Randstad Jobs Scraper: Germany, Switzerland and the UK

Get the public job listings of Randstad in Germany (randstad.de), Switzerland (randstad.ch) and the United Kingdom (randstad.co.uk) as one clean dataset. Search by keyword, filter by location and posting date, and optionally add the full job description and the consultant contact. One row per job.

What you can do with it

• Track the temporary and permanent jobs one of the largest staffing companies posts in three countries, in one run.
• Build a daily feed of new jobs per field, for example nursing, logistics or IT, with the posting date on every row.
• Compare pay, hours and contract types across Germany, Switzerland and the UK.
• Find which employers use Randstad for their hiring, from the end client name where Randstad shows it.
• Feed job boards, lead lists, market reports or alerts with structured data instead of web pages.

What you get

One row per job. Fields that a country does not publish are null, lists are always [] when empty.

Field	What it holds
id, job_id, reference	Row id (randstad-<country>-<job id>), Randstad's job id and the vacancy reference
title	Job title
company_name, client_name	Company on the job (Randstad when the end client is hidden) and the end client when Randstad shows it
location, city, region, postal_code, country	Place of the job, country as ISO code (DE, CH, GB)
latitude, longitude, street	With job details (all countries; Swiss jobs give coordinates and postal code, no street)
employment_type, employment_type_normalized, hours, work_time, contract_duration	Contract type as Randstad shows it, a normalized value, working hours and duration
salary_min, salary_max, salary_currency, salary_period, salary_text, salary_note	Pay as published (Germany and UK; Switzerland publishes no pay in its listing)
description, description_html, description_truncated	Short text in the listing, full text and HTML with job details
responsibilities, requirements, benefits	Lists from the job page and the benefit tags of the listing
specialism, sub_specialism, industry, role, categories	Field and category tags
created_at, updated_at, expires_at, start_date, end_date	Dates (ISO 8601)
job_url, apply_url, external_url, alternate_urls	Links to the job on Randstad, the application and the original posting; Swiss jobs list the language versions
contact_name, contact_email, contact_phone, contact_office, contact_street, contact_city, contact_postcode	The consultant and branch, with job details (Germany and UK; Switzerland gives the consultant name and email)
language, detail_fetched, source_query, source_country, scraped_at	Language of the text, whether the job page was read, the search that found the job, the country you asked for and the collection time

Every job appears once per country, even when several searches or pages find it.

Input

Field	Type	Description
countries	string list	DE, CH, UK. Default ["DE"]. Lower case, GB, a comma separated text and country names are accepted
searchQueries	string list	Keywords such as a job title or field. Each keyword runs in every country. Empty means all jobs
location	string list	Optional places, for example Berlin, Zurich, Manchester or a postcode start such as 80. Only matching jobs are returned. In Germany and the UK, together with a keyword, the place is searched directly on Randstad, which is much faster
postedWithinDays	integer	Only jobs posted in the last N calendar days (UTC, today included). 0 means no limit
maxResults	integer	Total number of jobs over everything. Default 100, 0 means no limit
maxResultsPerQuery	integer	Jobs per country and keyword. Default 50, 0 means no limit
includeDetails	boolean	Read each job page for the extra fields listed above. Default false
startUrls	string list	Optional Randstad search URLs, each one runs as an extra search

Example: nursing jobs in Germany and the UK

{
  "countries": ["DE", "UK"],
  "searchQueries": ["pflege", "nurse"],
  "postedWithinDays": 7,
  "maxResultsPerQuery": 50,
  "maxResults": 200
}

Example: everything new in Switzerland

{
  "countries": ["CH"],
  "postedWithinDays": 2,
  "maxResults": 0
}

Example: full descriptions and contacts in one city

{
  "countries": ["DE"],
  "searchQueries": ["logistik"],
  "location": ["Hamburg"],
  "includeDetails": true,
  "maxResults": 40
}

Supported URLs

• https://www.randstad.de/jobs/ and https://www.randstad.de/jobs/q-pflege/
• https://www.randstad.ch/jobs/ and https://www.randstad.ch/jobs/q-pflege/ (language prefixes such as /de/jobs/ work too)
• https://www.randstad.co.uk/jobs/ and https://www.randstad.co.uk/jobs/q-nurse/

Page numbers in a URL are ignored, the actor reads all pages up to your limits. URLs of single jobs are not search URLs and are skipped with a note in the log.

Output

{
  "id": "randstad-de-c01319492",
  "job_id": "C01319492",
  "title": "Pflegefachkraft (m/w/d)",
  "company_name": "Randstad Deutschland",
  "client_name": null,
  "location": "Berlin",
  "city": "Berlin",
  "postal_code": "10115",
  "country": "DE",
  "employment_type": "Zeitarbeit",
  "hours": "Vollzeit",
  "salary_currency": "EUR",
  "created_at": "2026-11-02T08:15:00Z",
  "job_url": "https://www.randstad.de/jobs/pflegefachkraft_berlin_c01319492/",
  "source_query": "pflege",
  "source_country": "DE",
  "scraped_at": "2026-11-04T09:00:00Z"
}

The example is shortened and the values are illustrative. Real rows hold all fields of the table above.

Pricing

Pay per result, two events:

• apify-actor-start: $0.001 once per run.
• job-result: $0.001 per job in the dataset ($1.00 per 1,000 jobs).

Filtered-out and duplicate jobs are never charged. Set a maximum total charge on the run to stop at a budget; the run then ends cleanly with the message "Stopped at the max charge limit. N jobs delivered." The pricing box on the actor page always shows the current rates.

Use with AI Agents (MCP)

Connect this actor to any MCP-compatible AI client, such as Claude Desktop, Claude.ai, Cursor, VS Code, LangChain, LlamaIndex, or custom agents.

Example prompts once connected:

"Use randstad-jobs-scraper to find nursing jobs in Germany posted in the last 3 days and list title, city and hours."

"Use randstad-jobs-scraper for DE, CH and UK with the keyword logistics and compare the salary ranges."

Configure the MCP server with the Randstad Jobs Scraper preconfigured at mcp.apify.com?tools=nanoscrape/randstad-jobs-scraper:

{
    "mcpServers": {
        "apify": {
            "command": "npx",
            "args": [
                "mcp-remote",
                "https://mcp.apify.com/?tools=nanoscrape/randstad-jobs-scraper",
                "--header",
                "Authorization: Bearer <YOUR_API_TOKEN>"
            ]
        }
    }
}

See the auto-generated API tab on the actor page for language-specific examples (cURL, JavaScript, Python, and more).

FAQ

What does one run cost? A start fee plus one result fee per job in the dataset. Use maxResults or a maximum total charge to cap it.

Why is client_name often empty? Randstad names the end client only on some jobs. When it does not, company_name is Randstad and client_name is null.

What does includeDetails add? It reads each job page and adds the full description as text and HTML, requirements, coordinates, consultant name and contact and the expiry date. Germany and the UK add more: responsibilities, street, update dates, the application link and the exact country of the job. Switzerland adds the postal code, number of openings, the task list and the offer list from the description, and the contract type.

Why can a UK search return jobs in Ireland or Austria? The Randstad UK site also lists a few jobs of other countries. Without job details the row carries the country of the site; with includeDetails the exact country is filled.

How does the location filter work? In Germany and the UK, a keyword plus a place such as Hamburg, München or London is searched on Randstad itself, so only the jobs of that place are read (a state such as Bayern works too). If Randstad does not know the place, or you give a postcode start, or you search without a keyword, the actor reads the keyword search and keeps the jobs whose city, region, location text or postal code start matches one of your places. This is also the way for Switzerland. Every row is checked against your place before it is delivered, so jobs elsewhere are never charged. The filtered search stops after 40 pages in a row without a match, so combine it with a keyword on large sites.

Can I get only new jobs? Yes, set postedWithinDays. Without a keyword, the listing is newest first and the search stops at the first page that holds only older jobs.

What if a search finds nothing? The run ends successfully with the message "0 jobs." and the reason, for example no hits or no job inside your filters.

Limits

• At most 500 pages (15,000 jobs) are read per search. Use keywords to split larger searches.
• Pages are read one after the other with a short pause per site, so very large runs take a while.
• Swiss listings carry no pay, hours, start date or application link, and Swiss job pages do not add them: the page only says things like "attractive hourly pay" in the text, and its apply button leads to the general Randstad sign-up page. apply_url is null for Swiss jobs, use job_url. Coordinates, tasks, offer and the consultant come with includeDetails.
• Swiss pages occasionally come back without their result block. The actor reads such a page up to 2 more times and only skips it when all tries fail (noted in the log).
• Only public listing data is returned. Pay is shown as Randstad publishes it and can be an estimate range.
• You are responsible for using the data lawfully, including the GDPR/DSGVO, national data-protection laws and unfair-competition rules for marketing and outreach.

Support

Questions, feature requests or a field you miss: contact@nanoscrape.com.

Built by NanoScrape. Not affiliated with Randstad.