Yandex Search Engine Scraper (SERP)

Search Yandex by keyword and region and get one row per result slot: rank, title, link, the domain Yandex shows, the snippet, sitelinks, and whether the slot is organic, a Direct ad or a featured block.

The same query put to Yandex from Moscow and from Istanbul does not come back as the same page. Yandex ranks by region, the region is a code in the URL. The honest part: Yandex does not answer every query, and one it will not serve comes back as an uncharged diagnostic row instead of results, so a run can end with far fewer rows than you asked for. There is also no total-result count on a modern Yandex page, so nothing here can report one.

Input	Keywords, or full Yandex search URLs, with optional date, language and family filters
Output	One row per result slot: rank, title, link, displayed domain, snippet, slot type, the query Yandex really searched
Ceiling	20,000 results per run, up to 40 pages per query
Account needed	None, and no API key
Price	$3.40 per 1,000 results on the Free plan, a little less on paid plans. The free plan's $5 a month covers about 1,400

🔍 What Yandex Search Engine Scraper does

Give it keywords and a region code and it pages through the results, writing a row for every slot on the page in the order Yandex put them.

The region code is the point. lr=213 is Moscow, 225 is Russia, 11508 is Istanbul, 163 is Astana. Set one and you get that region's ranking. Leave it empty and Yandex picks, which is fine for a look and wrong for a comparison.

Three kinds of slot come back, and all three take up a rank position: organic listings, Yandex Direct ads, and featured blocks like video carousels and image strips. Each row carries position for the slot's place on the page and organicPosition for its place among the organic rows only, which is the one rank trackers want.

You can also narrow a search the way Yandex's own tools do: by date, newest first, by page language, with the family filter, or to the exact spelling you typed.

📋 What data you get from each Yandex search result

What you get	Field
The slot's rank across the run, on its page, and among organic rows only	position, positionOnPage, organicPosition
Organic, ad or featured, and which kind of featured block	type, featureType, isAd
The listing as Yandex renders it	title, url, snippet
The breadcrumb Yandex prints, the real domain behind it, and the brand	displayedUrl, displayedDomain, brandName
The extra links under a listing	sitelinks
The region Yandex served and the one you asked for	region, regionRequested
The front end that answered and the one you picked	domain, domainRequested
The query, its page, and the exact search address	searchQuery, page, searchUrl
What Yandex really searched, when it corrected your spelling	correctedQuery
When Yandex says the page was published, as a date and as printed	date, dateText
The result's picture and the short facts under it, where Yandex shows them	thumbnail, richSnippet
Related searches, as a list (optional row)	relatedSearches
The knowledge card beside the results (optional row)	title, snippet, entityFacts, entitySources

▶️ How to scrape Yandex search results

1. Open Yandex Search Engine Scraper and click Try for free.
2. Put your keywords into Search queries, replacing the example.
3. Pick the market you care about in Region (pick by name), or type any code into Region (Yandex lr code). This is the setting that changes the answer.
4. Set Max results in total low for the first run, then click Start.
5. Download the dataset as JSON, CSV or Excel, or read it from the Apify API.

Run it with no queries and no URLs and you get one labelled sample row, not charged.

💰 How much does it cost to scrape Yandex search results?

$3.40 per 1,000 results on the Free plan. Paid plans pay a little less: $3.35 on Starter, $3.30 on Scale and $3.25 on Business. On the free plan, the $5 Apify gives you each month covers about 1,400 results. A run of 100 results costs about $0.34 on the Free plan.

Every row with a rank counts, ads and featured blocks included, because they take up a rank slot. If you only want organic listings, turn includeAds and includeFeatured off and you stop paying for them. With adsOnly you pay for the ads alone. The optional related-searches and knowledge-card rows cost one result each.

Sample rows, notice rows and diagnostic rows are not charged, and a query Yandex will not serve costs you nothing.

📥 What you give it

{
  "queries": ["купить пластиковые окна", "ergonomic office chair"],
  "domain": "yandex.com.tr",
  "region": 213,
  "maxItemsPerQuery": 20,
  "maxItems": 100
}

Field	Default	What it is
queries	none, the box holds купить пластиковые окна	Keywords, in any language Yandex handles. Russian, Turkish, English, Kazakh.
startUrls	empty	Full Yandex search URLs instead. The query, region and start page are read out of the address. To carry on a deep search where it stopped, add &p= to the address: Yandex counts pages from 0, so &p=4 starts on page 5.
domain	yandex.com.tr	Which Yandex front end to ask. They share one index. yandex.com.tr is the default because it honours the region code. yandex.uz ignores the region code and always ranks for Uzbekistan.
regionPreset	empty, the box holds Moscow (213)	The region picked by name from the 28 verified codes.
region	empty	Any lr code, typed as a number. It wins over regionPreset. Verified ones include 225 Russia, 213 Moscow, 2 Saint Petersburg, 43 Kazan, 187 Ukraine, 143 Kyiv, 149 Belarus, 159 Kazakhstan, 10393 Almaty, 983 Turkey, 11508 Istanbul, 171 Uzbekistan, 168 Azerbaijan, 169 Georgia, 84 USA, 102 Germany, 95 United Kingdom.
timeRange	any	day, twoWeeks or month: only pages Yandex dates within that window.
sortByDate	off	Newest first instead of by relevance. Ranks then mean freshness.
resultsLanguage	empty	Only pages in these languages: ru, en, tr, uk, be, kk, tt, de, fr, id.
safeSearch	off	Yandex's strict family filter.
noAutocorrect	off	Search exactly what you typed, even when Yandex thinks it is a typo.
maxItemsPerQuery	20	Results per query, up to 2,000. Yandex serves about 10 to 20 a page.
maxItems	100	Hard cap across every query. Every returned result counts, so this is your budget cap.
maxPagesPerQuery	15	How deep to page for one query, up to 40.
includeAds	on	Include Direct ad slots, typed ad.
includeFeatured	on	Include carousels and image strips, typed featured. Turn it off for the ten blue links only.
adsOnly	off	Only the ad slots. Everything else is dropped before it is charged.
includeRelatedSearches	off	One extra row per query with Yandex's related searches in relatedSearches.
includeEntityCard	off	One extra row per query with the knowledge card, when Yandex shows one.
failIfEmpty	off	End the run as failed when no result came back, so a schedule or webhook watching run status notices.
proxyConfiguration	leave it alone	Optional. Not needed for a normal run.

📤 What you get back

A real organic row from a run on 17 October 2026 for Moscow, with a typo in the query, which is why correctedQuery is filled:

{
  "searchQuery": "купить ноутбукк",
  "page": 1,
  "position": 3,
  "positionOnPage": 3,
  "organicPosition": 3,
  "type": "organic",
  "featureType": null,
  "isAd": false,
  "title": "Ноутбуки купить в интернет-магазине DNS. Ноутбуки...",
  "url": "https://www.dns-shop.ru/catalog/17a892f816404e77/noutbuki/",
  "displayedUrl": "dns-shop.ru › catalog › 17a892f816404e77 › noutbuki",
  "displayedDomain": "dns-shop.ru",
  "brandName": "Dns-shop.ru",
  "snippet": "Купить Ноутбуки по самым выгодным ценам в интернет-магазине DNS.",
  "sitelinks": [],
  "region": 213,
  "regionRequested": 213,
  "domain": "yandex.com.tr",
  "domainRequested": "yandex.com.tr",
  "searchUrl": "https://yandex.com.tr/search/?text=%D0%BA%D1%83%D0%BF%D0%B8%D1%82%D1%8C%20%D0%BD%D0%BE%D1%83%D1%82%D0%B1%D1%83%D0%BA%D0%BA&lr=213",
  "scrapedAt": "2026-10-17T17:42:44.253Z",
  "correctedQuery": "купить ноутбук",
  "date": null,
  "dateText": null,
  "thumbnail": null,
  "richSnippet": null
}

Field	How to read it
position, organicPosition	position counts every slot, organicPosition only the organic ones. They differ when an ad or a featured block sits above.
url	The listing's link. On an ad it is Yandex's click tracker, and the advertiser is in displayedDomain.
region, regionRequested	What Yandex actually served, and what you asked for. Compare them before trusting a ranking.
sitelinks	The extra links under a listing, as {title, url}. Empty on most rows.
snippet	On a featured row, the start of the block's text.
date, dateText	dateText is Yandex's own label, in Turkish on yandex.com.tr; date is that label as a calendar day. Empty when Yandex prints no date.
correctedQuery	Empty unless Yandex fixed your spelling and searched its own version. Turn on noAutocorrect to stop that.

🧾 Reading the output

Three kinds of row can land in your dataset.

Row	How to spot it
A result	It has a position and a type
A related-searches or knowledge-card row	type is featured, featureType is related or entity-card, and position is empty because it sits outside the ranked list
The sample row	Written only when there were no queries and no URLs
A notice or diagnostic	It carries an errorCode

Diagnostic and notice rows are never charged. Each run also saves a summary as the OUTPUT record in its key-value store: results by type, pages read, empty or unserved queries, and why it stopped.

Code	What it means
BAD_URL	That entry is not a Yandex search URL. The other jobs carry on.
BAD_INPUT	A search option holds a value that is not on its list. Nothing was searched.
DOMAIN_FAILOVER	The host you picked would not serve that query, so it was asked on yandex.com.tr at the same region instead. The domain field on the rows says who answered.
NOT_SERVED	Yandex did not serve that query on this run. Re-run it later. Until 12 October 2026 this code was called CAPTCHA, so update any filter that used the old name.
FETCH_FAILED	Yandex was unreachable or answered badly.
NO_RESULTS	The query returned nothing usable. Under a date, language or family filter, or with adsOnly, that can simply mean nothing matched.
ABORTED	Three queries in a row came back dead, so the rest were skipped rather than burned through.
TIME_LIMIT	The run was close to its time limit, so it stopped and kept everything it had already delivered. Give it more time, or ask for fewer results, to get the rest. If nothing had come back by then, the run ends as failed.
PROXY_GROUP_IGNORED	A setting in proxyConfiguration could not be used, so the run carried on without it.

💡 What people use it for

• Tracking where a site ranks in Moscow, Almaty or Istanbul, where a Google-only view says nothing.
• Watching which advertisers are buying a keyword, through displayedDomain on the ad rows.
• Pulling the snippet Yandex writes for your pages, which is not always the meta description.
• Comparing the same keyword across regions by running it once per lr code.
• Checking whether featured blocks are pushing the organic results down a market you care about.
• Following a topic over the last day, newest first.

🚧 What it does not do

• No total result count. Yandex does not print one any more, so nothing can report it.
• Leave region empty and each query keeps its own. Every query pins to whatever region its first page came back with, so two queries in one run can be ranked for different places. Set region whenever you are comparing anything.
• yandex.uz ignores the region code and always ranks for Uzbekistan.
• No row for Yandex's AI answer box. It's left out and never charged, and the rows under it keep their real position.
• Some queries will not be served. You get an uncharged row saying so, and a run can end with very few results.
• No site: searches in practice. Yandex did not serve those queries in our tests.
• Dates only where Yandex prints them, mostly on fresh pages.
• About 20 pages is the realistic depth for one query, whatever the page cap allows.
• Web results only. No images, video, maps or news verticals.
• A ranking is a snapshot, and Yandex still moves daily.
• It is not an official Yandex tool. Dami's Studio is independent and is not affiliated with or endorsed by Yandex, or by any other company named on this page.

🧭 Which Yandex scraper do you need?

If you want	Use
Businesses on Yandex Maps, with hours and coordinates	Yandex Maps Scraper
Reviews for a place on Yandex Maps	Yandex Maps Reviews Scraper
Yandex search results by keyword and region	This one
Products and prices on Yandex Market	Yandex Market Scraper
The same job on Google, with People Also Ask	Google Search Results Scraper
What a search box suggests as people type	Keyword Autocomplete Scraper

❓ Questions people ask

Do I need a Yandex account or an API key to scrape Yandex?

No. Nothing to apply for and no quota to watch.

Which region code should I use?

The one for the market you sell in. 213 for Moscow, 225 for Russia as a whole, 11508 for Istanbul, 163 for Astana. The list in the input has the verified ones.

Why don't the rankings match what I see on Yandex?

Usually the region. Your browser gets your own city's ranking; the run gets the lr code you set, or one Yandex picks. To compare like with like, open a private browser window and put the same lr in the search address.

Why did I get fewer rows than I asked for?

Either the query ran out of pages, or Yandex would not serve it. The diagnostic row tells you which, and neither costs you anything.

Can I call it from code or connect it to an AI assistant?

Yes. The API tab has ready-made code for Python, JavaScript and the command line. For Claude, ChatGPT or another MCP client, connect https://mcp.apify.com/?tools=fetch-actor-details,dami_studio/yandex-search-engine-scraper. Either way the run happens on your Apify account at the same price.

Is scraping Yandex search results legal?

They are public pages. Apify's write-up on the legality of web scraping is a good starting point, and we are not lawyers.

🆘 If something breaks

Open the Issues tab on the actor page. Send the query, the region code and the run ID. The errorCode on the diagnostic row usually names the problem on its own.