Google Keyword Suggestions Scraper

Expand seed keywords into Google's public autocomplete suggestions — ordered the way Google orders them — and profile specific keywords with their suggestion neighborhood. Every row is flat, typed and null-safe for SEO research, content planning, spreadsheets and pipelines.

Why this actor?

• Two buyer-shaped modes: expand seeds into keyword ideas, or profile a keyword list.
• One flat row per keyword: the keyword text, Google's own ordering position and relevance weight, the suggestion type, and a ready Google search URL.
• Hard local result limits keep delivery and spend bounded.
• Invalid input, a keyword with no suggestions, and blocked/5xx/timeout fetches are free error rows.
• Public suggest endpoint only: no login, no cookies, no challenge bypass.

Input

{"mode":"keyword","keywords":["web scraping","keyword research"],"maxResults":100}

For keyword profiles, set mode to detail:

{"mode":"detail","keywords":["web scraping","keyword research"]}

Input	Meaning
mode	keyword (expand seeds into suggestions) or detail (profile specific keywords).
keywords	Seeds to expand (up to 20) or keywords to profile (up to 50); trimmed, deduplicated, capped. One suggest request per keyword.
maxResults	Hard local cap on delivered rows and charges in keyword mode (1–1000).
language	Language code for the query (hl), e.g. en, de, pt-BR.
country	Country code for the query (gl), e.g. us, gb, br.
proxyConfiguration	Proxy for the requests; datacenter is the validated default.

Output

One keyword-mode row:

{"mode":"keyword","input":"web scraping","keyword":"web scraping meaning","seed":"web scraping","kind":"suggestion","position":0,"suggest_relevance":1250,"suggest_type":"QUERY","suggest_subtypes":[512],"verbatim_relevance":null,"language":"en","country":"us","url":"https://www.google.com/search?q=web%20scraping%20meaning","related":[],"error":null,"scraped_at":"2026-09-11T10:11:03.240Z"}

One detail-mode row (the related list is truncated here; the field carries Google's full ordered suggestion list for the keyword):

{"mode":"detail","input":"web scraping","keyword":"web scraping","seed":"web scraping","kind":"detail","position":null,"suggest_relevance":null,"suggest_type":null,"suggest_subtypes":[],"verbatim_relevance":1300,"language":"en","country":"us","url":"https://www.google.com/search?q=web%20scraping","related":[{"keyword":"web scraping meaning","position":0,"relevance":1250,"suggest_type":"QUERY"},{"keyword":"web scraping python","position":1,"relevance":950,"suggest_type":"QUERY"}],"error":null,"scraped_at":"2026-09-11T10:11:03.240Z"}

Field	Meaning
mode	Which mode produced the row: keyword or detail.
input	The seed keyword (keyword mode) or profiled keyword (detail mode) this row came from.
keyword	The delivered keyword: a suggestion in keyword mode, the profiled keyword in detail mode.
seed	The seed keyword this row was suggested for.
kind	suggestion (keyword mode) or detail (detail mode).
position	0-based rank in Google's suggestion list, keyword mode only.
suggest_relevance	Google's own ordering weight for this suggestion (google:suggestrelevance); null when Google omits it.
suggest_type	Google's suggestion type, e.g. QUERY.
suggest_subtypes	Raw google:suggestsubtypes codes for the suggestion.
verbatim_relevance	detail mode only: google:verbatimrelevance for the keyword itself.
language	hl request parameter used for this row.
country	gl request parameter used for this row.
url	Canonical Google search URL for the keyword.
related	detail mode only: Google's ordered suggestions for the keyword; [] in keyword mode.
error	Free per-keyword error message; null on keyword rows.
scraped_at	Extraction timestamp.

Pricing

Pay per event, and the platform usage is on us — the price you see is the price you pay, with no compute bill on top.

Event	Price	What one charge buys
Keyword row	$0.019	One delivered keyword row: an ordered suggestion in keyword mode, or a keyword profile in detail mode. Invalid or empty input, a keyword with no suggestions, and any blocked, 5xx or timed-out fetch are free error rows and are never charged.
Actor Start	$0.002	One run, whatever it returns.

Higher Apify subscription tiers pay less on every event (Silver −20%, Gold −35%).

Proxy

Requests use the Apify proxy configured in proxyConfiguration. The validated default is datacenter. This actor does not bypass challenges, log in, or read private pages.

Use cases

• Grow a seed term into long-tail keyword ideas, in Google's own order.
• Profile an existing keyword list with Google's suggestion neighborhood.
• Export keyword ideas with positions and weights for scoring in a spreadsheet.
• Refresh keyword research for a niche on a schedule (task runs).

FAQ

Does this return absolute search volume? No — Google's public autocomplete surface does not expose absolute volume, and no number here is invented. Each row carries Google's own ordering position and relevance weight (suggest_relevance), which are the signals Google uses to rank suggestions. Any actor that claims absolute volume from this surface is estimating, not measuring.

Do I need a Google account? No. Only the public suggest endpoint is read.

Why is position empty in detail rows? position belongs to a suggestion inside a list; detail rows carry the list itself in related.

Why do some rows have a null suggest_relevance? Google omits the weight for some queries and regions; the key is always present, the value may be null.

This Actor is an independent tool and is not affiliated with Google.