SEEK, JobStreet & JobsDB Jobs Scraper

Scrape jobs from SEEK (Australia, New Zealand), JobStreet (Malaysia, Singapore, Philippines, Indonesia) and JobsDB (Hong Kong, Thailand): title, company, location, the full job description, pay parsed into numbers with the right currency, how many people have applied, the employer's screening questions, listing and expiry dates, and the employer's industry, size, rating and perks.

One SEEK search shows at most 550 jobs. This Actor splits bigger searches by work type and category on its own, so a search for every nurse job in Australia returns thousands of jobs, not 550. Search several keywords and cities in one run, or paste links straight from SEEK, JobStreet or JobsDB.

Try it now: keep the example input (software engineer in Sydney) and press Start.

What does SEEK, JobStreet & JobsDB Jobs Scraper do?

It searches SEEK's job boards by keywords and location, in any of the eight countries SEEK runs, and returns one row per job. It reads the same data SEEK's own websites load, so there is no login, no API key and no browser to wait for.

• All eight SEEK countries in one Actor: a SEEK scraper, a JobStreet scraper and a JobsDB scraper in one.
• No 550-job ceiling. Past 550 matches the Actor runs narrower searches (by work type, then category, then subcategory) and merges them without duplicates.
• Several searches in one run. Give more keywords and more locations; every keyword is searched in every place, and a job found twice is delivered once.
• Paste links. Search result pages keep the filters in their address, and job ads are fetched directly, from any of the eight sites.
• Pay you can sort and filter. $100,000 – $125,000 per year, RM 6,000 – RM 9,000 per month and Rp 8.000.000 per month become salaryMin, salaryMax, salaryCurrency and salaryPeriod.
• The right currency. A $ on JobStreet Philippines or JobsDB Thailand is usually US dollars, and on JobStreet Malaysia it can be Singapore dollars. The Actor reads the currency SEEK recorded for each job instead of guessing from the symbol.
• How much competition there is. The number of applicants, SEEK's application volume label, the share who sent a cover letter, and the questions the employer asks every applicant.
• Company details from SEEK company profiles: industry, size, employee rating, perks and benefits, a description and website.
• Keep or drop jobs by the words in them, or keep only ads that publish an email or phone number (about a third of Sydney software ads do, but only about 1 in 40 in Malaysia and the Philippines). Jobs left out are not charged.
• Change tracking for monitoring: each run marks jobs NEW, UPDATED (with the fields that changed), UNCHANGED or REAPPEARED, and reports jobs that have closed as EXPIRED, confirmed with SEEK rather than guessed from a job's absence. Only the changes are output and charged.

Because it runs on Apify, you can start it from the Console or the API, schedule it, export the results as JSON, CSV or Excel, and connect them to your own tools.

Which SEEK, JobStreet and JobsDB sites does it cover?

Country	Site	Address	Location examples
Australia	SEEK	www.seek.com.au	Sydney, Melbourne VIC, Western Australia, Parramatta
New Zealand	SEEK	www.seek.co.nz	Auckland, All Wellington, Christchurch
Malaysia	JobStreet	my.jobstreet.com	Kuala Lumpur, Selangor, Penang
Singapore	JobStreet	sg.jobstreet.com	Central Region, or leave empty
Philippines	JobStreet	ph.jobstreet.com	Metro Manila, Cebu, Makati City
Indonesia	JobStreet	id.jobstreet.com	Jakarta, Surabaya, Bandung
Hong Kong	JobsDB	hk.jobsdb.com	Kowloon, Central, or leave empty
Thailand	JobsDB	th.jobsdb.com	Bangkok, Chon Buri

What data can you extract from SEEK, JobStreet and JobsDB?

One row per job:

Group	Fields
Job	title, workType, workArrangement, classification, subclassification, roleId, teaser, bulletPoints, description, descriptionHtml, tags
Pay	salaryText, salaryMin, salaryMax, salaryCurrency, salaryPeriod
Competition	applicantCount, applicantVolume, applicantsWithResumePercent, applicantsWithCoverLetterPercent, screeningQuestions
Location	location, locationDetail (often with a postcode), country, site
Dates	listedAt, expiresAt, isExpired, scrapedAt
Company	company, advertiser, advertiserId, companyId, isPrivateAdvertiser, advertiserVerified, companyLogo, companyProfileUrl, companyIndustry, companySize, companyRating, companyReviewCount, companyWebsite, companyDescription, companyPerks
Listing	jobId, url, isFeatured, isPromoted, appliesOnExternalSite, searchQuery, searchLocation
Contacts (optional)	contactEmails, contactPhones
Changes (with change tracking)	changeType, changedFields, firstSeenAt, lastSeenAt, repostOfJobId

advertiser differs from company when a recruitment agency posts the job for an employer. A confidential employer shows as isPrivateAdvertiser: true with no company name, rather than as a company called "Private Advertiser". advertiserId and companyId stay the same across runs, so you can join runs on them. tags holds SEEK's own labels, such as Urgently hiring, Be an early applicant and Expiring soon.

How to scrape SEEK, JobStreet and JobsDB jobs

1. Enter keywords, for example registered nurse or accountant. Leave them empty to list every job in a place.
2. Pick the country, then a location in it: a city (Sydney, Auckland, Kuala Lumpur), a region (Metro Manila, Kowloon) or a state (Queensland, Western Australia, or just QLD). Type remote for remote jobs anywhere in the country, or leave it empty for the whole country.
3. Set Max results. It is also your cost ceiling.
4. Optionally add more keywords and more locations, or paste links instead.
5. Optionally narrow it down under Filters: work type, on-site / hybrid / remote, job category, salary range, distance, how recently the job was listed, and words the title must or must not include.
6. Press Start, then open the Output tab or export the dataset.

Several searches in one run

Put data analyst in Keywords, data engineer in More keywords, Sydney in Location and Melbourne in More locations, and the run makes four searches. Max results is shared evenly between them, and whatever one search can't fill goes to the others. Each row's searchQuery and searchLocation say which search found it.

Paste SEEK, JobStreet or JobsDB links

Set up a search on SEEK's own site, copy the address from your browser and paste it into SEEK, JobStreet or JobsDB links. The Actor reads the keywords, place, category, work type, work arrangement, salary, date and sort order from the address, for example:

• https://www.seek.com.au/software-jobs/in-All-Sydney-NSW/full-time/hybrid?daterange=7
• https://my.jobstreet.com/data-analyst-jobs/in-Kuala-Lumpur
• https://id.jobstreet.com/jobs-in-information-communication-technology/in-Indonesia

Links to single job ads (https://www.seek.com.au/job/94701279) are fetched directly, with the same fields as a searched job. That works even after an ad has closed, since SEEK keeps closed ads: those rows have isExpired: true. Company pages (.../at-this-company) aren't supported yet.

Every job, not just 550

SEEK shows at most 550 jobs for any one search. Ask for more and the Actor splits the search: first by work type (full time, part time, contract, casual), then by job category, then by subcategory, until each piece fits. Every job matching your search can be reached this way unless a single subcategory alone holds more than 550 jobs, and the run's status message tells you when that happened.

With Newest first, a search past 550 jobs keeps to the fewest recent days that hold enough jobs (the last 1, 3, 7, 14 or 30), so asking for the 2,000 newest jobs in Australia returns jobs from the last few days rather than older ones from the biggest categories. SEEK's own newest-first list still mixes in a few featured ads of any age, as its website does; isFeatured marks them.

Track changes between runs

Switch on Track changes between runs and schedule the Actor, for example daily. Each run compares what it finds with the earlier runs of the same search, and outputs, and charges for, only what changed:

changeType	Meaning
NEW	No earlier run found it. repostOfJobId is set when it repeats an earlier ad: the same advertiser, title and place under a new job ID.
UPDATED	The ad changed. changedFields says what, for example ["salaryText"] or ["title", "description"].
REAPPEARED	It had closed, and is open again.
EXPIRED	An earlier run found it, and it has closed since. expiresAt is when it closed.
UNCHANGED	Found again, unchanged. Left out unless you turn on Also output unchanged jobs.

Closures are checked with SEEK, not guessed. SEEK keeps a closed ad readable, marked as expired with the time it closed, so the Actor looks up each job an earlier run found and this one didn't. A job that only dropped out of the results, because its ad was moved to another category or it sits past SEEK's 550-job limit, is not reported as closed. The check runs when the run covers the whole search, so set Max results above the number of jobs the search holds.

The first run outputs every job as NEW. Each different search (its keywords, locations, links and filters) keeps its own history automatically; give it a History name to keep one history while you adjust the search. Changes in applicant numbers and tags such as Expiring soon don't count, and a changed description needs Full descriptions and company details on. On an EXPIRED row, listedAt to expiresAt is how long the ad was up.

For a simpler feed, Only jobs you haven't seen before outputs each job once and never again. When nothing has changed, a run still succeeds and its status says so.

Run it from the API

curl -X POST "https://api.apify.com/v2/acts/akhil_rajesh~seek-jobstreet-jobsdb-scraper/run-sync-get-dataset-items" \
  -H "Authorization: Bearer YOUR_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "searchQuery": "data analyst",
    "extraKeywords": ["data engineer"],
    "country": "SG",
    "maxResults": 200,
    "postedWithin": "7"
  }'

The call waits for the run and returns the jobs directly. It waits up to 5 minutes, so for very large runs start the run instead and read its dataset when it finishes.

Run it from Python

from apify_client import ApifyClient  # pip install apify-client

client = ApifyClient("YOUR_API_TOKEN")
run = client.actor("akhil_rajesh/seek-jobstreet-jobsdb-scraper").call(run_input={
    "searchQuery": "registered nurse",
    "country": "AU",
    "location": "Queensland",
    "maxResults": 1000,
})
for job in client.dataset(run.default_dataset_id).iterate_items():
    print(job["title"], job["company"], job["salaryMin"], job["salaryMax"], job["salaryCurrency"], job["applicantCount"])

Use it from AI agents

Through the Apify MCP server, AI assistants such as Claude, ChatGPT and Cursor can run this Actor as a tool. Add https://mcp.apify.com?tools=akhil_rajesh/seek-jobstreet-jobsdb-scraper as an MCP server, then ask for jobs in plain words, for example "find hybrid data analyst jobs in Kuala Lumpur posted this week and compare the pay".

How much does it cost to scrape SEEK?

This Actor uses pay-per-event pricing, and platform usage is included, so there's nothing else to pay. You pay for each job returned, plus $0.001 each time a run starts. The price per job depends on your Apify plan:

Your plan	Per 1,000 jobs
Free	$1.00
Starter	$0.90
Scale	$0.75
Business and above	$0.60

Run	Jobs	On Free	On Business
The example input	100	$0.101	$0.061
1,000 nurse jobs in Queensland	1,000	$1.001	$0.601

Jobs your filters leave out, unchanged jobs when change tracking is on, and jobs an earlier run already delivered when Only jobs you haven't seen before is on, are not charged. To cap spending, set Max results or a maximum cost per run in the run options; the Actor stops cleanly at the limit.

Input

Field	Type	Default	What it does
searchQuery	string	—	Keywords: job title, skill or company
country	string	AU	AU, NZ, MY, SG, PH, ID, HK or TH
location	string	—	City, suburb, region or state in that country. remote means remote jobs anywhere
maxResults	integer	100	Stop after this many jobs across all searches (up to 20,000). Also your cost ceiling
postedWithin	string	any time	1 (today), 3, 7, 14 or 31 days
sortBy	string	relevance	relevance or newest
extraKeywords	array	—	More keywords; each is searched in every location
extraLocations	array	—	More places in the same country
startUrls	array	—	SEEK, JobStreet or JobsDB search pages or job ads. When given, the keyword and location fields are ignored
workTypes	array	all	fullTime, partTime, contract, casual
workArrangements	array	all	onSite, hybrid, remote
classifications	array	all	SEEK category IDs, chosen by name in the form
minSalary	integer	—	Only jobs paying at least this much
maxSalary	integer	—	Only jobs paying at most this much
salaryType	string	automatic	annual, monthly or hourly; automatic means yearly in Australia and New Zealand, monthly elsewhere
radiusKm	integer	SEEK's 50	Distance around a town or suburb, 1 to 100 km
includeKeywords	array	—	Keep only jobs whose title has one of these words
excludeKeywords	array	—	Leave out jobs whose title has any of these words
keywordsInDescription	boolean	false	Check the two word lists against the description too
requireContact	string	none	any, email or phone: only ads that publish contact details
includeDetails	boolean	true	Full description, expiry date, pay currency, applicants, screening questions and company profile
includeContacts	boolean	false	Emails and phone numbers published in the ad
trackChanges	boolean	false	Mark jobs NEW, UPDATED, UNCHANGED, REAPPEARED or EXPIRED against earlier runs, and output only the changes
includeUnchanged	boolean	false	With tracking, also output unchanged jobs
includeClosed	boolean	true	With tracking, report jobs that have closed since, checked with SEEK
skipReposts	boolean	false	With tracking, leave out new jobs that repeat an earlier ad
onlyNewJobs	boolean	false	Skip jobs an earlier run already delivered
historyName	string	automatic	The key-value store that remembers earlier runs; empty gives each search its own
proxy	object	none	Optional; SEEK answers ordinary requests

Sample output

A job from the example search, software engineer in Sydney (long text is shortened here):

{
  "jobId": "94701279",
  "url": "https://www.seek.com.au/job/94701279",
  "title": "She Lives Tech: Software Development Graduate Program",
  "company": "FDM Group",
  "advertiser": "FDM Group Australia Pty Ltd.",
  "location": "Sydney NSW",
  "locationDetail": "All Sydney NSW",
  "country": "AU",
  "site": "SEEK Australia",
  "workType": "Full time",
  "workArrangement": "On-site",
  "classification": "Information & Communication Technology",
  "subclassification": "Developers/Programmers",
  "salaryText": "$60,000 – $75,000 per year",
  "salaryMin": 60000,
  "salaryMax": 75000,
  "salaryCurrency": "AUD",
  "salaryPeriod": "YEAR",
  "listedAt": "2026-09-17T07:44:03Z",
  "expiresAt": "2026-10-17T12:59:59Z",
  "isExpired": false,
  "teaser": "Curious about AI and technology? The FDM Graduate Program gives graduates hands-on training, mentorship and real-world experience.",
  "bulletPoints": null,
  "description": "Build real software. Build a career in tech one step at a time.\n\nCurious about AI and technology? The FDM Graduate Program gives women hands on training, mentorship and real world experience. No IT degree? No problem! ...",
  "descriptionHtml": "<p><strong>Build real software. Build a career in tech one step at a time.</strong></p> ...",
  "tags": null,
  "isFeatured": false,
  "isPromoted": false,
  "appliesOnExternalSite": false,
  "applicantCount": 147,
  "applicantVolume": "Medium application volume",
  "applicantsWithResumePercent": 100,
  "applicantsWithCoverLetterPercent": 44,
  "screeningQuestions": [
    "Which of the following statements best describes your right to work in Australia?",
    "What are your pronouns? (How do you wish to be addressed)",
    "Are you geographically flexible to relocate within Australia (either Sydney or Melbourne) for this role?",
    "Have you completed your first Bachelor's before or after 2020?"
  ],
  "advertiserVerified": true,
  "isPrivateAdvertiser": false,
  "advertiserId": "36988816",
  "companyId": "158457779534327",
  "companyLogo": "https://bx-branding-gateway.cloud.seek.com.au/edac112c-b3ae-40b7-8b9c-cf5c7ab1ad6a.1/jdpLogo",
  "companyProfileUrl": "https://au.seek.com/companies/fdm-group-158457779534327",
  "companyIndustry": "Information Technology Services",
  "companySize": "101-1,000 employees",
  "companyRating": null,
  "companyReviewCount": null,
  "companyWebsite": null,
  "companyDescription": "FDM Group equips graduates with real-world tech and business skills in just 3 months. ...",
  "companyPerks": [
    "Support and Development",
    "Employee Assistance Program (EAP)",
    "Online Learning and Development",
    "Employee Recognition Awards"
  ],
  "roleId": "software-developer",
  "searchQuery": "software engineer",
  "searchLocation": "Sydney",
  "scrapedAt": "2026-09-24T01:45:12Z"
}

Fields without a value are null. Of the example search's 100 jobs (September 2026), 65 showed applicant numbers, 48 had screening questions, 41 came from employers with a SEEK company profile and 39 had pay that parsed into numbers.

Good to know

• About 44% of listings show a salary. Across 11,500 listings in the eight countries (September 2026), the Actor turned 93% of those into numbers; the rest are text such as Competitive salary and stay in salaryText. Coverage is highest in Asia, where most employers publish a monthly range.
• salaryPeriod is only set when the listing says (per hour, per month, p.a., fortnightly and so on). $70,000 alone keeps its number but no period, and a period that belongs to something else, as in + monthly bonus, is not read as the salary's.
• Applicant numbers appear once an ad has been up a while and takes applications on SEEK; brand-new ads and ads that send you to the employer's own site don't have them.
• How fast it runs. 100 jobs with full details take about 10 seconds. SEEK hands out job details at about 500 a minute, and the Actor keeps to that pace, so full details for 3,000 jobs take about 6 minutes. With Full descriptions and company details off, 5,000 jobs take about 2 minutes.
• SEEK's location matching is SEEK's own. A town or suburb covers SEEK's default 50 km around it (set 1 to 100 km with Distance from the location); an area such as All Sydney NSW or a state covers exactly that area. NSW, VIC and the other Australian state abbreviations are sent as the whole state (SEEK itself reads NSW as Sydney and WA as Wagga Wagga).
• Listings change constantly: a job can close between runs. expiresAt says when each one is due to close.

Use cases

• Job market research: pay ranges by role, city and country across Australia, New Zealand and South-East Asia
• Recruitment and sales prospecting: who is hiring, for what, and whether through an agency
• Salary benchmarking: numeric pay with the right currency and period, ready for a spreadsheet
• Competition analysis: how many people apply for each role, and what employers screen for
• Job alerts and aggregators: a daily feed of new, changed and closed listings with change tracking
• Hiring-speed research: which ads close fast, from listedAt to expiresAt on the EXPIRED rows, and which employers keep reposting the same role
• Employer research: company size, industry, perks and employee ratings next to their open roles

FAQ

How many results can I get?

As many as SEEK has for your search, up to 20,000 per run. One SEEK search stops at 550 jobs; this Actor splits bigger searches into narrower ones and merges them, so a search for every job in Australia can go well past that. The only jobs out of reach are those in a single subcategory that alone holds more than 550 matches, and the status message says when that happens.

Can I scrape several keywords or cities at once?

Yes. Add them under More keywords and More locations: every keyword is searched in every place, Max results is shared evenly, and a job found by two searches is delivered once.

How do I get only new or changed jobs?

Turn on Track changes between runs and schedule the Actor. The first run outputs every job; after that, each run outputs only new, updated, reappeared and closed jobs, and you pay only for those.

How does it know a job has closed?

It asks SEEK. A closed ad stays readable on SEEK, marked as expired with the time it closed. The Actor looks up every job an earlier run found and this one didn't, and reports only those SEEK says have closed, so a job that just left the search results isn't counted as closed.

Can I paste a SEEK, JobStreet or JobsDB URL?

Yes, search result pages and single job ads, from any of the eight sites. A search link keeps its own filters; the form's filters fill in any it doesn't set.

Does it need a SEEK login or API key?

No. It reads the same public data SEEK's websites load for any visitor.

Why did my run stop with a location message?

SEEK didn't recognise the place you typed, so it would have returned nothing. The Actor first tries a simpler form (Perth, Western Australia becomes Perth WA), and the message lists SEEK's suggestions when it has any. Use a city, region or state in the country you picked, such as Brisbane, All Auckland, Selangor or Cebu. With several locations, an unknown one is skipped and named in the status while the rest still run.

Why do some jobs have no currency?

With Full descriptions and company details off, a bare $ on JobStreet or JobsDB outside Singapore and Hong Kong is left without a currency, because it is often US dollars. Turn the details on (the default) and the Actor uses the currency SEEK recorded.

Is it legal to scrape SEEK?

This Actor reads only publicly available job listings. Job ads can still contain personal data, such as a recruiter's name or phone number, which privacy laws (for example Australia's Privacy Act) protect. Contact details are left out unless you turn them on; use them only for a legitimate purpose, and ask a lawyer if you're unsure.

Other Actors by this developer

• Indeed Jobs Scraper: job listings from Indeed in Canada, the US and six more countries, with full descriptions, pay and the employer's apply link
• Wellfound Jobs Scraper: startup jobs with salary and equity ranges and an exact Y Combinator flag
• Canadian Job Market Scraper - Job Bank: Canadian job postings with salary, employer, location and NOC code
• AutoTrader.ca Scraper: Canadian vehicle listings with price, mileage, trim and dealer
• Price Scraper for Bot-Protected Stores: live price comparison across major Canadian retailers
• Tennis Scraper: live scores, results, match statistics, point-by-point and bookmaker odds for ATP, WTA, Challenger and ITF

Data source

All data comes from SEEK, JobStreet and JobsDB, public job boards run by SEEK Limited. This Actor reads publicly available listings only and stores nothing beyond the dataset you request.

Disclaimer: This is an independent tool. It is not affiliated with, endorsed by, or sponsored by SEEK Limited, JobStreet or JobsDB. All product and company names are the property of their respective owners.