Google Jobs Scraper — Descriptions & Apply Links

What does Google Jobs Scraper do?

Google Jobs employer grouping: find hiring companies by job keyword and turn public Google Jobs search results into structured records. This private prototype is an unofficial Google Jobs API alternative, not a Google product. It does not submit applications, collect recruiter contacts, access login-only content, or guarantee every listing on every job board.

Why use Google Jobs search data?

Recruitment research, job-board feeds and labour-market monitoring all start with consistent role and employer data. Export job rows or group unique matching jobs by employer name for hiring signals. Employer-name grouping is not legal-entity verification. Apify offers API access, scheduling and integrations; scheduling should wait until coverage is validated.

What data can it extract?

Field	Type	Meaning
title	string	Public job title
company, location	string or null	Employer and displayed location
salaryText	string or null	Source salary text; no annualization
postedAtText	string or null	Displayed posting age or JSON-LD posting date
employmentType	string or null	Visible employment type
description	string or null	Description if present in the returned HTML
applyLinks	array	Public HTTPS destinations with source labels
jobId	string or null	Explicit source identifier when available
query, requestedLocation, collectedAt	strings	Search provenance and ISO collection time

Company mode returns company, jobCount, jobTitles, locations, applyLinks (URL array) and collectedAt. Missing salary, description, employer or apply links are not guessed. Results may expose only snippets: this implementation does not call undocumented job-detail endpoints.

How to scrape Google Jobs

1. Enter one role such as software engineer and a location such as London, UK.
2. Keep maxResults at 10 and maxPages at 1 for the first run.
3. Choose job rows or hiring-company groups.
4. Inspect SUMMARY for warnings before relying on coverage.
5. Download the dataset in JSON, HTML, CSV, or Excel.

Input

See the input tab for full configuration options.

{"queries":["software engineer"],"location":"London, UK","country":"gb","language":"en","maxResults":10,"maxPages":1,"outputMode":"jobs"}

queries accepts 1–10 searches. maxResults (1–1000) limits unique jobs before company grouping. maxPages (1–100) is the total request budget across all searches, not per query. More queries may remain unprocessed when that budget is reached. Only Apify's GOOGLE_SERP proxy is supported; no residential or browser fallback is attempted. Search country affects Google domain selection and the country parameter, but does not guarantee that Google interprets every location precisely. Interface parsing is English-first.

Output example (illustrative fixture, not live data)

[
  {"title":"Software Engineer (fixture)","company":"Example Ltd","location":"London","salaryText":null,"applyLinks":[{"source":"Example Jobs","url":"https://jobs.example.org/one"}]},
  {"title":"Data Engineer (fixture)","company":"Example Ltd","location":"Remote","salaryText":null,"applyLinks":[]}
]

Real rows include all fields listed above. SUMMARY reports requests, deduplication, output counts, per-page counts, resultLimitReached, coverage, stopReason and warnings. A SUCCEEDED run may have PARTIAL coverage; it does not guarantee that the requested result cap was reached. SOURCE-HTML holds the last private response for diagnostics. Data stays in the Actor run's default stores.

How much does Google Jobs scraping cost?

No Store monetization or price is configured. Private testing consumes platform compute, storage and Google SERP proxy usage. Apify's public pricing page currently lists up to $2.50 per 1,000 SERP requests depending on plan, plus compute. One request is not one guaranteed job: yield and missing fields vary. Proposed future pricing is $2 per 1,000 unique jobs, subject to the owner's approval and measured costs. Company output uses the same underlying unique-job unit, not the number of grouped company rows. The unique-job hook respects the run charge cap; unpriced/private runs do not activate fees. No margin or coverage is promised.

Limits, FAQ and support

Why did a search fail?

Consent, CAPTCHA, blocking and unrecognized HTML fail explicitly rather than silently returning an empty dataset. The Actor does not bypass consent, solve challenges, or retry with more expensive proxies. Google can change its markup at any time. A valid zero-result search requires an explicit empty-results marker.

Can it paginate?

Only ordinary public next-page links are followed, within the request budget. The tested Jobs responses had no such links. Standard Search start offsets were tested but did not paginate Jobs reliably, so they are not used as a fallback. JavaScript-only scrolling is not implemented: expect about ten jobs per individual search in the tested format. Opt-in expandQueries: true plans narrower senior/junior/lead/contract/graduate searches and deduplicates their first pages. Set maxPages to the total request budget (for example six) and maxVariantsPerQuery to the plan size. This is search fan-out, NOT true pagination or exhaustive collection; modifiers change the result mix. Each row retains both actual query and originalQuery. A result cap is a ceiling, not a promise. No undocumented async API is called. Jobs are deduplicated by explicit ID, otherwise normalized employer/title/location, with apply links merged.

Is this ready to launch?

This is a private technical beta with Store-ready draft documentation, not a validated commercial service. On 8 November 2026, cloud tests returned ten software-engineering jobs each for London, New York and Johannesburg with descriptions and apply destinations. A later opt-in London fan-out test returned 35 unique jobs, 35 descriptions and 35 apply-link sets from four searches, removing three duplicates. This improves breadth beyond ten results per original query but is not true pagination: role modifiers change the result mix. Exhaustive single-query coverage, larger-scale reliability and customer validation remain unproven. Use the Issues tab for feedback and the API tab for integration once access is approved.

Respect Google's terms, robots directives, source attribution and applicable privacy laws. Job descriptions may incidentally contain personal information; use a lawful purpose, minimize retention and do not use this tool to harvest personal contacts. Apply links may expire and employer names may be ambiguous. This Actor is not affiliated with Google or any source job board.

Expanded-search example (private validation)

{"queries":["software engineer"],"location":"London, UK","country":"gb","maxResults":35,"maxPages":6,"expandQueries":true,"maxVariantsPerQuery":6}

One generated search consumes one SERP request even if it yields only duplicates. Expansion is off by default to protect small-run costs. It samples related roles rather than guaranteeing all matches. No pricing change is configured.