Labour Market Insights API

Labour-market insights computed live from deduplicated job postings - occupations, industries, skills, compensation, and AI exposure.

GET  https://api.jobspipe.dev/v1/insights/...
POST https://api.jobspipe.dev/v1/insights/salary/benchmark

The Insights endpoints turn the postings corpus into aggregate labour-market answers: which occupations and industries are hiring, what skills they ask for, what they pay, and how exposed each group is to AI demand. Every number is measured live from deduplicated, liveness-tracked job ads at request time - never cached, never modeled. Occupations use ISCO-08 codes; industries use ISIC Rev.4 divisions. Requires authentication; each call costs 1 credit from your monthly quota.

Endpoints

Method	Path	Returns
`GET`	`/v1/insights/occupations`	ISCO-08 occupations ranked by posting volume
`GET`	`/v1/insights/occupations/{code}`	Snapshot: count, remote share, top companies, titles, countries
`GET`	`/v1/insights/occupations/{code}/compensation`	Advertised-salary percentiles (p10-p90)
`GET`	`/v1/insights/occupations/{code}/skills/top`	Skills ranked by mention share
`GET`	`/v1/insights/occupations/{code}/skills/trending`	Skills growing or declining vs the prior window
`GET`	`/v1/insights/industries`	ISIC divisions ranked by posting volume
`GET`	`/v1/insights/industries/{division}/skills/trending`	Trending skills inside one industry
`GET`	`/v1/insights/skills/{slug}/trend`	Monthly mention-share series for one skill
`GET`	`/v1/insights/technology/ai-exposure`	AI-skill share and salary premium per occupation or industry
`POST`	`/v1/insights/salary/benchmark`	Salary percentiles for a code or title match

Common parameters

Param	Applies to	Description
`window_days`	all	Look-back window over posting dates. Default `90`, max `365`.
`{code}`	occupation endpoints	ISCO-08 code: 4 digits exact (`2512`), or 1-3 digits to roll up a whole group (`25`).
`{division}`	industry endpoints	ISIC Rev.4 division, 2 digits (`62`).
`direction`	trending endpoints	`up` (default) for growing skills, `down` for declining. The scan covers twice `window_days` to compare against the preceding period.
`by`	ai-exposure	`occupation` or `industry`.

Sample sizes and suppression

Insight quality is stated, never implied:

List endpoints only return groups with at least 30 postings in the window.
Percentile endpoints return suppressed: true with percentiles: null when the sample is under 30 - small-sample statistics are never leaked.
Every percentile response carries a confidence tier: low (30-99 postings), medium (100-999), high (1000+), alongside the exact count.

Example

curl "https://api.jobspipe.dev/v1/insights/occupations/2512/compensation?country=US" \
  -H "Authorization: Bearer jp_live_your_key_here"

200 OK

{
  "occupation_code": "2512",
  "occupation_label": "Software Developers",
  "window_days": 90,
  "country": "US",
  "count": 8945,
  "currency": "USD",
  "percentiles": { "p10": 98000, "p25": 130000, "p50": 175500, "p75": 210000, "p90": 245000 },
  "confidence": "high",
  "suppressed": false
}

Salary figures are advertised pay from postings that state it, normalized to annual USD - a measurement of what employers offer publicly, which is not identical to survey-based earnings statistics.

Status codes

Status	Meaning
`200`	Success (possibly `suppressed: true`)
`400`	Invalid parameter
`401`	Missing or invalid API key
`402`	Monthly request quota exceeded
`404`	Unknown code, or no postings in the window
`429`	Per-second rate limit exceeded

See the error reference for handling guidance.

Endpoints

Common parameters

Sample sizes and suppression

Example

Status codes

On this page