Townwork Jobs Scraper — Japan Part-Time & Casual Jobs

Townwork Jobs Scraper for Japan Part-Time Jobs

Extract part-time, casual and arubaito job listings from Townwork (townwork.net) by keyword or search/area URL, with titles, employers, JPY wages, employment types, access, URLs and parse confidence. No Townwork login or API key is required.

• Build recruiter and HR-tech feeds with one flat record per listing.
• Compare wage periods and parsed JPY bounds for benchmarking and labor-market analysis.
• Create lead-generation datasets with employer, access, URL and optional area provenance.
• Search with Japanese or English keywords, paste Townwork URLs and paginate result pages.
• Add optional detail-page descriptions and filter incomplete rows with parse_confidence.

What you get

The Actor writes one dataset record for each parsed Townwork listing. Nullable fields can be empty when Townwork does not show the value.

Field	Type	Description	Availability
job_id	string or null	Townwork listing ID from the job URL	Listing records
title	string or null	Job title shown on Townwork	Listing records
company	string or null	Employer or store text	Listing records
salary_text	string or null	Raw wage label from Townwork	Listing records
salary_period	string or null	hourly, daily, weekly, monthly, annual, or per_minute	Listing records
salary_min	integer or null	Lower wage bound in JPY	Listing records
salary_max	integer or null	Upper wage bound in JPY; null when none is quoted	Listing records
currency	string	Currency returned by the wage parser: JPY	Listing records
employment_type	string or null	Employment label such as アルバイト・パート	Listing records
access	string or null	Access, station or location text from the card	Listing records
url	string or null	Townwork job-detail URL	Listing records
area	string or null	Optional input label stamped on records	When supplied
search_url	string	Normalized base search URL for the query	Listing records
page_number	integer	Result page for the listing	Listing records
scraped_at	string	UTC timestamp for the run's scrape batch	Listing records
description	string	Plain-text detail content, truncated to 6,000 characters	When enabled and returned
parse_confidence	number	Fraction of four core fields present: title, company, salary text and job URL	Listing records

Wage parsing recognizes 時給, 日給, 週給, 月給, 年俸, 年収 and 分給. A single value becomes both bounds; 以上 sets only a minimum and 以下 sets only a maximum.

Sample output

This is one real record from the dataset sample.

{
  "job_id": "cd8d9a6abeb6224f",
  "title": "gong cha(ゴンチャ)のカフェクルー",
  "company": "移転オープン◎週1日×3h〜★クルードリンク制度あり♪",
  "salary_text": "時給1,220円～1,260円",
  "salary_period": "hourly",
  "salary_min": 1220,
  "salary_max": 1260,
  "currency": "JPY",
  "employment_type": "アルバイト・パート",
  "access": "ゴンチャ 富山MAROOT店",
  "url": "https://townwork.net/viewjob/jobid_cd8d9a6abeb6224f/",
  "area": null,
  "search_url": "https://townwork.net/job_search/kw/%E3%82%AB%E3%83%95%E3%82%A7/",
  "page_number": 1,
  "scraped_at": "2026-10-28T03:21:52.116384+00:00",
  "parse_confidence": 1
}

company and access hold the text Townwork shows on the card. In this record company is a store tagline and access is a store name, so do not read them as clean employer or station fields.

Pricing

This Actor uses pay-per-event pricing. A job-result event represents one Townwork job listing. A job-description event represents full description text fetched from a detail page and is charged only when descriptions are enabled and returned.

Plan	job-result price per event	job-description price per event
Free	$0.00405	$0.005
Bronze	$0.0040095	$0.00495
Silver	$0.003969	$0.0049
Gold	$0.0039285	$0.00485
Platinum	$0.003888	$0.0048
Diamond	$0.0038475	$0.00475

Example on Free, descriptions disabled: 1,200 results = $4.86 on Free.

Free-plan totals for the event counts below:

Events on Free	job-result cost per event count on Free	job-description cost per event count on Free
1,000	$4.05	$5
10,000	$40.50	$50

Enabling descriptions can add one job-description event for each listing whose detail text is returned.

The Apify Free plan includes $5 of monthly platform usage credit. At the Free-plan job-result price, that credit corresponds to 1,234 job-result events. Non-paying accounts receive a 30-listing preview cap in this Actor, so a run can deliver no more than 30 listings on that path.

Platform usage, including Apify Proxy traffic such as residential traffic, is included in the event prices and paid by the Actor developer. This Actor does not use "Pay per event + usage", so platform usage is not separately charged to the buyer.

How to use

1. Enter a Japanese or English keyword, or paste Townwork search or area URLs in searchUrls. You can provide both; duplicate normalized URLs are removed.
2. Set pagesToFetch from 1 to 50 and use maxItems to limit total listings across queries. Townwork shows approximately 20 listings per result page.
3. Add an optional area label such as Tokyo or Osaka to stamp the same provenance label on every record. It does not filter the search.
4. Enable includeDescription for plain-text detail-page descriptions. This adds a detail-page request for each listing and can add a separate event charge.
5. Keep the default proxyConfiguration unless needed otherwise. It uses the Apify RESIDENTIAL group with country JP, required for reliable Townwork access from the cloud. Supply your own Apify account; no Townwork credentials are requested.
6. Run the Actor and inspect records as they are pushed to the dataset.
7. Repeat the run through Apify Schedules; daily is a reasonable starting point for a job-listing collection.

Example input:

{
  "keyword": "カフェ",
  "maxItems": 100,
  "pagesToFetch": 2,
  "includeDescription": false,
  "area": "Tokyo",
  "disableUsageStats": false
}

Export the dataset as JSON or CSV, or read it through the Apify dataset API.

Input parameters

Parameter	What it does	Default
keyword	Builds a Townwork keyword search from Japanese or English text	Not set
searchUrls	Accepts Townwork search or area URLs, paginated independently	Empty list
maxItems	Maximum total listings across queries; 0 means no explicit maximum in the input, subject to the preview cap	100
pagesToFetch	Result pages fetched per query	1
includeDescription	Fetches plain-text detail-page content	false
area	Adds a free-text provenance label; it does not filter	Not set
proxyConfiguration	Configures the Apify proxy used for requests	Apify RESIDENTIAL, country JP
disableUsageStats	Opts out of the usage-statistics record	false

Limits and edge cases

Condition	What you see	What to do
Neither keyword nor a usable searchUrls value is provided	The run logs No keyword and no searchUrls provided and writes {"error":"no_queries"} to OUTPUT	Provide a keyword or Townwork search/area URL
A supplied URL is not accepted as a Townwork URL	It is ignored; if no query remains, the run follows the no-query path	Use a URL from townwork.net
pagesToFetch is outside the input range	Only values from 1 through 50 are accepted; the scraper also applies a 50-page safety ceiling	Use a value from 1 through 50
A non-paying run reaches the preview cap	The Actor limits the run to 30 listing records even if maxItems is higher or 0	Use the preview or a paid Apify plan
A page has no parsed job cards	The query logs 0 cards and stops; earlier records remain in the dataset	Check the URL, keyword and page depth, then run again
Townwork returns a session interstitial or navigation is blocked	The Actor rotates the residential session and retries the page, up to four rotations per query; then the query stops	Keep the default Japan residential configuration and retry
The browser tier is unavailable	The run logs the browser error and ends the current query; records already pushed remain available	Retry later
A detail page has no usable description text	The listing can be returned without description; no description event is charged for missing text	Keep descriptions disabled when card fields are sufficient
A result page repeats seen job IDs	Duplicate records are skipped; if all IDs on a later page were seen, that query stops	Reduce page depth or use a distinct query
Townwork does not expose a wage bound or recognized period	salary_text preserves the raw label; parsed salary fields can be null, and salary_max is null for an 以上 wage without an upper bound	Use salary_text with the parsed fields and parse_confidence
Legal use of the public listing data	See the legality answer in the FAQ	Review Townwork's terms and applicable law

FAQ

How much does it cost to scrape Townwork jobs?

See the Pricing table. Without descriptions on Free, 1,200 results = $4.86 on Free.

Why did my run return no jobs?

The Actor needs a keyword or accepted Townwork search URL. A query can also stop when a page has no parsed cards or after its block-retry limit. Check the input, use a Townwork URL copied from the site, and keep the default Japan residential proxy configuration.

Why is the Actor not working?

Townwork listing pages are rendered in a browser session. Blocked pages are retried with rotating residential sessions, but a query can stop after four rotations, and the current query ends if the browser tier is unavailable. Review the run log and retry with the default proxy configuration.

Can I collect full job descriptions?

Yes. Set includeDescription to true. Usable plain-text content is returned up to 6,000 characters, and one additional job-description event can be charged per returned description.

Is scraping Townwork data legal or allowed?

The Actor reads public data from publicly visible pages. Check Townwork's terms and applicable law (the law that applies to you) for your use case before running it.

Related actors

• Arbeitsagentur Jobs Scraper — German Jobs API | PPE — related job-listing scraper for German jobs.
• Google Jobs Scraper | Postings + Multi-Board Apply Links — related scraper for Google Jobs postings and multi-board apply links.
• Jobstreet Jobs Scraper — SEA: MY, SG, PH, ID, TH, HK — related job-listing scraper for Southeast Asian markets.

Usage statistics

This Actor creates a small, content-free summary at the end of each run, used only to monitor reliability and improve this Actor. No input values, credentials or scraped data are sent.

A copy of the exact record for your run is saved as USAGE_STATS in your own Apify key-value store.

To opt out, set disableUsageStats to true in the input. Nothing is sent then.

Full list of collected fields