Glassdoor Salary Data by Job Title

Get the median pay for any job title on Glassdoor — base pay, total pay, the full percentile spread, and how many reported salaries each figure is based on.

Why use this actor

• Median base pay and median total pay, kept separate — most tools blur the two
• Full percentile spread (10th / 25th / 50th / 75th / 90th) so you can see the range, not just the midpoint
• Sample size on every row — a median from 58,000 reports means something different from one built on 40
• Additional pay broken out (bonus, commission, profit sharing) where Glassdoor publishes it
• No account and no key required
• Stable JSON output, ready for a spreadsheet or a database

How it works

1. You give the actor a list of job titles (or paste Glassdoor salary page addresses directly).
2. It looks up each title's salary page and reads the published pay figures.
3. Every row records the job title, the pay figures, the currency, the pay period and the number of salary reports behind it.
4. Results stream into your dataset, ready to download as JSON, CSV or Excel.

You don't need to manage scrapers or infrastructure. Use the default US residential proxy — Glassdoor limits how many salary pages a single address may request in a short window, and a fresh address per request is what keeps results complete.

Input

{
  "jobTitles": ["data scientist", "product manager"],
  "maxItems": 20,
  "proxyConfiguration": {
    "useApifyProxy": true,
    "apifyProxyGroups": ["RESIDENTIAL"],
    "apifyProxyCountry": "US"
  }
}

Field	Type	Description
jobTitles	array	Job titles to look up, e.g. "software engineer". One row per title.
startUrls	array	Optional. Glassdoor salary page addresses, for company- or location-specific pages.
maxItems	integer	Maximum titles/pages per run. Default 20.
proxyConfiguration	object	Defaults to US residential — recommended, see above.

Output

Real sample from a live run:

{
  "_input": "data scientist",
  "_source": "S1-detailed",
  "_scrapedAt": "2026-08-26T08:12:44Z",
  "recordType": "SALARY_AGGREGATE",
  "url": "https://www.glassdoor.com/Salaries/data-scientist-salary-SRCH_KO0,14.htm",
  "jobTitle": "Data Scientist",
  "salariesCount": 58116,
  "payPeriod": "ANNUAL",
  "currency": "USD",
  "baseMedian": 120678.43,
  "basePay": { "percentile10": 79345.37, "percentile25": 96771.16, "median": 120678.43, "percentile75": 150491.99 },
  "totalMedian": 157630.82,
  "totalPay": { "percentile25": 124485.45, "median": 157630.82, "percentile75": 202225.34 },
  "additionalPay": { "percentile10": 21432.39, "percentile25": 27714.29, "median": 36952.39, "percentile75": 51733.35 }
}

Field	Type	Description
jobTitle	string	The job title as Glassdoor labels it.
salariesCount	integer	How many reported salaries the figures are based on.
baseMedian	number	Median base pay.
basePay	object	Base pay percentiles (10th–90th, where published).
totalMedian	number	Median total pay (base plus additional).
totalPay	object	Total pay percentiles.
additionalPay	object	Bonus / commission / profit sharing percentiles.
payPeriod	string	Usually ANNUAL.
currency	string	Currency code, e.g. USD.
url	string	The page the figures came from.

Notes and limits

• Coverage measured 5 of 6 titles on a live run (nurse, accountant, marketing manager, software developer, teacher all returned full figures; software engineer did not). A handful of very high-traffic pages are served in a layout that carries no aggregate — software developer works where software engineer does not, so try a close synonym if a title comes back empty.
• A title with no published data returns an error row, not an empty one. If Glassdoor has no aggregate for a title (or serves a reduced page because the address has been requesting too quickly), the row carries _error explaining which happened, so a gap never looks like a zero.
• Base pay is only available on the richer page layout. When only the summary layout is served, total pay is reported and baseMedian is left empty rather than quietly filling it with a total-pay figure — the two are genuinely different numbers.
• Run titles in modest batches. Requesting many pages very quickly from one address triggers a cool-down; the residential proxy default handles this for you.
• Figures are Glassdoor's own estimates from self-reported salaries, not verified payroll data.