Google Autocomplete Scraper — Bulk

What this Actor scrapes

For every (seed × language × country) combination you submit, this google autocomplete scraper queries Google's Suggest endpoint (suggestqueries.google.com/complete/search) and emits one row per returned suggestion. The default client=chrome mode returns up to 15 suggestions per call plus relevance scores and subtype tags; the lean client=firefox mode returns up to 10 suggestions without relevance scores. Each row carries the seed, the suggestion, its 0-indexed position, the locale used, and an is_expansion flag so alphabet-expanded long-tail suggestions are easy to filter or aggregate.

What we handle for you

• Bulk seeds — submit up to 500 seed queries in a single run; the Actor fans out across every language × country combination you specify.
• Multi-locale — query the same seed across many languages (hl) and many countries (gl) in one run. The Suggest endpoint serves locale-specific results: "bestes auto" in de-de returns German automotive suggestions; "best laptop" in en-us returns US tech suggestions.
• A–Z alphabet expansion — toggle enableAlphabetExpansion to also query <seed> a, <seed> b, …, <seed> z for every seed. This is the canonical long-tail discovery technique: one seed × 26 letters reveals every common Google query that starts with your seed.
• Two client variants — chrome (default) returns up to 15 suggestions with relevance scores and subtype tags; firefox returns up to 10 suggestions in a leaner payload.
• Denormalised output — every row carries the seed, locale, and provenance fields so a flat CSV is self-contained for SQL / BI / spreadsheet workflows without joins.
• We rotate browser fingerprints — curl-cffi replays a real Chrome TLS+H2 fingerprint so requests look indistinguishable from a real browser, not a Python script.
• We retry with exponential backoff — 408 / 429 / 5xx responses trigger automatic retries (up to 5 attempts, honours Retry-After). You get results; we handle the turbulence.
• We rotate proxies on block — Apify Proxy is wired in; on any block a fresh session ID and fresh exit IP take over automatically.
• Conservative concurrency — 2 in-flight requests per run by default; the Actor backs off gracefully when the target rate-limits and surfaces partial results with a clear status message rather than returning an empty dataset.
• Pydantic v2 input validation — locale codes are normalised to lowercase, length-validated, and rejected up-front if invalid before any network call.

Use cases

• Long-tail SEO keyword discovery — enable alphabet expansion to enumerate every common Google query that starts with your seed; pipe into a search-volume tool to prioritise content briefs.
• AnswerThePublic alternative — get the same autocomplete suggestions direct from Google for a fraction of the cost. 1,000 suggestions = $1.20 here vs $89–199/mo for a SaaS dashboard that wraps the same endpoint.
• PPC negative-keyword mining — pull alphabet-expanded suggestions across the seed for your category, then filter for adjacent products or intents you want to exclude from broad-match campaigns.
• Content gap analysis — query your brand + competitor names across en-us, en-gb, en-au to find geo-specific concerns and product comparisons real users are searching for.
• Multi-market localisation research — query the same product seed across 5–10 locales to see how local users phrase the same intent in their language; informs product copy, landing pages, and marketplace listings.
• Trend monitoring — schedule a recurring run on a fixed seed list and diff the suggestion list day-over-day to detect emerging interest spikes before they hit the SERP.
• AI training-data curation — bulk-export real human search phrasings for fine-tuning intent classifiers or query-understanding models; the suggestions are post-processed real queries, free from PII.
• Free keyword research API replacement — developers who were paying per-call for a keyword research API can replace that spend with a single Apify Actor at $1.20 per 1,000 results.

Input

Paste this into the Apify Console, or send it as the run input over the API. Proxy settings are on by default; you rarely need to touch them.

{
  "seeds": [
    "best laptop",
    "best espresso machine"
  ],
  "languages": [
    "en"
  ],
  "countries": [
    "us"
  ],
  "client": "chrome",
  "maxSuggestionsPerSeed": 10,
  "enableAlphabetExpansion": false,
  "useProxy": false
}

Output

One row per result, schema-validated before it is written. Export JSON, CSV, Excel or XML from the run, or read it over the API.

seedsuggestionpositionrelevancesubtypelanguagecountryclientis_expansionparent_seedscraped_at

{
  "seed": "best laptop",
  "suggestion": "best laptop 2026",
  "position": 1,
  "relevance": 900,
  "subtype": "QUERY",
  "language": "en",
  "country": "us",
  "client": "chrome",
  "is_expansion": false,
  "parent_seed": null,
  "scraped_at": "2026-05-16T12:00:00.000Z"
}

Pricing

You pay only for results that land. Cap any run with maxTotalChargeUsd. See pricing & billing for worked examples.

Limitations

• Suggest endpoint, not SERP. This Actor returns autocomplete suggestions — the dropdown under the search box. It does not scrape the actual Google search results page. For SERP scraping use a dedicated SERP Actor.
• No search-volume data. Google Suggest does not return monthly search-volume estimates — only the suggestions themselves. Pair with a keyword-volume Actor or Google Keyword Planner for volume.
• Locale is a request hint, not a guarantee. Google may return general suggestions when locale-specific signal is sparse; obscure seeds in long-tail locales may return small or empty suggestion lists.
• A–Z expansion only. This Actor expands seeds with a..z suffixes. Number suffixes (0..9), question modifiers (why, how, when), or comparison modifiers (vs, or) are not built-in — they can be passed as explicit seeds.
• Concurrency capped at 2. The Actor deliberately limits in-flight requests to stay well within the Suggest endpoint's tolerances. Very large runs (10k+ tuples) will take time; budget accordingly.
• 7-day default storage retention on the Apify FREE tier. Export your dataset immediately after the run or upgrade for longer retention.
• No private / signed-in suggestions. Personalised suggestions tied to a Google account are out of scope — this Actor reflects the unauthenticated public Suggest endpoint.