Scrape Upwork Job and Freelancer Pages Without Managing Browsers
Upwork job boards are built to resist automated pagination: they combine JavaScript rendering, aggressive bot checks, and IP-based throttling.
A managed scraping API lets you send a single HTTP request with an Upwork URL and receive rendered HTML or parsed JSON back, so your team can focus on the data rather than on proxy pools and headless browser fleets.
What You Can Collect from Upwork
Upwork exposes three broad datasets that are useful for talent-market and gig-economy research:
- Job listings — job title, client information, budget, hourly rate range, required skills, and posting date. Recruiters and market researchers use these to track which skills are in demand and how budgets move over time.
- Freelancer profiles — skills, hourly rate, total earnings, success score, and certification badges. Sourcing teams use these to build shortlists and compare candidates on measurable signals.
- Public project milestones — milestone data, feedback scores, and contract close-out signals per public project. Analysts use these to see how engagements are structured and how often they reach completion.
A single request can return rendered HTML ready to parse, or structured JSON if you supply extraction rules. The API also supports screenshots for visual verification, Markdown output for LLM pipelines, and JavaScript scenarios that click, scroll, or wait for dynamic content before capture.
How a Request Works
The workflow has four steps. First, create an account and obtain an API key; any free-credit allowance or trial terms should be confirmed on the current signup page before you rely on them. Second, install an SDK or use cURL. SDKs are published for Python, Node.js, Go, PHP, Ruby, and Java, and plain cURL works too, so each Upwork scrape becomes a single function call. Third, send a GET request with your API key and an Upwork URL on upwork.com. JavaScript rendering and residential proxy rotation are handled for you, and you receive rendered HTML back. Fourth, add advanced parameters when needed: render_js enables JavaScript rendering, stealth_proxy helps with CAPTCHA-heavy pages, country_code sets geo-targeting, and ai_extract_rules accepts a natural-language extraction spec. If you want structured output, pass extract_rules with CSS selectors mapped to field names, or use ai_extract_rules with a natural-language description. The response comes back as parsed JSON.
Choosing an Extraction Method
Use this checklist to pick the right output mode for your workflow:
- Rendered HTML — best when you already have a parser and want full control over selectors. You receive the page after JavaScript execution.
- `extract_rules` with CSS selectors — best when the page structure is stable and you want deterministic field mapping. You map selectors to field names and receive parsed JSON.
- `ai_extract_rules` with a natural-language spec — best when selectors change often or when you want to describe fields in plain language. No selectors are required.
- Markdown output — best when the next step is an LLM or a retrieval pipeline rather than a database.
- Screenshots — best for visual verification or archival, not for structured parsing.
A practical rule: start with extract_rules for stable job-listing pages, and switch to ai_extract_rules for freelancer profiles or project pages whose markup changes more often.
Field and Control Map
The table below maps request options to their effect and credit cost. Credit costs are per successful request and depend on which options are enabled.
| Option | What it does | Credit cost |
|---|---|---|
| Base request | Fetches the page | Starts at 1 credit |
render_js | Executes page JavaScript | 5 credits |
premium_proxy without JS | Routes through premium proxies | 10 credits |
premium_proxy with JS | Routes through premium proxies with JS rendering | 25 credits |
stealth_proxy | Higher-stealth proxy for CAPTCHA-heavy pages | 75 credits |
country_code | Geo-targets the request | Included in plan |
extract_rules | CSS/XPath extraction to JSON | Included in plan |
ai_extract_rules | Natural-language extraction to JSON | Included in plan |
| Screenshots | Full-page capture | Included in plan |
API credits are valid for one month and do not roll over. Concurrency limits are set per plan, not per target site.
Worked Example: A Job-Listing Request
Suppose you want to collect current contract postings for a specific skill. A minimal request sends an Upwork job-listing URL with render_js enabled so the page's JavaScript executes. The response is rendered HTML. If you add extract_rules mapping selectors for job title, client info, budget, hourly rate range, required skills, and posting date to field names, the response becomes parsed JSON with those fields.
If the listing page changes markup frequently, replace extract_rules with ai_extract_rules and a short natural-language description of the same fields. The trade-off is less deterministic control over selectors in exchange for resilience to layout changes. For pages that trigger CAPTCHA challenges, add stealth_proxy, which costs 75 credits per successful request.
Concurrency Limits vs. Per-Target Rate Limits
These are two different constraints and are often confused:
- Concurrency limits are set by your plan. They cap how many requests can be in flight at the same time. The Hobby plan allows 25 concurrent requests; higher plans allow more.
- Per-target rate limits are not imposed by the API. If Upwork throttles aggressive request patterns, space out requests or use multiple sessions.
If you see throttling from the target site, lowering concurrency or adding delays between requests is usually more effective than upgrading your plan.
Plans and What They Include
Standard plans include JavaScript rendering, rotating and premium proxies, geo-targeting, and screenshots and extraction rules. Priority email support, a dedicated account manager, and team management are included starting with the Business plan. The plan tiers differ mainly in monthly API credits and concurrent requests: Hobby is $19/month with 75,000 credits and 25 concurrent requests; Freelance is $49/month with 250,000 credits and 50 concurrent requests; Startup is $99/month with 1,000,000 credits and 100 concurrent requests; Business is $249/month with 3,000,000 credits and 200 concurrent requests; Business + is $599/month with 8,000,000 credits and 400 concurrent requests; Enterprise 14 is $999/month with 14,000,000 credits and 500 concurrent requests; Enterprise 24 is $1,599/month with 24,000,000 credits and 600 concurrent requests; Enterprise 41 is $2,399/month with 41,000,000 credits and 700 concurrent requests; Enterprise 69 is $3,599/month with 69,000,000 credits and 800 concurrent requests; Enterprise 120 is $5,799/month with 120,000,000 credits and 900 concurrent requests. All prices are exclusive of VAT. Any free-credit offer or trial terms should be checked on the current signup page before purchase. For volumes beyond the listed plans, contact sales.
Reliability and Compliance Notes
The platform is designed to scale across large numbers of headless browsers and rotating proxies; specific throughput or latency figures should be confirmed against current service documentation before you depend on them. Proxy rotation reduces IP-based blocking, and headless browser rendering mimics real user browsing behavior to lower the risk of blocks. On data handling, review the current privacy and data-processing documentation to determine how personal data is treated. Public-facing pages can usually be scraped, but you should review the target site's terms of service before deploying and avoid post-login content. Legal compliance is the responsibility of the person or team running the scrape.
| Plan | Monthly price | API credits | Concurrent requests | Priority support | Account manager | Team management |
|---|---|---|---|---|---|---|
| Hobby | $19 | 75,000 | 25 | Not included | Not included | Not included |
| Freelance | $49 | 250,000 | 50 | Not included | Not included | Not included |
| Startup | $99 | 1,000,000 | 100 | Not included | Not included | Not included |
| Business | $249 | 3,000,000 | 200 | Included | Included | Included |
| Business + | $599 | 8,000,000 | 400 | Included | Included | Included |
| Enterprise 14 | $999 | 14,000,000 | 500 | Included | Included | Included |
| Enterprise 24 | $1,599 | 24,000,000 | 600 | Included | Included | Included |
| Enterprise 41 | $2,399 | 41,000,000 | 700 | Included | Included | Included |
| Enterprise 69 | $3,599 | 69,000,000 | 800 | Included | Included | Included |
| Enterprise 120 | $5,799 | 120,000,000 | 900 | Included | Included | Included |
Prices are exclusive of VAT. Credits are valid for one month and do not roll over. Plan details reflect the captured page and may change; check current pricing before purchasing.
Frequently asked questions
Do I need an Upwork account to scrape Upwork?
No. You do not need an Upwork account. You sign up for an API key with the scraping service, which includes 1,000 free credits and does not require a card, then send the Upwork URL. The service handles the rest.
Is scraping Upwork legal?
Public-facing pages can usually be scraped, but you should review Upwork's terms of service before deploying and avoid scraping content behind a login. Legal compliance is your responsibility as the operator.
What is the rate limit for scraping Upwork?
The API enforces concurrency limits per plan, not per-target rate limits. If Upwork throttles aggressive request patterns, space out requests or use multiple sessions.
How do I get structured JSON from Upwork?
Pass extract_rules with CSS selectors mapped to field names, or use ai_extract_rules with a natural-language description of the fields you want. The response comes back as parsed JSON.
Which programming languages can I use to scrape Upwork?
Any language that can send an HTTP request. SDKs are published for Python, Node.js, Go, PHP, Ruby, and Java, and plain cURL works too. Pick whatever your stack already uses.
How much does it cost to scrape Upwork?
There are 1,000 free credits on signup with no card required. Paid plans start at $19/month for higher volume and concurrency. The credit cost per request depends on options: 5 credits for JavaScript rendering, 25 for premium_proxy with JS, and 75 for stealth_proxy.
What is the difference between extract_rules and ai_extract_rules?
extract_rules uses CSS or XPath selectors that you map to field names, giving you deterministic control over what is extracted. ai_extract_rules accepts a natural-language description of the fields you want and does not require selectors, which is more resilient to markup changes but less deterministic.
Do API credits roll over to the next month?
No. API credits are valid for one month and do not roll over. Each request starts at 1 credit, and enabling options such as JavaScript rendering, premium proxies, or stealth proxies increases the credit cost per successful request.