Methodology
How the RoleWorth Ghost Job Index is calculated.
The index is built for citation, not accusation. It reports aggregate ghost-risk signals across RoleWorth-evaluated postings and suppresses posting-level data.
Source window
The public index uses the last 30 days of RoleWorth job evaluations. A monthly cron refresh stores a snapshot for the current calendar month. Repost calculations look back 90 days.
Aggregation level
Public rows are aggregated by source domain only when a source URL is available. When source URLs are absent, rows are explicitly labeled as normalized company keys, not domains. A row is published only when at least 3 evaluated postings exist in the 30-day window. This prevents a single candidate submission from naming a domain or company key in the top-20 table.
No PII and no posting-level leak
The aggregation excludes user IDs, IPs, emails, resumes, raw descriptions, candidate profile fields, and raw job URLs. The API returns only the monthly snapshot, summary counts, and company aggregate rows.
Ghost-risk flag
A posting counts as ghost-risk when the RoleWorth evaluation contains at least one risk flag or when the posting-legitimacy tier is "Proceed with Caution" or "Suspicious." The detector is pattern-based, not a claim about employer intent.
Metrics
Median posting age is calculated from the source posting date when available, otherwise the RoleWorth job record creation time. Salary disclosure counts a posting when either salary minimum or maximum is present. Source URL coverage is the share of evaluated postings with a stored source URL in the 30-day window. Reposted more than 3 times in 90 days means the same aggregate key and normalized title appears more than 3 times in that lookback window.
Known limitations
Aggregates reflect RoleWorth-evaluated postings, not the entire job market. Candidate source mix can bias the sample toward specific job boards, industries, or seniority bands. A ghost-risk flag is a screening signal, not proof that a company is acting in bad faith.
Product proof
See the product surface behind the claim.
Each page carries the matching RoleWorth surface in a glass-framed proof card: the radar, extension overlay, ATS matrix, review queue, dashboard, or package flow behind the promise.
The public promise, visible above the fold: score first, package second, approve before anything leaves.
Batch scan, ghost-risk skipped count, high-worth jobs, and Apply/Maybe/Skip routing in one product surface.