Google Keyword Suggestions Scraper - Autocomplete Data

Collect the public keyword ideas Google shows while a user types a search. Give the Actor one or more seed queries and it returns ranked autocomplete suggestions with the seed, language, country context, rank, source, and collection time. The output is ready for keyword research, content planning, search-intent discovery, marketplace taxonomy work, and repeatable SEO research.

The Actor uses Google's public suggestion response rather than opening a full browser or scraping a search-results page. That keeps runs fast, predictable, and inexpensive. No Google account, cookie export, or external API key is required. Results are written to the default Apify dataset and can be exported as JSON, CSV, Excel, XML, or RSS, fetched through the API, or sent into an automation workflow.

What this Actor returns

Each dataset item is one autocomplete suggestion. The result preserves the seed query that produced it, so several seeds can be processed in one run without losing provenance.

Typical fields include:

• seed_query: the word or phrase submitted to the suggestion service;
• suggestion: one public autocomplete phrase returned for that seed;
• rank: the suggestion's order in the response, starting at one;
• language: the Google interface-language context used for the request;
• country: the country context used for the request;
• source: Google Autocomplete for clear provenance;
• scraped_at: the UTC collection timestamp.

The dataset schema displays the most useful fields as a table in Apify Console, while the complete records remain available in every supported export format.

Common use cases

Long-tail keyword discovery

Start with a broad commercial or informational phrase and collect the longer queries people may type around it. This is useful when creating a first topic map or expanding an existing keyword list.

Content briefs and editorial planning

Use suggestion phrases as research inputs for article outlines, FAQ sections, video topics, documentation pages, or support-center content. Autocomplete is a discovery signal, not a complete content strategy, so combine it with business relevance and firsthand subject knowledge.

Product and marketplace taxonomy

Teams can compare the language used around product categories, services, locations, or problems. The seed-to-suggestion relationship makes it easy to group phrases into navigation labels, filters, or internal search synonyms.

Localized research

Run the same seeds with different language and country settings to observe how public suggestions vary by market. Country and language parameters influence the request context but do not guarantee that every returned phrase is unique to that location.

Scheduled change monitoring

Save a task with a stable list of seeds and run it weekly or monthly. Compare datasets by seed_query, suggestion, and rank to identify newly appearing or disappearing phrases.

Input

The default input is intentionally small so a first run finishes quickly and produces a useful dataset.

{
  "queries": ["web scraping", "data extraction api"],
  "language": "en",
  "country": "us",
  "maxSuggestionsPerQuery": 10
}

Input fields

• queries: a list of seed words or phrases. The Actor accepts up to 100 non-empty seeds in one run.
• language: a short Google interface-language code such as en, de, es, or fr.
• country: a short country code such as us, gb, de, or es.
• maxSuggestionsPerQuery: the maximum number of ranked rows saved for each seed, from 1 to 50. Google's response may contain fewer rows.

Keep related seeds together when the resulting dataset will feed one project. Split unrelated markets into separate tasks when you need cleaner scheduling, cost tracking, or change comparisons.

Example output

{
  "seed_query": "web scraping",
  "suggestion": "web scraping in python",
  "rank": 4,
  "language": "en",
  "country": "us",
  "source": "Google Autocomplete",
  "scraped_at": "2026-07-10T12:00:00+00:00"
}

Suggestions are public phrases returned at collection time. A rank is only the order in that response; it is not search volume, competition, CPC, trend growth, or a guarantee of search demand. The Actor does not invent those metrics.

Pricing and cost control

This Actor uses pay per event pricing:

• a small Actor-start event is charged once per run;
• the result event is charged once for each suggestion written to the default dataset.

The visible maximum-per-query input makes the upper result count predictable. For example, 20 seeds with a limit of 10 can write at most 200 rows, although the source may return fewer. Apify platform usage is shown separately by Apify.

Run from the API

After creating an Apify API token, call the Actor through the standard Actor run endpoint. Keep the token in an environment variable or secret manager rather than embedding it in source code.

curl -X POST \
  "https://api.apify.com/v2/acts/benthepythondev~google-keyword-suggestions-scraper/runs?token=$APIFY_TOKEN" \
  -H "content-type: application/json" \
  -d '{"queries":["remote jobs","python developer"],"language":"en","country":"gb","maxSuggestionsPerQuery":10}'

The run response contains links to the default dataset. You can also create an Apify task for a saved input, schedule that task, and use a webhook when the run finishes.

Automation ideas

• Append weekly suggestions to Google Sheets and compare rank changes.
• Send new phrases into Notion or an editorial backlog after deduplication.
• Group suggestions by seed before passing them to an internal research or RAG workflow.
• Combine phrase discovery with a separate news or content source to validate whether a topic has timely material.
• Use separate tasks for countries so costs and outputs remain attributable.

Reliability and responsible use

The Actor retries transient request failures and fails the run explicitly when no suggestions are returned. An empty successful dataset would hide source or input problems, so it is treated as an error. Defaults are bounded for daily auto-testing, and the Actor requests only the public suggestion response needed for the supplied seeds.

Autocomplete phrases can include sensitive, inaccurate, or unexpected wording. Review results before publication or automated decision-making. Do not treat a suggestion as an endorsement by Google, proof of a person's intent, or a factual claim. Respect applicable search-provider terms, privacy rules, and the laws in your jurisdiction.

Limitations

• The source may change suggestions between runs.
• Language and country settings influence context but do not simulate every user's exact location, history, or personalization.
• The Actor does not return search volume, CPC, keyword difficulty, SERP rankings, or trends.
• A phrase can appear for multiple seeds; that relationship is intentionally preserved rather than globally deduplicated.
• Google may return fewer suggestions than the configured maximum.

Frequently asked questions

Does this need a Google API key?

No. It uses the public autocomplete response and does not require a Google account or external API credential.

Can I use several seed queries?

Yes. Add them to queries; each row records its original seed_query.

Can I research German or Spanish phrases?

Yes. Set the language and country codes for the market you want to examine. The output remains UTF-8 and can contain non-English text.

Does rank equal search volume?

No. Rank is only response order. Use a separate licensed keyword-data provider when exact volume or advertising metrics are required.

Can I schedule it?

Yes. Save the input as an Apify task, attach a schedule, and consume the dataset through exports, API calls, webhooks, or integrations.

What should I do if a run returns no data?

Check that the seeds are non-empty and use common language/country codes. The Actor reports a failed run rather than silently succeeding with an empty dataset. If valid inputs repeatedly fail, open an Actor issue with the run ID.

Related Actors

• Google News Scraper for articles and source monitoring by query.
• Website Contact Extractor for public business contact research from websites.
• Webpage Text Extractor for clean text from pages used in research pipelines.

If this Actor saves time in a real workflow, a short honest Store review helps other users evaluate it. For a reproducible problem, open an issue with the run ID, non-secret input, expected behavior, and the affected seed query.

Keywords: Google keyword suggestions scraper, Google autocomplete scraper, keyword ideas API, long-tail keyword research, SEO keyword scraper, search query suggestions, autocomplete data export, content research API, Apify SEO Actor.