Methodology

How Gaming Job Tracker collects and computes every figure on this site. Data last refreshed 2026-09-14. 10434 active jobs are tracked across 555 gaming studios right now.

What's counted

Active listings are scraped hourly directly from 555+ tracked studio applicant-tracking-system (ATS) boards (Greenhouse, Lever, Workday, Teamtailor, and others) — not from third-party job aggregators. A listing counts as "active" from the hour it's first seen until the studio's own ATS stops serving it.

How closures are handled

When a listing disappears from a studio's ATS board, it is marked closed and logged (ClosedJobLog: company, job family, and days-open) — it is never deleted from the underlying dataset. The public job-detail URL for a closed listing returns HTTP 410 Gone rather than 404, so search engines and readers can tell "this specific job closed" apart from "this URL never existed." Reports like Recently Closed Jobs and Ghost Jobs are built entirely from this retained history.

What tag-derived stats denominate on

Some fields (tools, platforms, education requirements) are extracted from each listing's description text by a separate tagging pass that hasn't reached every listing yet. Reports built on those fields (Tools, Platforms, Education) state their percentages against the number of listings tag extraction has actually reached, not against every active listing — a listing with no tags yet is not the same as a listing confirmed to mention no tools, and this site never conflates the two. Fields that are never null (like AI/ML detection) are instead stated as a share of all active listings directly.

Salary figures

The game industry salary hub and role pages compute their figures separately from the all-currency Salary Report above: they cover US, USD-denominated postings only, over a trailing 12-month window (postings first seen in the last 365 days). A repost of the same listing (same studio, a normalized version of the title, and the same disclosed range) is collapsed to a single posting rather than counted twice. Only annual and hourly pay periods are used — hourly ranges are annualized at 2,080 hours/year and labeled "annualized" wherever that conversion is used; monthly and daily pay periods are excluded entirely rather than approximated. A posting's contributed figure is the midpoint of its disclosed range (or its single value when only one bound is disclosed). A role or experience level needs at least 8 disclosed-pay postings to show a figure at all, and at least 20 to earn its own indexed page — smaller groups are labeled "not enough data" rather than shown as a precise-looking number. Most disclosed pay comes from postings in pay-transparency states (California, New York, Washington, Colorado, and similar) that require a posted range by law, not a random sample of the whole US job market — treat these figures as what disclosed postings say, not a scientific salary survey. Because roughly six in ten of these postings are in California specifically, the hub and every role page also show an "Outside California" median (postings with a known non-California US state) alongside the headline figure, so the California-driven skew is explicit rather than left for a reader to infer. Sources are the same structured ATS postings (Greenhouse, Lever, Ashby, and others) every other report on this site is built from.

Composite indices

Four indices are computed weekly from the same underlying WeeklySnapshot/Job/ClosedJobLog data as every other report:

Gaming Hiring Index (GHI)

A weighted composite on a 0-1000 scale (500 = neutral baseline): 40% Active Listings (current active count vs. an all-time baseline week), 30% Posting Velocity (this week's net change vs. the historical average net change), 20% Market Breadth (share of tracked studios that posted a new listing this week), and 10% Sentiment (ratio of new postings to closures this week). See Gaming Hiring Index.

Studio Hiring Intensity

A per-studio composite on a 0-100 scale, computed only for studios with 5+ active listings that week: 40% Velocity (recent 4-week average new postings vs. that studio's own longer-run historical average, capped at 100 — 0 when there isn't yet enough history for a real baseline), 30% Fill Speed (100 minus the average days-open of that studio's recently closed listings), and 30% Breadth (the share of the site's 13 job families the studio is actively hiring across). See Who's Hiring Hardest.

Studio Stability

A per-studio composite on a 0-100 scale, equally weighted across four components: Posting Consistency (low week-to-week variance in new postings), Ghost-Job Score (inverse of the share of that studio's active listings open 90+ days), Net Hiring Direction (a smoothed trend of the last 4 weeks' net change), and Fill Health (inverse of average days-open for recently closed listings). The composite score is tiered Strong & Stable (80+), Healthy (60-79), Watch List (40-59), Elevated Risk (20-39), or Significant Concern (below 20). See Studio Stability Scores.

Hiring Concentration (HHI)

The standard Herfindahl-Hirschman Index: the sum of each studio's squared share of total active listings, scaled to a 0-10,000 range. Below 1,500 is considered a competitive/unconcentrated market, 1,500-2,500 moderately concentrated, and above 2,500 highly concentrated — the same bands U.S. antitrust merger guidelines use. See Hiring Concentration.

Limitations

• Studio-type and parent-company classifications (see Studio Types and Parent Companies) come from a small, hand-curated metadata table, not every tracked studio — those reports state their coverage against the classified subset, not the whole market.
• Salary figures (Salary Report) are computed from all disclosed salaries observed over the last 6 months, across both active listings and listings closed in that window (deduplicated so a closed-then-reposted listing isn't counted twice). Each currency is reported on its own with no FX conversion applied — converting to a single currency would mean modeling an exchange rate rather than reporting what studios posted, and currencies below a minimum sample size are named but not tabled.
• Grouped tables (by job family, by experience level, by studio) suppress any group with fewer than 5 listings as statistically unreliable, rather than showing a misleadingly precise small-sample figure.
• Composite indices (GHI, intensity, stability, concentration) are calculated once per week; between runs, the reports above show the most recently completed week's snapshot, not a live recalculation.
• All figures reflect what this site's scrapers can see on public ATS boards — they are not a substitute for a studio's own hiring data.

Data last refreshed 2026-09-14.