Every job seeker has the same suspicion after the 14th application disappears that a meaningful share of what they applied to was never really a job. The suspicion is usually treated as morale, something to be managed rather than measured.

We had a way to measure a piece of it so we did.

Between 16 and 29 July 2026 we collected 528 job postings from 410 distinct companies and then for each one tried to answer a narrow question Is there specific human being attached to it and can we reach them ? Not the company .Not careers portal .A person

We could do it on 219 of the 528

These figures comes from 528 job postings mainly from linkedin with the reminder from We Work Remotely , Hacker news , Remote Ok

What we were actually testing

We are not claiming that 309 postings had no human behind them .We are claiming something narrower and more useful : on 309 of 528 postings a pipeline built specifically to find the hiring contact could not find one.

That is the point If a system designed to do nothing else fails on 58.5% of postings , an individual job seeker with browser and lunch break is not going to do better.

Why the number is this high
Four reasons :

  1. Postings that are collecting not hiring
  2. Reposts and evergreen listings
  3. Anonymous recruiting accounts
  4. Aggregation strips the author

What this means for how you apply

  1. Freshness is doing more work than tailoring - the postings where we found human skewed noticeably towards recently published
  2. 41% that are reachable deserve disproportionate effort
  3. Volume is rational response to opacity not laziness

If u want to look at raw end of this we build a free tool from our learnings that can help you filter ghost jobs Free job leads