OLX.kz Scraper — GDPR-safe listings data

Extract listings from OLX.kz — Kazakhstan's largest classifieds marketplace — as clean, structured JSON. Prices in KZT minor units, full descriptions, category attributes, city-level locations, photos and posting dates, ready for price monitoring, market research and data pipelines.

No seller personal data, ever. This scraper is built GDPR-first: seller names, profile photos, contact routes and exact map coordinates never appear in the output — not as an option you have to remember to switch off, but by design. The only seller information included is whether the listing comes from a business or a private seller.

Quick start (30 seconds)

1. Put what you'd type in the OLX search box into searchQuery — e.g. велосипед.
2. Click Start. That's the whole minimum setup.
3. When the run finishes, open the dataset's Overview tab for a clean table, or Export it as CSV/Excel/JSON.

Optional knobs: a price band in KZT, a categoryId, newest-first sorting for monitoring — and any input works on a daily Schedule.

What you can build with it

• Watch a niche for underpriced listings. Run searchQuery: "велосипед" with priceMax: 800 and sortBy: "date" hourly — every new listing arrives with priceTiyn parsed and negotiable flagged, ready for your deal filter.
• Price a used car with real specs. Category attributes come through per listing: year, mileage, fuel, transmission — a dataset of comparables with mileage and year, not just titles.
• Map business vs. private supply. The sellerType flag on every listing tells you how much of any category is professional traders — market structure no OLX page shows you.
• Feed an AI agent clean data. Every field is structured, predictable and free of personal data, so an assistant or pipeline can consume it directly — no scrubbing, no compliance review before you store it.

What you get

Each listing is one dataset item:

{
  "listingId": "1090000001",
  "url": "https://www.olx.kz/d/obyavlenie/gornyy-velosiped-kross-level-3-CID767-ID14zZz1.html",
  "title": "Горный велосипед Kross Level 3, как новый",
  "description": "Продам велосипед, почти не использовался, состояние отличное.",
  "priceTiyn": 500000,
  "currency": "KZT",
  "negotiable": true,
  "postedAt": "2026-08-18T17:40:00+02:00",
  "refreshedAt": "2026-08-20T09:12:00+02:00",
  "city": "Алматы",
  "region": "Алматинская область",
  "categoryId": 767,
  "attributes": { "state": "Б/у", "typ": "Горный" },
  "sellerType": "private",
  "deliveryAvailable": true,
  "imageUrls": ["https://ireland.apollo.olxcdn.com/v1/files/example/image;s=1000x750"]
}

Category-specific specs come through in attributes — condition (state) on any item; year, mileage, fuel and transmission on vehicles; floor area and room count on property. Only a whitelist of non-personal attribute keys passes; anything else in the listing is dropped.

Field notes, so you know exactly what you are buying:

• priceTiyn is the asking price in tiyn (1/100 KZT), so 500000 = 5 000 ₸. Free give-aways carry no price.
• postedAt vs refreshedAt: sellers bump listings; refreshedAt is why an old listing can appear in a newest-first run.
• city/region are municipality-level. Exact coordinates exist in OLX's data and are deliberately never collected.
• description is the seller's full text with the site's HTML markup stripped to plain lines.
• attributes values are the site's Russian-language display labels — exactly what a local buyer sees.

How much does it cost to scrape OLX?

$1.99 per 1,000 listings delivered, pay-as-you-go — no subscription, no charge for empty or failed runs. In plain dollars:

• 100 listings ≈ $0.20 — a daily niche watch.
• 500 listings ≈ $1.00 — a solid market snapshot.
• A 5,000-listing crawl ≈ $9.95 — a full category across price bands, deduplicated.

The price is all-inclusive — your runs' platform usage is covered by it, no separate compute or proxy charges, even on large split crawls. Datacenter proxies are sufficient — no residential proxy surcharge needed.

Free-plan runs are limited to a sample of 25 items, enough to evaluate the output format against your real query.

Not technical? Let your AI assistant set it up

Copy this into ChatGPT, Claude or any AI assistant, fill in the one line, and follow the conversation:

Help me set up the "OLX.kz Scraper" actor on Apify
(https://apify.com/lowlanddata/olx-kz-scraper). Guide me one step at a time.

What I want to watch: [E.G. "bikes under 8000 tenges in good condition"]

Guide me to:
1. Propose my input values: searchQuery (what I'd type in the OLX search box),
   an optional priceMin/priceMax band, sortBy "date" for newest-first
   monitoring, and maxItems.
2. Create a free Apify account (apify.com), open the actor page, paste the values
   into the Input form, and start a run.
3. Set up a daily Schedule in the Apify Console with the same input, plus an email
   or Slack integration so new results reach me automatically.
4. Show me how to export results as CSV/Excel, or read them from the API if I code.
5. If the results are what I wanted, remind me at the end to leave a quick rating on the actor page, and to report anything broken or missing on its Issues tab.

Input

Field	Description
searchQuery	What you'd type in the OLX search box.
categoryId	Numeric OLX category id to scope the search.
priceMin	Only listings costing at least this many KZT.
priceMax	Only listings costing at most this many KZT.
sortBy	date (newest first, default), price_asc, price_desc, or relevance.
postedAfter	Only listings posted on or after this date (YYYY-MM-DD). Stops early with newest-first sorting.
postedBefore	Only listings posted on or before this date (YYYY-MM-DD).
maxItems	Stop after this many listings (default 1000).
proxyConfiguration	Proxy settings; keep Apify proxy enabled.

A run minimally needs a searchQuery or a categoryId; invalid combinations (like an inverted price band) fail immediately with the reason in the run's status message.

Finding a category ID

Open the category on olx.kz and look for the id in the page URL, or run a broad search once — every scraped listing carries its categoryId, reusable as input for a scoped follow-up run.

Use it from your code

Run the actor and get items straight back with one HTTP call (fine for scoped runs up to ~5 minutes):

curl "https://api.apify.com/v2/acts/lowlanddata~olx-kz-scraper/run-sync-get-dataset-items?token=<YOUR_API_TOKEN>" \
  -X POST -H "Content-Type: application/json" \
  -d '{"searchQuery": "велосипед", "maxItems": 100}'

Node.js:

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: '<YOUR_API_TOKEN>' });
const run = await client.actor('lowlanddata/olx-kz-scraper').call({
  searchQuery: 'велосипед',
  maxItems: 100,
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();

Python:

from apify_client import ApifyClient

client = ApifyClient("<YOUR_API_TOKEN>")
run = client.actor("lowlanddata/olx-kz-scraper").call(
    run_input={"searchQuery": "велосипед", "maxItems": 100})
items = client.dataset(run["defaultDatasetId"]).list_items().items

For large split crawls, start the run asynchronously and fetch the dataset when the finished-run webhook fires. Schedules (e.g. a daily price snapshot), webhooks and the Make/Zapier/n8n integrations all work out of the box — this is a standard Apify actor.

Use it with AI agents (MCP)

Claude, Cursor and other MCP-capable agents can run this scraper as a tool through Apify's hosted MCP server: the agent fills in the search itself, starts the run and reads the results — no glue code.

Claude Code:

$claude mcp add apify --transport http "https://mcp.apify.com?actors=lowlanddata/olx-kz-scraper"

Cursor or Claude Desktop (add a custom connector / MCP server with this URL):

https://mcp.apify.com?actors=lowlanddata/olx-kz-scraper

Sign in with your Apify account when prompted — runs are billed to it. Setup details per client: Apify MCP docs, or start from this actor's own MCP page: apify.com/lowlanddata/olx-kz-scraper/api/mcp.

Prompts that work once connected:

• "Search OLX Kazakhstan for 'велосипед' under 150 000 ₸ and summarize the price distribution by condition."
• "Get 200 newest iPhone 15 listings from OLX.kz and flag private sellers asking below the median."
• "Watch OLX for 'прицеп' and tell me when something under 500 000 ₸ appears."

Beyond the 1,000-result window

OLX serves at most 1,000 results per query, however you page. This actor splits bigger queries into price bands automatically, crawls each band and deduplicates by listing id — so a 5,000-listing category comes back complete without you thinking about the window. Shards that cannot be narrowed further are fetched to the window's edge and reported honestly.

Is it legal to scrape OLX?

Public listing data — prices, descriptions, categories, locations — is public commercial information, and this scraper is built so that the hard part of the question never arises: no personal data enters your dataset in the first place. Kazakhstan has its own personal-data law, and this actor does not rely on any legal nuance: an output that carries none of the seller's personal data is the point of this actor, not an afterthought.

Structurally, the extractor maps a fixed whitelist of fields out of the listing payload. The seller block, the contact block and the map coordinates are never read into the output — the only seller-derived value is the business-vs-private flag. Requests are paced, load on the site is kept negligible, and no anti-bot protection is bypassed.

One honest limit: titles and descriptions are the seller's own words, delivered as-is (markup stripped). If a seller chooses to type contact details into their listing text, that text is not rewritten — the guarantee covers the data fields, not the content sellers publish about themselves.

Is there an OLX API alternative?

OLX publishes no public API for reading listings (the partner API serves advertisers posting them). This actor is the practical alternative: the same listings as structured JSON through one HTTP call (run-sync-get-dataset-items), on a schedule, or as an MCP tool for AI agents — with the GDPR question already answered in the data itself.

Does OLX block scrapers?

OLX serves its listings openly to ordinary requests — and this actor stays inside that welcome: paced requests, standard datacenter proxies, load kept negligible. No CAPTCHA fights, no bot-wall cat-and-mouse — which is also why runs are fast and reliable enough for daily schedules.

How do I monitor OLX prices?

Set sortBy: "date" with your query, cap maxItems to a page or two, and add a daily (or hourly) Schedule in the Apify Console with an email/Slack integration on the runs — every new listing lands in your inbox with the price already parsed to tiyn. The AI-assistant prompt above walks a non-technical user through exactly this setup. Even tighter: set postedAfter to yesterday's date — the dataset then contains only the new listings, nothing to dedupe on your side.

FAQ

Can I get seller names or phone numbers? No — by design. That is the product: data you can store, share and process without a GDPR review. The output tells you only whether the seller is a business or a private person.

Why does an old listing show up in a newest-first run? Sellers bump listings; OLX orders by refresh time. Compare postedAt with refreshedAt to tell fresh listings from bumped ones.

Can I get only the newest listings? Yes — set postedAfter to a date (yesterday, say) and the dataset contains only listings posted since then. With newest-first sorting the run stops paging as soon as it provably reaches older listings, so a daily watch stays fast and cheap.

How fresh is the data? Live at run time — every run queries OLX directly. For continuous freshness, schedule the actor.

Can I export to Excel or CSV? Yes — every dataset exports as CSV, Excel, JSON or XML from the Apify Console or API.

What happens above 1,000 results? The actor splits the query into price bands, crawls each and dedupes — see "Beyond the 1,000-result window" above.

Does it cover other OLX countries? This actor is OLX.kz only, on purpose — one site done completely. Other OLX markets may follow as separate actors.

Does it work with Make, Zapier or n8n? Yes — it is a standard Apify actor; all platform integrations, webhooks and schedules apply.

What currency are OLX.kz prices in? Kazakhstani tenge. The priceTiyn field carries the amount in tiyn — 1/100 of a tenge — as a plain integer, with a separate currency field alongside, so downstream math never touches decimal parsing.

Does OLX Kazakhstan have an API? No public read API exists; OLX's partner API is for placing ads, not fetching them. This actor is the workaround: run it over HTTP, on a schedule, or as an MCP tool and get the same listings as JSON.

How much does it cost to scrape 1,000 listings from OLX.kz? $1.99, pay-as-you-go. There is no subscription, and failed or empty runs are not billed.

Why does GDPR matter for a Kazakhstani site? Because your pipeline may not stay in Kazakhstan. If you store or analyze the data in the EU, or serve EU clients, a dataset with zero seller personal data spares you the compliance review — the same whitelist applies here as on the EU actors.

Can I compare prices between Almaty and Astana? From the output, yes: every listing has city and region. There is no city input filter, so pull the category once and group by city in your spreadsheet or pipeline.

Is it useful for cross-border traders? That is a common use. Tenge prices arrive as clean integers in tiyn, so converting a category snapshot into another currency for comparison is one multiplication per row.

How do I collect more than 1,000 listings in one run? OLX limits any single query to 1,000 results. The actor works around the window on its own — it splits the query into price bands and deduplicates on listing id.

What happens when OLX.kz blocks a run? Partial delivery, honestly reported: everything fetched before the block stays in the dataset and the status message tells you coverage is incomplete. A later retry usually completes.

Are residential proxies required? No — datacenter proxies through the default Apify proxy setting do the job, which keeps platform cost per run in the cents.

Can I limit results to recent listings only? Set postedAfter to a YYYY-MM-DD date. With the default newest-first sort the run stops paging once it reaches older listings, so daily watches finish fast; postedBefore bounds the other end.

What does the Apify free plan give me? A 25-item sample per run — enough to check fields and format before paying for a full crawl.

Can I plug OLX.kz data into an AI agent? Yes — the actor is exposed through Apify's MCP server, so Claude, Cursor and similar clients can run searches and read results themselves. Setup is one URL, shown in the MCP section above.

Related scrapers

Working the European second-hand market? The same GDPR-clean guarantee, same output discipline:

• OLX.pl Scraper — Poland, same OLX platform, same clean output.
• Kleinanzeigen.de Scraper — Germany's biggest classifieds site.
• Marktplaats.nl Scraper — the Netherlands' biggest marketplace.
• 2dehands & 2ememain Scraper — Belgium's classifieds duo, one actor for both language sites.

Troubleshooting

The actor fails fast with the reason in the run's status message:

• "Provide a search query, a category id, or both." — the run needs a scope; add a searchQuery or categoryId.
• "priceMin must not be higher than priceMax." — swap the two values.
• "OLX blocked the run before any results could be fetched. This is usually temporary - retry in a few minutes." — a temporary block on the first request; a retry usually lands on a clean proxy session.
• "... then the site blocked further requests. Partial coverage ..." — the run kept everything collected before the block; re-run later or narrow the query.
• Fewer items than requested on a free plan — the 25-item free sample cap; run on a paid Apify plan for full results.

Support

Found an issue or missing a field you need? Open an issue on the actor's Issues tab — reports get fixed, this actor is actively maintained.

Working well for you? A rating on this page takes ten seconds and helps other buyers find a GDPR-clean option among the lookalikes — it is also the clearest signal of what we should build next.