Semrush Keyword Overview Scraper — Volume, KD, CPC & Intent

Pull Semrush keyword overview data at scale — search volume, Keyword Difficulty, CPC, search intent, 12-month trend, global per-country breakdowns, related keyword ideas, and the full organic SERP — with no Semrush subscription, no API key, and pay only per successful result.

This Semrush keyword scraper delivers the complete keyword overview dataset for any list of keywords. Feed it your keyword list, choose a country database, and get back a structured record per keyword with every metric the Keyword Overview tool surfaces — plus structured global volumes across ~120 countries and optional organic SERP results that competing tools don't include. It runs entirely on Apify's residential proxy for reliable access, so you never need to configure anything on your end.

Features

• Full keyword overview metrics — monthly search volume, CPC (USD), competition level, Keyword Difficulty (numeric 0–100), number of search results, referring domains median, and search intent (commercial / informational / navigational / transactional).
• 12-month volume trend — a month-by-month array showing how search demand for the keyword moves over time, useful for seasonality analysis.
• Structured global volumes — per-country breakdown (volume, CPC, competition, KD) across ~120 country databases, returned as a structured array. Not a flat string like some alternatives — every country is a queryable object, sorted by volume descending.
• Keyword ideas (optional, on by default) — up to 30 related keyword suggestions per keyword, each with its own volume and KD score.
• Organic SERP (optional, off by default) — up to 100 organic results per keyword, including position, domain, URL, and SERP features present on each result.
• SERP features list — which rich results appear on the SERP (featured snippet, video, local pack, image pack, etc.).
• No Semrush account or API key required — the actor handles access internally.
• Pay only for successful results — keywords that return no Semrush data are shown in the output with an error note but are never billed.
• Fault-tolerant delivery — each result row is pushed to the dataset the instant it is ready, so a run timeout never discards work already completed.

What data you get

One record per keyword. All fields verified on a live cloud run:

Field	Description
keyword	The input keyword phrase
database	The country database used (e.g. us, uk, de)
volume	Monthly search volume in the selected country
cpc	Cost-per-click in USD
competition_level	Ad competition level (0–1 float)
keyword_difficulty	Numeric KD score (0–100)
results_count	Number of results returned by the search engine
referring_domains_median	Median referring domains of the top-10 ranking pages
intents / intent_labels	Search intent codes and human-readable labels
serp_features	Array of rich SERP features active for this keyword
trend	Array of 12 monthly search volume data points
global_volumes	Per-country array: database, volume, cpc, competition_level, keyword_difficulty
global_country_count	Number of countries with non-zero volume (~120 max)
ideas	Related keyword suggestions, each with phrase, volume, and KD
serp	Organic SERP rows: position, domain, url, serp_features (when enabled)
checked_at	ISO timestamp when the keyword was processed
error	Populated only on failed lookups (not billed)

Example result for "seo tools": volume 40 500 · KD 90 · CPC $7.04 · results 128 · rds_median 589 · intent commercial · global_country_count 120 · ideas[30] · serp[100].

This actor is a strict superset of the leading alternative: where that tool ships global multi-country volume as an unstructured flat string, we return a fully structured global_volumes[] array per country. We also add ideas[], a full organic serp[], and a numeric referring_domains_median that the market leader does not expose.

Input

{
  "keywords": ["seo tools", "keyword research", "content marketing"],
  "database": "us",
  "includeIdeas": true,
  "maxIdeas": 30,
  "includeSerp": false,
  "proxyConfiguration": {
    "useApifyProxy": true,
    "apifyProxyGroups": ["RESIDENTIAL"]
  }
}

Parameter	Type	Default	Description
keywords	array	required	List of keywords to look up
database	string	us	Two-letter country database code (us, uk, de, in, …)
includeIdeas	boolean	true	Return related keyword ideas per keyword
maxIdeas	integer	30	Cap on ideas per keyword
includeSerp	boolean	false	Return the full organic SERP per keyword
maxItems	integer	0	Cap total keywords processed (0 = no cap)
proxyConfiguration	object	RESIDENTIAL	Proxy setting — defaults to Apify RESIDENTIAL

Pricing

Pay per result, not per run.

• $0.018 per keyword that returns metrics (charge event: keyword-result).
• Keywords that fail to return data are included in the output for transparency but are not billed.
• At 100 keywords, the run costs ~$1.80. The organic SERP option adds no extra cost — it is included in the same per-keyword charge.
• This rate is approximately 10% below the comparable market-leader's per-result price.

FAQ

Do I need a Semrush subscription or API key? No. This actor handles access internally using a residential proxy for reliable access. You do not need a Semrush account, a paid plan, or any external API credentials.

How many keywords can I process per run? The actor is throughput-capped by design — it is built for focused, high-value keyword research rather than mass bulk extraction. It works well for typical research workloads (tens to low hundreds of keywords per run). Very large batches may run into session limits before completing; if you have a large list, split it across multiple runs and the actor will pick up reliably each time. Each successful row is pushed immediately, so no work is lost.

Which country databases are supported? Any two-letter Semrush database code: us (United States), uk (United Kingdom), de (Germany), fr (France), in (India), ca (Canada), au (Australia), and many more. Set the database field to the code you need. Regardless of which database you pick, the global_volumes[] field returns volume data across ~120 countries in a single call.

What is referring_domains_median and why does it matter? It is the median number of referring (linking) domains that the top-10 ranking pages have. It is a direct proxy for the link authority you need to compete for a keyword — more useful for practical SEO planning than raw KD alone, and not available in comparable scrapers.

Use cases

Keyword research and content planning — bulk-fetch volume, KD, CPC, and intent for any keyword list without opening the Semrush UI. Filter by KD range or intent to find content opportunities that match your site's authority level.

International and global SEO — use global_volumes[] to compare how the same keyword performs across 120 countries in a single run. Identify markets where a keyword has high volume but lower competition before committing to a geo-expansion strategy.

SERP analysis and competitor mapping — enable includeSerp to retrieve up to 100 organic results per keyword. See exactly which domains dominate each SERP, identify SERP features that affect click-through, and map content gaps.

Keyword clustering and topic modeling — combine the ideas[] field (related suggestions with volume + KD) across multiple seed keywords to build keyword clusters and topic maps programmatically, without paying per idea lookup.

Trend and seasonality detection — the trend[] array gives 12 months of monthly search volume. Use it to detect seasonal peaks, declining topics, or emerging demand before publishing content.