Yandex Wordstat API — Keyword Volume, Trends & Regions

Get Yandex Wordstat keyword data in bulk as clean JSON: the monthly search volume of every keyword (broad, phrase and exact match), up to 2,000 related keywords with their volumes, similar queries, search trends (monthly since 2018, weekly or daily) and demand by city or region. Data comes straight from Yandex — Russia's leading search engine — so it is the keyword data you need for SEO and PPC in Russia, Belarus, Kazakhstan and Uzbekistan. No Yandex account, API key, proxy or captcha solving needed.

Quick facts
Price	$10.20 per 1,000 rows on the Free plan, $6.80 on Gold+; no start fee
One row	a keyword's volume, one related/similar keyword, a whole trend series, or one region
Depth	up to 2,000 related keywords and 20 similar keywords per keyword
Filters	1,100+ regions (city, region or country), device, excluded words, minimum volume
History	monthly from January 2018, weekly, or daily for the last 60 days
Speed	about 100 keywords per hour (Yandex's current hourly request limit); longer lists wait automatically

Jump to Pricing · FAQ · Limitations

Why this Actor

• Yandex's own numbers: the same counts you see on wordstat.yandex.ru, fetched in bulk — no manual copying, no browser automation, no blocked accounts.
• Real ("clean") frequency in one row: broad, "phrase" and "!exact" match volumes side by side, charged as one row.
• Keyword research in one run: up to 2,000 related keywords per seed (most searched first), plus the 20 similar queries Yandex users also searched.
• Seasonality and growth: a trend row per keyword with the full series, average, peak month, change % and year-over-year % (monthly trends of 13+ months).
• Local demand: searches per city or region with Yandex's affinity index (over 100 = more popular there than on average), or limit every count to the regions you target.
• Fair, predictable billing: you pay per delivered row; a keyword nobody searches costs only a $0.0016 no-data fee, invalid keywords and failed requests are free, and your maximum cost per run is always respected.

Compared with alternatives

There is no other Actor with Yandex search volumes in the Apify Store (searched "wordstat", "yandex keyword" and "yandex search volume", October 2026; one small Actor returns Yandex autocomplete suggestions without volumes). The closest listings return Google keyword volumes. Actor fees of public Store listings (Free plan / Gold plan), checked October 2026:

Option	Data	1,000 keyword volumes	Related keywords
This Actor	Yandex	$10.20 / $6.80	yes, up to 2,000 per keyword ($0.0102 / $0.0068 each)
Most-used keyword volume Actor (~1,500 users)	Google	$12 / $8	yes, at the same per-row price
Keyword research Actor (~400 users)	Google	$10	yes, $0.01 per row
Batch volume Actor (~800 users)	Google	$1.99 / $0.40 per batch of 1,000	no

This Actor is 15% cheaper per row than the most-used keyword volume Actor on every plan. If you only need Google volumes for a list of keywords, a batch Google Actor is cheaper — but none of them has Yandex data.

Not included

• Google, Bing or other search engines — Yandex searches only.
• CPC, competition and keyword difficulty — Wordstat does not provide them.
• Search results or rankings — use a Yandex SERP Actor for positions.
• More than 2,000 related keywords or 20 similar keywords per keyword — Yandex's own limits.
• Daily data older than 60 days or monthly data before January 2018 — Yandex's own limits.

What data you get

Every row has id (stable across runs), type, keyword (your input), phrase, searches, position, regionFilter, device and scrapedAt. Extra fields by row type:

Row type (type)	Extra fields	Example
volume — 1 per keyword	searches (broad), phraseMatchSearches, exactMatchSearches (null when not requested), topRelatedKeyword	купить диван: 663,892 / 26,008 / 23,289
related — queries containing the keyword	searches, position, shareOfKeywordPercent, wordCount	купить диван недорого: 76,468 (11.52%)
similar — other queries of the same people	searches, position, wordCount	мебельный магазин: 252,693
trend — 1 per keyword	period, fromDate, toDate, latestDate, averageSearches, peakDate, peakSearches, changePercent, yearOverYearPercent (monthly, 13+ months; else null), series[]	monthly, peak 2026-08
region — demand by place	regionId, regionName, regionLevel, regionSharePercent, affinityIndex	Moscow: 83,899, affinity 113.9

searches is the number of Yandex searches in the last 30 days (for trend rows: the latest period). The dataset has ready-made views: All rows, Search volume, Related & similar keywords, Trends, Regions. Each view shows the columns for one row type and starts with the type column (all rows stay in one dataset, so sort or filter by type in the table or your export), plus a RUN_SUMMARY record in the key-value store.

How to use

1. Open the Actor and click Try for free.
2. Enter your Keywords, one per line (Russian, English and other languages; Wordstat operators work).
3. Pick the Reports — search volume and related keywords are selected by default.
4. Optionally limit counts to Regions (e.g. Moscow, Kazan, 213) or a Device.
5. Click Start and download the results as JSON, CSV, Excel or HTML, or read them through the API.

The prefilled input (1 keyword, volume + 20 related keywords) finishes in a few seconds. Plan big lists at about 100 keywords per hour — 1,000 keywords take about 10 hours in one run (it waits for the hourly limit by itself; keep Maximum wait at 60 minutes or more). One run takes up to 2,000 statistics requests (volume + related = 1 request per keyword; each extra match type, the trend and the regions report add 1 — the run log shows the count and the expected hours at the start; the default run timeout is 24 hours); split bigger lists across runs.

Ready-to-paste input:

{
    "keywords": ["купить диван", "ремонт квартир"],
    "reports": ["volume", "related", "trend"],
    "matchTypes": ["broad", "exact"],
    "maxPhrasesPerKeyword": 100,
    "regions": ["Moscow"],
    "excludeWords": ["бесплатно"]
}

API (curl) — runs the Actor and returns the rows when it finishes:

curl -X POST "https://api.apify.com/v2/acts/cheapapi~yandex-wordstat-keyword-tool/run-sync-get-dataset-items?token=<YOUR_TOKEN>" \
  -H "Content-Type: application/json" \
  -d '{"keywords": ["купить диван"], "reports": ["volume", "related"], "maxPhrasesPerKeyword": 50}'

JavaScript (npm install apify-client):

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: '<YOUR_TOKEN>' });
const run = await client.actor('cheapapi/yandex-wordstat-keyword-tool').call({
    keywords: ['купить диван', 'ремонт квартир'],
    reports: ['volume', 'related'],
    maxPhrasesPerKeyword: 50,
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();
console.log(items.filter((r) => r.type === 'volume'));

Python (pip install apify-client):

from apify_client import ApifyClient

client = ApifyClient("<YOUR_TOKEN>")
run = client.actor("cheapapi/yandex-wordstat-keyword-tool").call(run_input={
    "keywords": ["купить диван", "ремонт квартир"],
    "reports": ["volume", "related"],
    "maxPhrasesPerKeyword": 50,
})
for row in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(row["type"], row["phrase"], row["searches"])

Use cases

• SEO for Russian-speaking markets: build a semantic core from seed keywords, with real (exact-match) frequencies to prioritise pages.
• Yandex Direct (PPC) planning: size demand per keyword, city and device before launching campaigns; find negative keywords in related queries.
• Seasonality: see which months a product peaks and plan content, stock and budgets with year-over-year growth.
• Regional expansion: compare demand and affinity across cities to choose where to open, advertise or localise.
• Market and trend research: track brands, products or topics monthly — schedule the Actor and compare runs.
• Content planning: turn related and similar queries into article topics and FAQ sections.

Advanced options

Option	Default	What it does
keywords	— (required)	One keyword per line, up to 400 characters; up to 2,000 statistics requests per run (2,000 keywords with the default reports, about 20 hours). Operators: "…", !word, [word order], -word.
reports	volume, related	Any of volume, related, similar, trend, regions.
regions	all	City, region or country names (English or Russian) or Yandex region IDs; up to 100. Applies to volume, related, similar and trend.
device	all	all, desktop, phone, tablet.
maxPhrasesPerKeyword	50	Related keywords per keyword, 1–2,000 (each is a charged row).
minSearches	0	0 or more: skip related/similar keywords with fewer monthly searches (skipped ones are free).
excludeWords	—	Leave out queries containing these words (added as -word).
matchTypes	broad	Volume columns: broad, phrase ("…"), exact ("!…"). Still one charged row; phrase/exact stay empty for keywords that already use operators.
maxSimilarPerKeyword	20	Similar keywords per keyword, 1–20.
trendPeriod	monthly	monthly (from 2018), weekly, daily (last 60 days).
trendFrom / trendTo	last 24 months / 52 weeks / 30 days	Dates YYYY-MM-DD; moved to whole months or Monday–Sunday weeks when needed.
regionLevel	cities	Regions report: cities, regions or all.
maxRegionsPerKeyword	20	Places per keyword in the Regions report, most searches first (each is a charged row).
maxWaitMinutes	65	Minutes each request may wait for Yandex's hourly request limit (0–120) before the run stops; the limit resets within 60 minutes.

Output example

Real rows from a run for купить диван (all reports with matchTypes broad/phrase/exact, 3 related, 2 similar, 2 regions and a custom monthly trend from trendFrom: 2025-07-01; October 2026; some rows and trend points left out):

[
  { "id": "1e40eed40219cd70", "type": "volume", "keyword": "купить диван", "phrase": "купить диван", "searches": 663892, "position": null,
    "regionFilter": "All regions", "device": "all", "scrapedAt": "2026-10-05T14:46:36.494Z",
    "phraseMatchSearches": 26008, "exactMatchSearches": 23289, "topRelatedKeyword": "купить диван недорого" },
  { "id": "095c907d78e099a0", "type": "related", "keyword": "купить диван", "phrase": "купить диван недорого", "searches": 76468, "position": 1,
    "regionFilter": "All regions", "device": "all", "scrapedAt": "2026-10-05T14:46:36.494Z", "shareOfKeywordPercent": 11.52, "wordCount": 3 },
  { "id": "9dea7c84e2e4d2fd", "type": "similar", "keyword": "купить диван", "phrase": "мебельный магазин", "searches": 252693, "position": 1,
    "regionFilter": "All regions", "device": "all", "scrapedAt": "2026-10-05T14:46:36.494Z", "shareOfKeywordPercent": null, "wordCount": 2 },
  { "id": "3f6a080f4ad36ba2", "type": "trend", "keyword": "купить диван", "phrase": "купить диван", "searches": 690667, "position": null,
    "regionFilter": "All regions", "device": "all", "scrapedAt": "2026-10-05T14:46:36.494Z",
    "period": "monthly", "fromDate": "2025-07-01", "toDate": "2026-09-30", "latestDate": "2026-09-01", "averageSearches": 650007,
    "peakDate": "2026-08-01", "peakSearches": 911485, "changePercent": -9.7, "yearOverYearPercent": -1.2,
    "series": [ { "date": "2025-07-01", "searches": 764886, "sharePercent": 0.007228 }, { "date": "2026-08-01", "searches": 911485, "sharePercent": 0.008115 },
                { "date": "2026-09-01", "searches": 690667, "sharePercent": 0.005944 } ] },
  { "id": "3e360e6cc07ecbae", "type": "region", "keyword": "купить диван", "phrase": "купить диван", "searches": 83899, "position": 1,
    "regionFilter": "All regions", "device": "all", "scrapedAt": "2026-10-05T14:46:36.494Z",
    "regionLevel": "cities", "regionId": "213", "regionName": "Moscow", "regionSharePercent": 0.006557, "affinityIndex": 113.9 }
]

Pricing

Pay per event — you pay per delivered row, or a small $0.0016 no-data fee per request for a keyword with nothing to deliver, with no start fee. Apify platform usage is billed separately by Apify. This Actor uses 256 MB of memory; our test runs used about $0.0005 of platform usage each, and a run that waits at Yandex's hourly limit adds roughly $0.10 per hour of waiting (at $0.40 per compute unit).

Event	Free	Bronze	Silver	Gold, Platinum, Diamond
Keyword row (keyword-row)	$0.0102	$0.00906	$0.00793	$0.0068
Lookup without data (empty-lookup)	$0.0016	$0.0016	$0.0016	$0.0016

• A volume row is one row however many match types you choose. A trend row contains the whole series.
• Lookup without data: only when a keyword gives nothing to deliver (nobody searched it, or your filters removed everything) — charged once per request made for that keyword. Search volume (with all its match types counted as separate requests), trend and regions are one request each; related and similar share the volume request. With the Search volume report, such a keyword still gets its volume row (all zeros) for this fee only.
• Free: invalid keywords, keywords Yandex rejects, failed requests, duplicates, and related keywords below your minimum. If one report of a keyword fails, the rows of its other reports are still delivered and charged as usual.

Worked examples (Free / Gold):

• Search volume for 1,000 keywords (1,000 rows): $10.20 / $6.80 (about 10 hours at the current hourly limit, plus about $1 of platform usage for the waiting).
• 100 seed keywords × (volume + 50 related keywords) (5,100 rows): $52.02 / $34.68.
• Full research for one keyword — volume, 200 related, 20 similar, monthly trend and the top 20 cities (242 rows): $2.47 / $1.65.

Set Maximum cost per run in the run options to cap any run; the Actor stops cleanly at the limit and tells you what was skipped.

Apify Free plan: up to $0.25 of results per month (enough to try the prefilled input — 1 keyword with volume and 20 related keywords costs $0.21). Any paid Apify plan removes this limit. When many users are running at the same time, even a trial run may wait for Yandex's hourly limit before it returns rows (it waits up to Maximum wait, and nothing is charged while it waits).

Integrations

• Google Sheets, Excel, CSV: export the dataset or use the Google Sheets integration to fill a sheet after every run.
• Make, Zapier, n8n: start the Actor and receive rows in your scenarios.
• Webhooks: get a call when a run finishes.
• Scheduling: run monthly to track trends; every row has a stable id.
• API and MCP: call it from code (see above) or from AI agents through Apify's MCP server.

FAQ

Is it legal? The Actor gets its data through Yandex's official, paid keyword statistics API (Yandex Cloud Search API), not by scraping the Wordstat website. The data is aggregated search counts with no personal data. You are responsible for how you use the results, including Yandex's terms of use if you republish the numbers.

How fresh is the data? Volumes are for the last 30 days, fetched live at run time. Monthly trends end with the last finished month, weekly with the last finished week, daily with yesterday.

Why do the volume row and the trend row show different numbers? The volume row counts the last 30 days; the trend row's searches is the last calendar month (or week/day).

What is the difference between broad, phrase and exact match? Broad counts every query containing the words (купить диван includes купить угловой диван недорого). Phrase ("купить диван") counts only queries made of exactly these words. Exact ("!купить !диван") also fixes each word form. SEO specialists usually plan with phrase or exact match.

Why did I get fewer related keywords than I asked for? Yandex returns fewer for rare keywords; minSearches, excludeWords and duplicates also remove some. The keyword itself is never repeated as a related keyword (it is the volume row), so with the maximum of 2,000 you get up to 1,999 related keywords plus the volume row.

Why was I charged for a keyword with no searches? Yandex still processed the lookup, so a small $0.0016 no-data fee applies per request made: $0.0016 with the default reports, up to $0.008 with every report and match type enabled (5 requests). Invalid keywords and failed requests are free.

Can I limit counts to a city? Yes — put city, region or country names in Regions. The Regions report itself always ranks every place (it shows where the demand is).

The run says it is waiting at the hourly limit — is something wrong? No. Yandex limits how many statistics requests can be made per hour (about 100 at the moment, shared by all users of this Actor); the run waits and continues automatically (up to Maximum wait for each request), and nothing is charged while it waits. Platform usage while waiting is about $0.10 per hour.

Does it work for Turkey or other countries? Wordstat counts searches made on Yandex. They are large in Russia, Belarus, Kazakhstan and Uzbekistan, and small elsewhere (Turkish keywords often show only tens of searches).

Do I need a Yandex account or API key? No. Everything is included in the price.

Support

Found a problem or need a feature? Open an issue on the Actor's Issues tab with the run ID; we aim to answer within one working day.

Every run also saves a RUN_SUMMARY record in its key-value store: keywords requested, completed, without data, partial, failed and skipped by your cost limit, rows delivered, no-data fees charged, waits at the hourly limit, and a problems list with the reason for each keyword that was skipped or incomplete.

Limitations

• Throughput: Yandex currently allows this Actor about 100 statistics requests per hour. Volume + related costs 1 request per keyword; each extra match type, the trend and the regions report add 1 each. Large lists are processed at that pace — the run waits automatically, or you can split the list across runs.
• Counts are Yandex's own statistics: estimates, not exact logs; very rare keywords may show 0.
• Up to 2,000 related and 20 similar keywords per keyword; daily trends only for the last 60 days; monthly trends from January 2018.
• The region filter applies to volume, related, similar and trend; the Regions report always covers every place.
• Region names are matched in English or Russian; when several places share a name, the larger one is used — give the numeric Yandex region ID to be exact.