Google Jobs Scraper - Salary, Apply Links & Dates, No Login

Google Jobs Scraper searches Google Jobs for your job titles and a location and returns one row per job: title, company, location, salary, job type, posting age, full description and the apply links to LinkedIn, Indeed or the employer site. It reads the public Google Jobs results page, so you need no Google login and no API key. The default input (python developer in Berlin, 10 jobs) usually takes 10–60 seconds and costs $0.033.

Use it for job market research, recruiting lead lists, salary checks and job feeds. The smallest run is one job title in the Job searches field.

Table of contents

• What it does
• Quick start
• Use cases
• Input
• Output
• Pricing and limits
• API, export and integrations
• Troubleshooting
• FAQ
• Related Actors
• Support

What it does

1. You add one or more job searches, for example python developer or registered nurse, and a location.
2. The Actor opens the Google Jobs results page for each search in the right Google country.
3. It reads every job card: title, company, location, the site Google found the job on (via), posting age, salary and job type.
4. It reads the job panel too: the full description, Google's highlights and all apply links, without tracking parameters.
5. It removes duplicate jobs (same Google job id) and saves one Dataset row per job.
6. Fast mode reads one Google page per search, about 10 jobs. Deep mode repeats each search with Google's date and job-type filters and removes duplicates, so you get several times more jobs.

Good fit

• Job market research: who hires for a role in a city, and on which job boards.
• Lead lists for recruiters and staffing agencies: companies with open roles, with the link to each job.
• Salary checks: Google's salary text where the job has one.
• A daily job feed for a niche job board, a Slack channel or a spreadsheet.

Not a fit / limits

• 8 countries only: United States, United Kingdom, Canada, India, Germany, France, Spain and Brazil (countryCode). Google shows the texts in the language of that country, for example vor 1 Tag and Vollzeit in Germany.
• Google shows about 10 jobs per search page. Fast mode gives about 10 jobs per search; deep mode about 20–70 (our tests). For more jobs, add more searches (up to 50) or more specific titles.
• salary is often null: Google shows a salary only for some jobs (0–40 % in our tests).
• highlights is often null in fast mode: Google does not send this block on many pages. The full description is still there.
• No recruiter names, emails or phone numbers. The Actor saves only the public job post.
• Google decides which jobs it shows. Two runs a few minutes apart can return a slightly different set or order.
• Google's terms do not allow automated access to Google Search. You are responsible for how you use the data.

Quick start

1. Open the Input tab.
2. In Job searches (queries), keep python developer or add your own job title.
3. Keep Location Berlin and Max jobs 10, or change them. Country auto picks the country from the location.
4. Click Start.
5. Open the Output tab. The Jobs table shows title, company, location, salary and the apply links.

Expected result: about 10 Python developer jobs in Berlin in Google's order, each with company, posting age and at least one apply link.

Use cases

Buyer job	Input path	Output entity	Key fields	Next integration
Job market research	queries + location, mode: "deep"	Job	title, company, via, postedAt	CSV or Excel export
Recruiting lead list	queries like "sales manager" + datePosted: "week"	Job	company, title, location, applyLinks	Google Sheets or CRM through Apify integrations
Salary check	queries + location	Job	salary, employmentType, title	Spreadsheet
Daily job feed	queries + datePosted: "today" on a schedule	Job	jobId, title, googleJobsUrl	Slack, email or a website through webhooks

Input

Input fields

Field	Plain name	Type	Required	Default / prefill	Allowed values	Effect	Example	Recommendation
queries	Job searches	array of strings	yes	prefill ["python developer"]	1–50 searches	One Google Jobs search per item	["data analyst", "nurse"]	Use job titles people really write in job posts
location	Location	string	no	prefill Berlin	City, region or remote	Added to every search	"Austin, TX"	Leave empty to search the whole country
countryCode	Country	string	no	auto	auto, us, gb, ca, in, de, fr, es, br	Google country and language	"gb"	Keep auto when the location is a known city
datePosted	Date posted	string	no	any	any, today, 3days, week, month	Only jobs posted in this period	"week"	Use today for a daily feed
employmentType	Job type	string	no	any	any, fulltime, parttime, contractor, internship	Only jobs of this type	"fulltime"
mode	Search depth	string	no	fast	fast, deep	deep repeats each search with Google's filters	"deep"	Use deep when 10 jobs per search are not enough
maxJobs	Max jobs	integer	no	default 50, prefill 10	1–1,000	Stops the run at this many jobs in total	200	Keep the default 900-second timeout for more than 300 jobs
includeDescription	Include description	boolean	no	true	true / false	false leaves out description and highlights	false	Turn off for smaller rows; the price is the same

Duplicate searches (same text, any case) run only once. A job that two searches find is saved once. If queries is empty, the run fails at once with INVALID_INPUT and makes no requests. A country outside the list above is rejected by the input form.

Interacting settings

When	Input A	Input B	Effective behavior	Example	Cost/result effect
Empty location	location: ""	countryCode	The country name is added to the search, because Google Jobs needs a place	"nurse" → nurse USA	none
Country auto	countryCode: "auto"	location	Country from the location; unknown places use the United States	"Lyon" → France	none
Date or type filter	datePosted, employmentType	Google's results	Google's own filter, then jobs whose shown date or type does not match are dropped (filteredOut in RUN_SUMMARY). Jobs without a shown date or type are kept	datePosted: "week"	Dropped jobs are free; the page is still charged
Deep mode	mode: "deep"	datePosted, employmentType	Up to 9 Google pages per search: the base search, plus one per date filter if datePosted is any, plus one per job type if employmentType is any. With both filters set, deep mode is the same as fast	mode: "deep"	Up to 9 charged pages per search, several times more jobs
Job limit	maxJobs	queries	Searches run in order and stop at maxJobs; later searches may not run	maxJobs: 10 with 3 searches	RUN_SUMMARY.stopReason: "MAX_JOBS"

Input examples

Python jobs in Berlin (prefill)

{
  "queries": ["python developer"],
  "location": "Berlin",
  "maxJobs": 10
}

Expected behavior: one Google Jobs page in Germany (countryCode auto → de). Expected output: 10 job rows with title, company, via, postedAt and applyLinks. Limit or cost note: $0.033 on the Free plan; about 10 seconds in our last test.

More jobs per search: deep mode in Houston

{
  "queries": ["registered nurse", "pharmacist"],
  "location": "Houston, TX",
  "mode": "deep",
  "maxJobs": 100
}

Expected behavior: up to 9 Google pages per search (base, 4 date filters, 4 job types); duplicates removed. In our test these two searches found 60 and 59 jobs. Expected output: 100 jobs with searchVariant showing which search found each job. Limit or cost note: at most $0.34 on the Free plan (1 start, up to 18 pages, 100 jobs); usually a few minutes.

Output

The Actor saves one row per job in the default Dataset. A search with no jobs gets one free not_found row, and a search that failed gets one free error row, so you always see what happened to each search. A run overview is saved as RUN_SUMMARY in the default Key-value store.

Output fields

Field	Meaning	Type	Can be empty?	Example / enum	Source / notes
status	Row result	string	no	complete, partial, not_found, error	partial = deep search where some Google pages did not answer
title	Job title	string	yes	Python Team Lead, Integration team	As the employer wrote it
company	Employer	string	yes	Fliff Inc
location	Job location as Google shows it	string	yes	Berlin, Beliebiger Ort (anywhere)	In the language of the country
via	Site Google found the job on	string	yes	LinkedIn, Indeed
postedAt	Posting age as Google shows it	string	yes	vor 1 Tag, 2 days ago	null when Google shows no date
postedAtApprox	Approximate posting date	string (YYYY-MM-DD)	yes	2026-10-14	fetchedAt minus postedAt
salary	Salary as Google shows it	string	yes	65.000 € bis 85.000 € pro Jahr	null = Google shows no salary (not zero)
employmentType	Job type as Google shows it	string	yes	Full-time, Vollzeit
workFromHome	Remote job tag	boolean	yes	true	null when Google does not say
description	Full job description, plain text	string	yes	We're looking for a Python Team Lead…	null when includeDescription is off
highlights	Qualifications, Responsibilities, Benefits	array	yes	[{"title": "Qualifications", "items": ["…"]}]	Often null in fast mode
applyLinks	Where to apply	array	yes	[{"title": "LinkedIn", "url": "https://…"}]	Tracking parameters removed
jobId	Google's job id	string	yes	eAHxzL0ATotpbjXSAAAAAA==	Same job, same id across runs
googleJobsUrl	Google page that opens this job	string	yes	https://www.google.de/search?ibp=htl;jobs&…
position	Order in this search after removing duplicates	integer	yes	2	null on not_found and error rows
query, searchLocation, countryCode	Your search, your location, the Google country	string	searchLocation yes	python developer, Berlin, de
searchVariant	Which deep search found the job	string	yes	datePosted:week	null in fast mode
source, sourceUrl	Source name and the Google URL that was read	string	no	google_jobs
fetchedAt	When the page was read, UTC	string (ISO 8601)	no	2026-10-15T06:20:55.335Z
statusReason, errorCode	Why a row is not complete	string	yes	SOURCE_TIMEOUT	null on complete rows

null means "Google shows no value"; it never means zero.

Output entities and surfaces

Situation	Dataset/API/KVS/live view	Result shape	User action
Normal item	Dataset, view Jobs	status: "complete" job row	Use title, company, applyLinks
No jobs for a search	Dataset	one free not_found row with statusReason	Try a wider location or another title
Partial result	Dataset	status: "partial" rows of a deep search	Run again later for the missing pages
Technical failure	Dataset + RUN_SUMMARY	free error row with errorCode; run status PARTIAL or ERROR	Run again later; see Troubleshooting
Run overview	Key-value store record RUN_SUMMARY	status, stop reason, counts per search, charged pages and jobs	Check status, stopReason and counts

Output example

{
  "status": "complete",
  "title": "Python Team Lead, Integration team",
  "company": "Fliff Inc",
  "location": "Beliebiger Ort",
  "via": "LinkedIn",
  "postedAt": "vor 1 Tag",
  "postedAtApprox": "2026-10-14",
  "salary": null,
  "employmentType": "Vollzeit",
  "workFromHome": true,
  "description": "We’re looking for a Python Team Lead to join our Integration team, which owns a broad set of critical integrations across the product…",
  "highlights": null,
  "applyLinks": [
    { "title": "LinkedIn", "url": "https://de.linkedin.com/jobs/view/python-team-lead-integration-team-at-fliff-inc-4479228357" },
    { "title": "Startup Jobs", "url": "https://startup.jobs/python-team-lead-integration-team-fliff-inc-10437417" }
  ],
  "jobId": "eAHxzL0ATotpbjXSAAAAAA==",
  "googleJobsUrl": "https://www.google.de/search?ibp=htl;jobs&q=python+developer+Berlin&htidocid=eAHxzL0ATotpbjXSAAAAAA%3D%3D&hl=de-DE#…",
  "position": 2,
  "query": "python developer",
  "searchLocation": "Berlin",
  "countryCode": "de",
  "searchVariant": null,
  "source": "google_jobs",
  "sourceUrl": "https://www.google.de/search?q=python+developer+Berlin&udm=8&hl=de&gl=de",
  "fetchedAt": "2026-10-15T06:20:55.335Z",
  "statusReason": null,
  "errorCode": null
}

This is one row from the prefill run. The description, one apply link and the end of googleJobsUrl are shortened here. This job has no salary on Google, so salary is null.

Moving from another Google Jobs Actor

Field names are different from other Google Jobs Actors, so this Actor is not a drop-in replacement. Common changes: company_name → company, share_link → googleJobsUrl, job_highlights → highlights, apply_options → applyLinks (with url instead of link), job_id → jobId (Google's own id), search_timestamp → fetchedAt. Google's tag list (extensions) is split into salary, employmentType, postedAt and workFromHome.

Pricing and limits

Pricing unit

This Actor uses pay-per-event pricing. You pay for:

Event	When it is charged	Price
Actor start (apify-actor-start)	Once per run (runs up to 1 GB memory)	$0.01
Search page (search-page)	Each Google Jobs page that Google answered (up to 10 jobs), also a page with no jobs	$0.013
Job (job)	Each unique job saved to the Dataset	$0.001 (Free plan), $0.00095 (Bronze), $0.0009 (Silver), $0.00085 (Gold and higher)

Duplicates, jobs dropped by your date or type filter, not_found rows and error rows are free. Pages that failed are not charged, and a page that answers after maxJobs is reached is not charged. A search with no jobs costs one search page ($0.013). Apify platform usage (compute, proxy, storage) is included; you do not pay it separately.

If you set a maximum charge for the run, the Actor stops before it goes over it: RUN_SUMMARY.status: "CAPPED", stopReason: "CHARGE_LIMIT".

Cost examples

Prices on the Free plan, from our test runs.

Scenario	Input scope	Pages	Jobs	Total
Prefill	1 search, fast, 10 jobs	1	10	$0.033
100 jobs	10 searches, fast	10	100	$0.24
1,000 jobs	22 searches, deep	196	1,000	$3.56
Nothing found	1 search, no jobs	1	0	$0.023

Measured run times: 10 jobs in about 10–60 seconds (up to about 3 minutes when Google answers slowly), 100 jobs in about 2 minutes, 500 jobs in about 8 minutes, 1,000 jobs in about 8–9 minutes. Google answers each page in 4–20 seconds; more memory does not make it faster.

The default run timeout is 900 seconds. If a run reaches its timeout, the Actor stops by itself and keeps what it saved (status: "CAPPED", stopReason: "DEADLINE"). For more than 300 jobs, keep the timeout at 600 seconds or more, or split the searches into several runs.

API, export and integrations

• API: start the Actor with the Apify API and read rows from the default Dataset (view jobs).
• Export: download the Dataset as JSON, CSV, Excel, XML or HTML from the Output tab.
• Scheduling: create a schedule with datePosted: "today" for a daily job feed. Use jobId to skip jobs you already have.
• Webhooks / no-code: send new rows to Slack, Google Sheets, Make, Zapier or n8n through Apify integrations.

Use with AI agents (MCP)

• MCP server with only this Actor: https://mcp.apify.com?tools=cliqtomedia/google-jobs-scraper.
• Smallest useful input:

{ "queries": ["python developer"], "location": "Berlin", "maxJobs": 10 }

• Key output fields: title (job title), company (employer), location, salary (text or null), applyLinks (where to apply), status (complete, partial, not_found or error).
• A 10-job run usually takes 10–60 seconds and costs $0.033 on the Free plan. Keep maxJobs small (10–50) and mode: "fast" for agent calls; 1,000 jobs take about 8 minutes.

Troubleshooting

The run failed at once

Symptom: run status FAILED in a few seconds, no rows. Likely reason: invalid input (INVALID_INPUT), for example no searches. Check: the run log and RUN_SUMMARY.message. Action: fix the input and start again.

A search returned one not_found row

Symptom: one row with status: "not_found". Likely reason: Google Jobs has no jobs for this search, place and filter. Check: statusReason and sourceUrl; open sourceUrl in a browser. Action: try a more common job title, a bigger city, or datePosted: "any".

Error rows with SOURCE_TIMEOUT or SOURCE_BLOCKED

Symptom: free error rows with errorCode SOURCE_TIMEOUT, SOURCE_UNAVAILABLE or SOURCE_BLOCKED. Likely reason: Google did not answer in 60 seconds after retries, or refused the request. Check: RUN_SUMMARY.http (requests, retries, timeouts, blocked). Action: run these searches again later. The Actor does not solve CAPTCHAs.

The run stopped with CAPPED and DEADLINE

Symptom: RUN_SUMMARY.status: "CAPPED", stopReason: "DEADLINE", fewer jobs than maxJobs. Likely reason: the run timeout was too short for the number of jobs. Check: the run timeout in the run options. Action: keep the default 900 seconds, or split the searches into several runs. Searches that did not start get free error rows.

Fewer jobs than maxJobs

Symptom: the run is COMPLETE with fewer jobs than you asked for. Likely reason: in fast mode Google shows about 10 jobs per search; filters drop jobs that do not match. Check: RUN_SUMMARY.counts (jobs, duplicates, filteredOut). Action: use mode: "deep" or add more searches.

FAQ

Do I need a Google account or API key?

No. The Actor reads the public Google Jobs results page without login, cookies or keys.

Is this the official Google Jobs API?

No. Google has no public API for Google Jobs search results. This Actor is not made by or linked to Google.

Why are some texts in German or French?

Google shows jobs in the language of the country you search. Search in Germany, and postedAt says vor 2 Tagen and employmentType says Vollzeit. postedAtApprox gives a date that is the same in every language.

Why is salary empty for most jobs?

Google shows a salary only when the job post or job board gives one. The Actor does not guess: no salary on Google means salary: null.

Can I get the recruiter's email or phone?

No. The Actor saves only the public job post and does not collect personal contact data.

Related Actors

• Remote Jobs Aggregator — remote jobs from RemoteOK, We Work Remotely and Himalayas.
• Greenhouse Jobs Scraper & Monitor — all jobs of a company that you found on Google, straight from its Greenhouse job board.
• Lever Job Scraper — the same for companies on Lever.

Support

For bugs, questions or feature requests, open an issue on the Issues tab of this Actor. Include the run ID, your input and the error text. Never send tokens, cookies or private data.