Upwork Jobs Scraper: New Jobs in 2 Minutes, Client Data
Pricing
from $0.50 / 1,000 jobs
Upwork Jobs Scraper: New Jobs in 2 Minutes, Client Data
Upwork Jobs Scraper for real-time job alerts: new jobs 2 to 3 minutes after posting, with client country, spend, rating and exact proposals count. Keywords or search URLs, filters, new-jobs-only monitoring. No account.
Pricing
from $0.50 / 1,000 jobs
Rating
0.0
(0)
Developer
Alexis Morain
Maintained by CommunityActor stats
0
Bookmarked
1
Total users
1
Monthly active users
4 days ago
Last modified
Categories
Features
Share
What does Upwork Jobs Scraper do?
Upwork Jobs Scraper reads the job search of Upwork and gives you new jobs 2 to 3 minutes after they are posted, each with the client's country, total spend, rating, hires and the exact number of proposals already received. Give it keywords or paste an Upwork search URL, set filters, press Start. No Upwork account, no browser, no cookies to paste.
It works as an Upwork jobs API: plain HTTP requests to the guest search service that the Upwork search page itself calls. Turn on New jobs only, schedule the run every few minutes, and you have real-time Upwork job alerts that bill only the jobs you have not seen yet.
How fast do new jobs show up?
Upwork publishes a job, then its search index catches up. We measured the delay between a job's publishTime and the first time a poll saw it: median 140 seconds, 90% under 175 seconds, slowest 248 seconds (16 polls at 20 second intervals, 19 new jobs, 12 November 2026). A run that reads one page of the newest jobs takes 3 to 5 seconds and costs about $0.0005 to run, so polling every 2 to 5 minutes is cheap.
Past the 5,050 results wall of a single search
One Upwork search never lets you read past offset 5,000: the server answers Paging offset greater than 5000. A scraper that pages through one search therefore stops at 5,050 jobs, whatever the page says it holds. This actor cuts a big search into slices that do not overlap (the 12 Upwork categories, then job type, experience level and client history if a slice is still too big) and reads them all.
Measured on the Apify platform on 12 November 2026:
| Search | Announced by Upwork | One search (the most it lets you page through) | Upwork Jobs Scraper | Requests | Run time |
|---|---|---|---|---|---|
| "video" | 10,346 | 5,050 | 10,346, no duplicates | 203 | 144 s |
| Same, run restarted by the platform in the middle | 10,346 | 5,050 | 10,345 distinct jobs (see the FAQ) | n/a | 198 s |
The 12 categories add up exactly to the whole market (94,196 open jobs that day), so a category slice never loses a job. Slicing by experience level can leave out the roughly 0.5% of jobs that have no experience level.
Why use Upwork Jobs Scraper?
- Freelancers and agencies. Be among the first proposals on jobs that match your skills. Filter by budget, client country, hires and spend, and skip clients who never paid anyone.
- Lead generation and sales. Every Upwork job is a company with a need and a budget. Find who is hiring for shopify developer, HubSpot, SEO audit or n8n, and how much they spend.
- Market research. Count jobs per skill, compare fixed budgets and hourly rates by country, watch demand for a technology week after week.
- Job boards and newsletters. Feed a curated Upwork digest from a dataset, a Google Sheet, Slack or a webhook.
- Pipelines. Rows go to the dataset in JSON, CSV or Excel, to Google Sheets, Slack, Zapier, Make or n8n, and a webhook tells you when a run ends. Apify adds scheduling, an API for every run and monitoring.
What data can Upwork Jobs Scraper extract?
Every row is one job.
| Field | Example | Description |
|---|---|---|
ciphertext, id, url | ~022120995318862614544 | Stable job key (deduplicate on it), numeric id and link |
title, description | Scraping specialist who... | Title and the complete description |
jobType | fixed | hourly or fixed |
budget, budgetCurrency | 150, USD | Fixed-price budget |
hourlyRateMin, hourlyRateMax | 20, 40 | Hourly range |
experienceLevel | expert | entry, intermediate or expert |
duration, durationWeeks, workload | 1 to 3 months, 9, part_time | Expected length and hours per week |
skills | ["Python", "Web Crawling"] | Skills the client tagged |
publishedAt, createdAt | 2026-11-12T21:42:38.710Z | When the job went public and when it was started (UTC) |
search | python | The keyword or URL that found it |
With Read the job details (or any client filter), each row also has:
| Field | Example | Description |
|---|---|---|
applicants | 12 | Exact number of proposals so far, not a range |
interviewing, invitesSent, unansweredInvites, hires, positionsToHire | 1, 2, 0, 0, 1 | What the client did with the proposals |
lastClientActivityAt | 2026-11-12T22:10:00Z | Last time the client was active on the job |
category, subcategory, occupation | Web, Mobile & Software Dev, Scripts & Utilities | Where Upwork files the job |
clientCountry, clientCity, clientTimezone | United States, Austin | Where the client is |
clientTotalSpent | 7108.87 | Total the client has spent on Upwork, USD |
clientRating, clientReviews | 4.99, 14 | Feedback score and number of reviews |
clientHires, clientJobsWithHires, clientActiveContracts, clientHoursBilled | 40, 23, 15, 40.5 | Client history |
clientPaymentVerified, clientMemberSince | true, 2024-05-01... | Payment method and age of the account |
clientIndustry, clientCompanySize | Software, 10 | Company profile when the client filled it in |
minJobSuccessScore, minHoursBilled, englishLevel, preferredCountries, requiresPortfolio, risingTalentOnly | 90 | What the client asks of freelancers |
detailsFetched | true | false when the details could not be read |
Empty fields. Upwork serves what it serves to a visitor, and the actor does not guess. clientTotalSpent is null for a client who never paid anyone (about 4 clients in 10), clientRating is null for a client with no feedback yet, and clientIndustry and clientCompanySize are often null. About 1 job in 10 is private: Upwork asks for a login to show its details. Those rows keep every search field (title, description, budget, skills, dates) and have detailsFetched: false and empty client fields, and they are billed as a plain job. With a client filter on (maximum proposals, minimum spend, minimum rating, payment verified), a private job cannot be checked, so it is dropped, not delivered and not billed; the end-of-run message says how many private jobs were dropped and how many were delivered. The client's hire rate, the number of jobs they posted and their average hourly rate are not served to visitors and are not in the data.
How to scrape Upwork jobs
- Click Try for free.
- Type one or more Keywords (one per line), or paste Upwork search URLs.
- Pick filters if you want them: job type, experience level, budget, hourly rate, client country, client history, categories.
- Tick Read the job details to get the client panel and the exact proposals count.
- Set Max jobs and press Start. A 20-job run with details finishes in about 5 seconds.
- Download the dataset as JSON, CSV or Excel, or send it on with an integration.
How to get Upwork job alerts every few minutes
- Fill in your keywords and filters as above.
- Turn on New jobs only (monitoring) and set Only jobs published in the last N minutes to
30, so the first run does not return a month of jobs. - Add a Webhook URL (a Slack or Discord incoming webhook, a Zapier or Make URL).
- In Apify, open Schedules and run the task every 2 to 5 minutes. Each run reads one page, skips the jobs it has seen and bills only the new ones.
- In your tool, read the dataset of the run that the webhook names, or add an Apify integration that sends new rows to Slack, e-mail or a Google Sheet.
For a "be first" alert, add Maximum proposals 5 and a Minimum client spend: a job that fails the filter is not delivered and not billed.
How much will it cost to scrape Upwork?
Pay per event, in dollars per 1,000, with volume discounts on the paid Apify plans (the price drops with your plan):
| Event | Free plan | Starter | Scale | Business and up |
|---|---|---|---|---|
| Job (search fields) | $1.00 | $0.90 | $0.70 | $0.50 |
| Job with detail (client panel, exact proposals) | $1.50 | $1.30 | $1.10 | $0.80 |
| Start of a run | $0.001 per GB of memory | same | same | same |
The default memory is 512 MB, so a run starts at $0.001. A job rejected by a client filter, a job already seen in New jobs only mode and a duplicate are never billed. A private job is billed as a plain job, unless a client filter drops it (then it is not billed).
Real examples, measured on the platform:
- 20 jobs with details: about $0.03 at the free-plan price.
- 10,346 jobs of the search "video": about $10.35 at the free-plan price, $5.17 at the Business price.
- A monitoring run every 3 minutes: 480 runs a day, $0.48 a day for the starts plus the few new jobs each run finds.
Set Max jobs to cap a run, and Apify's own maximum cost per run option stops the run when your limit is reached.
Input
See the Input tab for every option. The ones that need a word:
- Keywords and Upwork search URLs can be combined. The filters on the page (job type, level, budget...) apply to keywords. A pasted URL carries its own filters.
- Fixed budget only applies to fixed-price jobs and Hourly rate only to hourly jobs, so setting one limits the search to that type. Setting both runs two searches.
- Categories takes the 12 Upwork categories. With categories and no keyword you get every new job of those categories.
- Maximum proposals, Minimum client spend, Minimum client rating and Payment verified clients only are not offered by Upwork to a visitor, so the actor reads each job's details and applies them itself. Jobs that fail are not delivered and not billed.
- Stop after this many rejected reads (
maxRejectedReads, default 2,000) is a safety net for the client filters. A job read and rejected is free for you but costs the actor a request, so a filter that nothing passes would read the whole market for nothing. The run stops, with a clear message in its status, when 2,000 jobs were rejected and they outnumber the delivered ones 20 to 1. Set0for no guard. - Max jobs
0means no limit. Above 5,050 the search is cut into slices (see above), so jobs do not come in strict date order.
{"keywords": ["web scraping", "python scraper"],"jobType": "fixed","minBudget": 500,"clientCountries": ["United States", "United Kingdom"],"maxApplicants": 5,"newOnly": true,"maxAgeMinutes": 30,"maxItems": 100}
Output
You can download the dataset in various formats such as JSON, HTML, CSV or Excel. A row with details:
{"ciphertext": "~022120995318862614544","url": "https://www.upwork.com/jobs/~022120995318862614544","title": "Scraping specialist who has beaten Akamai bot protection","jobType": "fixed","budget": 150,"experienceLevel": "expert","duration": "More than 6 months","skills": ["Data Scraping", "Python", "Web Crawling", "Automation"],"publishedAt": "2026-11-12T21:42:38.710Z","applicants": 4,"interviewing": 3,"clientCountry": "Rwanda","clientPaymentVerified": true,"clientTotalSpent": 357,"clientRating": 4.97,"clientReviews": 13,"clientHires": 21,"detailsFetched": true,"search": "python scraper"}
A private job keeps its search fields only:
{"ciphertext": "~022120867209617137486","title": "Looking for a developer","jobType": "hourly","hourlyRateMin": 20,"hourlyRateMax": 40,"publishedAt": "2026-11-12T20:11:00.000Z","applicants": null,"clientCountry": null,"detailsFetched": false}
Run it from Python
from apify_client import ApifyClientclient = ApifyClient("<YOUR_APIFY_TOKEN>")run = client.actor("alexmorain/upwork-jobs-scraper").call(run_input={"keywords": ["n8n automation"],"includeDetails": True,"maxApplicants": 5,"maxAgeMinutes": 60,"maxItems": 50,})for job in client.dataset(run["defaultDatasetId"]).iterate_items():print(job["title"], job["clientCountry"], job["applicants"], job["url"])
Run it from JavaScript
import { ApifyClient } from 'apify-client';const client = new ApifyClient({ token: '<YOUR_APIFY_TOKEN>' });const run = await client.actor('alexmorain/upwork-jobs-scraper').call({keywords: ['shopify developer'],jobType: 'hourly',minHourlyRate: 40,maxItems: 50,});const { items } = await client.dataset(run.defaultDatasetId).listItems();console.log(items.length, 'jobs');
Run it with cURL
curl -X POST "https://api.apify.com/v2/acts/alexmorain~upwork-jobs-scraper/run-sync-get-dataset-items?token=<YOUR_APIFY_TOKEN>" \-H "Content-Type: application/json" \-d '{"keywords": ["web scraping"], "maxAgeMinutes": 30, "maxItems": 20}'
Tips
- Cheapest alert: New jobs only,
maxAgeMinutesof 30 to 60, no details. One page per run. Add details only on the filters that need them. - Be first:
maxApplicants3 to 5 and a run every 2 minutes. Upwork lists a job about 2 to 3 minutes after the client posts it. - Search by skill or by exact phrase: put quotes around a phrase, such as
"data entry". - Several clients in one go: put several keywords. A job found by two keywords is delivered and billed once, and
searchtells you the first keyword that found it. - Big exports: set Max jobs to
0and a category, a job type or an experience level, so slices stay small and finish faster.
FAQ, disclaimers and support
Can I get Upwork job alerts?
Yes, by polling. Turn on New jobs only, add a webhook URL and schedule the run in Apify every 2 to 5 minutes. A new job appears in the search about 2 to 3 minutes after it is posted (median 140 seconds in our measure), then your next run picks it up. There is no push from Upwork and no run faster than your schedule, so the delay you get is that of the search index plus your polling interval. See "How to get Upwork job alerts every few minutes" above.
How much does it cost?
$1.00 per 1,000 jobs, or $1.50 per 1,000 jobs with the client panel and exact proposals, on the free plan, down to $0.50 and $0.80 on the biggest plans, plus $0.001 per run start. Jobs you have already received in New jobs only mode, duplicates and jobs rejected by a client filter are not billed. The table above gives the exact prices and measured examples.
Which client data do I get?
Country, city, time zone, total spent, rating and number of reviews, hires, active contracts, payment verified, member since, industry and company size when the client filled them in. The client's name and contact details are not served to visitors and are not in the data. The hire rate, jobs posted and average hourly rate need a login and are not returned.
What are the limits?
No Upwork account, so private jobs (about 1 in 10) come with the search fields only. Upwork shows a visitor jobs 2 to 3 minutes after they are posted, not instantly. Past 5,050 results the search is cut into slices, so those jobs do not come in strict date order, and slicing by experience level can miss the roughly 0.5% of jobs with no level. Upwork limits requests per IP address, so very big runs take minutes. If Upwork changes its site, a run can fail until the actor is updated.
Do I need an Upwork account?
No. The actor reads what Upwork shows a visitor. It never logs in, so it cannot be tied to an Upwork account, and it cannot read what a login would add (the client's hire rate, number of jobs posted, proposal bids).
Why are some client fields empty?
See "Empty fields" above. The actor returns null rather than a guess: a client who never spent anything has no total, and a client with no feedback has no rating.
Can I get the freelancers' profiles or their bids?
No. The actor returns jobs and the client panel only. It does not read freelancer profiles.
Why did a restarted run bill a few jobs twice?
The platform can move a run to another machine, which restarts it. The actor remembers what it already delivered and skips it, but a batch of up to 50 jobs that was written just before the restart can be written again. We measured 50 repeated rows in 10,395 after a forced restart, and the ciphertext column tells you which. Billing follows the rows.
Why is my run slower than I expect on a big search?
Upwork refuses an IP address after roughly 130 to 145 requests in a few minutes. The actor changes IP address every 90 requests and as soon as Upwork answers "too many requests", so a big run keeps going, with a few retries visible in the log.
Is it legal?
Our Actors are ethical and do not extract any private user data, such as email addresses, gender, or location. They only extract what the user has chosen to share publicly. We therefore believe that our Actors, when used for ethical purposes by Apify users, are safe. However, you should be aware that your results could contain personal data. Personal data is protected by the GDPR in the European Union and by other regulations around the world. You should not scrape personal data unless you have a legitimate reason to do so. If you're unsure whether your reason is legitimate, consult your lawyers.
This actor returns job posts and the public profile of the hiring company (country, spend, rating), never a freelancer. Upwork's terms of use restrict automated access to the site and its robots.txt asks bots to stay out of its search interface. You are responsible for using the data in line with those terms and with the law where you are.
Support
Something missing, a field that came back empty on a job that has it, a feature you need? Open an issue on the Issues tab of the Actor page; it is read and answered. Everything on this page is also available through the API tab.
Guide and use cases
Use cases for outbound and automation, the output fields, pricing and common questions are on the Upwork Jobs Scraper page of my site. The Job boards family page puts it side by side with the other scrapers of the same family.