinely measurable
Three things hold up, and they are more actionable than a ghost-job rate.
Long-lived postings are an employer habit
Of postings re-confirmed within the last week, 8.2% have been live 90 days or more, and that concentrates by company rather than by role. Some employers run 30% or more of their postings past 90 days. Others sit at 0.0%. The same job at two companies gets a different lifespan, which is why age tells you about the employer, not the job.
Feeds go quiet while their listings stay visible
This is the concrete version of the ghost-job worry. Six of our twenty sources have added nothing in over a week, and they hold 27,046 active postings, 10.5% of our corpus. Those listings are still served and still look live. They are not fake, but nobody has re-verified them recently.
The date you are shown is not the employer's date
On the employer's own posted date the median live posting is already 28 days old, and only 0.2% were opened in the last day. Much of the 'this is stale' instinct is really a reaction to board date labels being generous.
We are part of the problem, so here is our number
It would be convenient to write this guide as though staleness is something other job boards do. It is not. Six of the twenty sources we ingest exhausted a monthly API quota in early August and have added nothing since. The 27,046 postings they hold are still marked active and still served, last re-verified in the first week of the month.
That is 10.5% of what we show. It is a supply problem rather than a deception problem, and it is exactly the kind of thing every board has and few publish. If you want the full accounting of how long postings survive and which feeds go quiet, it is on how long job postings live.
What to do instead of ghost-hunting
Stop reading silence as a verdict
A listing that never replied may have been filled, abandoned, mirrored from a page that no longer exists, or simply slow. None of those are about you, and you cannot tell them apart, so do not spend effort trying.
Apply on the employer's own site where you can
It removes the aggregator layer, which is where most mirroring and staleness is introduced. If the role is gone from the company's own careers page, that is the strongest available evidence, and it is stronger than any age heuristic.
Judge fit, not freshness
You cannot verify intent, but you can verify whether you clear the stated requirements. That is the variable you control and it is the one that moves callback rates.
Common questions
What percentage of job postings are ghost jobs?
We will not give you a number, because we cannot measure employer intent from listing data and neither can the studies that do quote one. What we can tell you is that 46.1% of live postings are over 30 days old and 10.5% of our own corpus sits on feeds that have gone quiet.
Do companies post jobs they never intend to fill?
Some do, for pipeline building, compliance and internal reasons. The part we can measure is that continuous visibility is an employer-level habit, so the companies whose postings stay up longest do it across their whole listing set rather than on particular roles.
Is it worth applying to old job postings?
Yes, because the alternative removes about half the market on a weak signal. Age on a listing mostly reflects the employer's posting habits and the board's refresh cadence, not whether the role is live.
How do I find jobs that are definitely real?
There is no certain test from the outside. The closest practical check is whether the role still appears on the employer's own careers page, since that removes the aggregator mirroring layer where most stale listings survive.
See how this applies to your resume
Upload your resume and see which roles match your skills, where the gaps are, and what to target next. Free, about 60 seconds.
Analyze my resumeData & insights
Career transitions