Paste an OLX search URL from any country the platform runs in - olx.ro, olx.pl, olx.kz, olx.in - and get the listings back as rows: the title and description, the price with the currency it is quoted in, where it is, who is selling and whether they are a business, every image, and a link to the offer.
The input is whatever OLX search you are looking at, on whichever country's site you are looking at it.
One row per listing, each tagged with the query it came from - so a run across several countries still reconciles back to your input list.
OLX runs a lot of national sites and this reads all of them through one form. Paste a Romanian search and a Polish one on consecutive lines and they come back in the same shape.
City, region and district each get their own column, with ids alongside the names, rather than arriving as one string you have to pull apart. The combined string is still there if you want it.
A dedicated column marks whether the listing is a private sale or a trader, which is the split that decides whether a price series is measuring a consumer market or a professional one.
Each row carries the listing as OLX's results show it: the title and description, the price and its currency, the location broken into parts, the seller and their account age, every image, and a link to the offer.
The column list below is the header row of real runs rather than a specification. Read the note under it before you build against it - this output has three shapes that catch people out, and one of them will quietly corrupt a price comparison if you assume the currency from the country you searched.
Taken from the header row of real runs, identical to the column list the platform ships. The JSON and the workbook carry the same names, so no format here is narrower than another.
business - the two do not track each other reliably in either direction.yes and no.YYYY-MM-DD. A rough proxy for how established they are.url on every row we have seen - the two are a duplicate pair, so pick one and ignore the other.yes and no."; " to get the list.YYYY-MM-DD. A real date - no arithmetic against your run date required.listing_url.images.150 000 km · 2022. It varies by category and carries trailing whitespace, so trim it before parsing.Three shapes in this output catch people out. First and most costly: currency must be read per row. One search on a single national site came back with prices in more than one currency, so a price series that assumes the currency from the domain is silently mixing units. Second: images is one string joined by a semicolon and a space, not an array - split on "; ", and note that image is just its first entry. Third: url and listing_url were identical on every row we have seen, as were the combined location and the split city and region columns, so several of the twenty-eight are the same fact twice. Also worth knowing before you write a parser: details is free text that changes shape by category and carries trailing whitespace, and company_name does not reliably follow the business flag in either direction.
A per-query limit decides what a run costs and how much lands in the file. It is chosen before the job starts, alongside how you hand over the list of searches.
A few examples of how teams use classifieds data to answer a question they actually have.
OLX covers a lot of countries with one shape of output, so the same search run against several national sites gives a cross-border price picture in a single file. Convert on the currency column rather than on the domain and the comparison holds.
The business flag splits a category into a consumer market and a professional one. Those two behave differently on price and on how fast stock turns, and averaging them together hides both.
City, region and district arrive as their own columns with ids attached, so geographic rollups do not start with string surgery. Join on the normalised names and local spelling stops being a problem.
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. Per-query limits, file upload and every export format included.
Roughly $2 per 1,000 listings. 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 continuous monitoring or very large historical pulls. Tell us your numbers and we will quote.
LIVESCRAPER10 at checkout.Poland's other marketplace, and one OLX competes with directly. Running both over the same category is how you stop mistaking one platform's supply for the whole market.
The same job on a US local marketplace - title, asking price, city and state, and a link to each listing. Useful if you are comparing how a category behaves across regions.
Short answer: yes for the public listing content - and this output has more seller detail in it than most, which is worth planning for.
Listings on OLX are published to be found. The title, the description, the price, the location, the images and the seller's display name are shown to anyone who runs the search, signed in or not. Collecting publicly visible listing data for price research is long-established practice, and nothing here touches a login or a paywall.
This schema carries more about the seller than a plain price scrape does: a display name, an account id, an account creation date, a business flag, and columns for a trading name and a phone number. OLX is huge in the EU, so the GDPR applies to what you do with those downstream, not to the fact that they were published. Two things follow. If your analysis is about prices and supply, drop the seller columns at ingest and the question stops being yours. And if you do keep them, treat the phone column as personal data by default, whatever any given run happens to return.
OLX's own terms restrict automated access, so this is a terms question as well as a legal one - if you have a contractual relationship with the platform, 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 OLX Scraper turns marketplace searches into listing data. You submit olx.* search URLs from any country the platform runs in - typed one per line, or uploaded as a CSV, XLSX, TXT or Parquet file - cap the rows per query if you want to, and download the results as a clean CSV, Excel or JSON file.
Each row carries the listing as OLX's results show it, across twenty-eight columns: the id, title and description, the price and the currency it is quoted in, the posting date, every image, a link to the offer, the city, region and district as separate columns with both display and normalised names, and the seller's display name, account id, account age and business flag. All of it appears in the workbook as well as the JSON.
Pricing teams run one search across several national sites to build a cross-border picture in a single file. Strategy teams use the business flag to split a category into its private and professional halves, which behave differently and mislead when averaged together. Researchers use the split location columns to roll supply up from district to region without string surgery.
Three shapes in this output are worth knowing before you build on it. The currency must be read per row - one search on a single national site returned more than one currency, so inferring it from the domain produces wrong numbers. The images column is one semicolon-joined string rather than an array. And several columns are the same fact twice: url and listing_url matched on every row we have seen, as did image and the first entry of images. Start free: your first 500 rows cost nothing and need no credit card, and after that it is $0.002 per row, flat.