Paste an Upwork job-search URL and get the postings back as rows: the title, a link to the posting, when it went up, the budget, whether it is fixed-price or hourly, the experience level asked for, how many proposals it has already drawn, the skill tags and the full description. Sorted by relevance or by recency, your choice.
The input is the search, not the job - you build the query on Upwork the way you always would, then hand us the URL instead of scrolling it.
One row per posting, tagged with the search it came from - so a run across a dozen skill queries still reconciles back to your input list.
Every filter Upwork gives you - keywords, category, budget, level - is already encoded in the search URL. Build the search on the site, paste the address here, and the scraper honours exactly the query you built. Nothing new to learn.
One run is a snapshot of what clients are asking for and what they will pay. Run the same searches weekly and the file becomes a series - which skills are appearing more often, and how the budgets attached to them move.
Choose relevance or most-recent, cap the rows per query, or walk a search from its oldest posting forward. Decisions made on the job rather than in a spreadsheet afterwards.
What the search results page shows about each job, flattened into a row: how it is titled and where it lives, when it went up, what the client will pay and on what basis, the experience level they asked for, how much competition the posting has already attracted, the skill tags, and the description in full.
One thing we would rather set straight than let you discover. This page lists the ten columns by name and says what each field is, and it stops there - it does not tell you how posted is formatted, whether proposals arrives as a number or a band, or which words job_type and experience_level use. Those are answers a real export gives, and we would rather publish nothing than publish a guess dressed up as a specification. Run a small free-tier job on your own search: the header row and the first few rows of that file are the definitive answer for the query you actually care about.
The export's header row, in order. Descriptions say what each field is on an Upwork posting; they deliberately claim nothing about its format - see the note underneath.
Names, not formats. The ten column names above are corroborated by two independent sources, so you can write your header mapping against them today. What is deliberately missing is everything a description could get wrong: no types, no date or currency formats, no enumerated values for job_type or experience_level, and no note on whether proposals is a count or a band. A free-tier run over your own search answers all of those in one file, and it answers them for your query rather than for somebody else's.
The search URL carries your filters. These decide how much of it you take, in what order, and how you hand the list over.
A few examples of what a table of job postings is actually good for.
Pull every posting for the skill you sell and read the budgets together. A hundred rows tell you more about the going rate than a week of scrolling, and they tell it to you in a form you can put in a deck.
Run the same searches on a schedule and count the postings each week. Which tools clients are asking for, and which have stopped appearing, is a leading indicator worth having before you invest in learning one.
Take a dozen related searches and export the lot. Job types, experience levels, budgets and skill tags in one file let you describe a market with evidence rather than impressions.
No subscription, no minimum, no recurring bill. Your first 500 rows are on us - after that, pay-as-you-go at the same flat rate as every other scraper here.
Every new account, one-time. No credit card required. Sorting, per-query limits, file upload and every export format included.
Roughly $2 per 1,000 postings. The pre-flight estimator shows the row count and credit cost before a run starts - no surprise bills, no compute units to translate.
Volume pricing, dedicated workers and an SLA for tracking many searches on a schedule. Tell us your numbers and we will quote.
LIVESCRAPER10 at checkout.An Upwork posting does not name the client, so there is no join to make here. These two cover the parts of the hiring picture this dataset cannot.
The in-house side of the same skills market - industry, size band, exact headcount and HQ for the companies hiring permanently rather than by the project.
What ranks for the same skill terms. Useful when you want to know who is competing for the attention of the clients writing these postings.
Short answer: yes for the public posting - and a job advert is business information, not somebody's profile.
Everything collected here appears on a public Upwork search results page: the title, the link, the posting date, the budget, the job type, the experience level asked for, the proposal count, the skill tags and the description the client wrote. No login, no paywall, nothing behind an account.
It is worth naming what is absent, because it is what keeps this simple. Upwork does not publish the client's name on a search listing, and there is no column for one - nor for any freelancer who has applied. The export describes advertised work, not the people on either side of it, which puts it on the business-data side of the line.
Upwork's terms restrict automated access, so this is a terms question as well as a legal one - if you have a contractual relationship with the site, check it. We run no third-party trackers on the data layer, and your exports auto-delete after 30 days.
The questions we hear most. Anything else? Talk to us - humans, not bots, write the answers.
500 one-time free rows on every new account - no expiry. After that it is $0.002 per row, pay-as-you-go - no card on file until you say so.
Livescraper's Upwork Jobs Scraper turns a job search into a dataset. You build the search on Upwork the way you normally would, submit the resulting URL - typed one per line, or uploaded as a CSV, XLSX, TXT or Parquet file - choose whether results come back by relevance or by recency, cap the rows per query if you want to, and download the file as clean CSV, Excel or JSON. One row per posting.
Each row carries what the search results page shows about a job: the title as the client wrote it, a link to the posting, when it went up, the budget, whether the work is fixed-price or hourly, the experience level the client asked for, how many proposals it has already drawn, the skill tags attached to it, and the description in full.
Freelancers and agencies pull every posting for the skill they sell and read the budgets together, which says more about the going rate than a week of scrolling. Strategy teams run the same searches on a schedule and count postings week over week, because a skill appearing more often is a leading indicator worth having early. Researchers export a dozen related searches at once to describe a category with evidence instead of impressions.
One note on what this page does and does not tell you. The ten column names are confirmed by two independent sources, so you can map your headers against them today - but we do not publish the format of the budget, the posting date or the proposal count, because those are answers a real export gives and a documentation page can only approximate. Start free and let the file tell you: your first 500 rows cost nothing and need no credit card, and after that it is $0.002 per row, flat.