Job posting statistics
Job posting statistics measure how long open job listings remain posted. Zeltru reads each posting from the employer’s own careers site, records the date the employer published it, and retains saved records during collection gaps. These figures describe 524,322 postings from 1,077 companies that carry an employer-supplied date, as of September 8, 2026.
Posting Longevity Index — September 2026
Three figures, each with the population it is taken over, computed from the saved dataset. The computation date does not confirm a successful check of every board.
As of September 8, 2026, the median tracked job posting has been listed for 32 days, across 524,322 postings carrying an employer-supplied date.
As of September 8, 2026, these figures are drawn from 1,077 companies that publish a posting date on their own careers site. Companies whose boards publish no date are excluded.
Posting longevity
A posting’s age here is the number of days since the date the employer published on their own careers site, as calculated in the dated dataset. The employer date is not a collection observation date. Where an employer publishes no date we record when we first saw the posting, and that is a floor rather than an age — those postings are excluded from every figure above.
These are composition figures. They say what the board looks like today, grouped by how long each posting has been up. They are not survival rates, and they cannot be read as “X% of postings close within 90 days” — that is a different measurement over a different population. The full distribution and industry breakdown sit on the pillar page, and the methodology states what is excluded and why.
Which postings are counted
Postings that carry a date supplied by the employer, and that are not flagged as evergreen or continuous-hiring. That is a subset of everything tracked: boards publishing no date are excluded, because an age measured from the day we first saw a posting is a floor, not the employer’s date. A figure on this site should always say which population it counts, and these three say so in the sentence.
Applicant-reported response rates
A company's reported no-response rate is shown only with at least 5 different applicants whose reports have settled. Each person's latest settled report supplies one data point. Individual reviews are not published. See the methodology for the calculation and publication minimums.
The term used in press coverage
Postings that stay open for months are sometimes called “ghost jobs” in press coverage — that is their term, not a category we measure; we report aggregate statistics and the method behind them and let readers draw conclusions.
A posting can stay open for many ordinary reasons: a backfill held open, continuous hiring for a high-turnover role, a slow internal process, or a board nobody has pruned. We publish how long each posting has been listed and link to the employer’s own page so anyone can check it.
Citing these figures
Free to cite with attribution to Zeltru. The aggregate figures behind this page are downloadable as CSV, read from the same source as the page so a citation and the page cannot disagree. Every figure states the date it was computed: September 8, 2026.