Google Autocomplete Scraper — Google, YouTube, Amazon & Bing

🇨🇳 谷歌自动补全关键词采集器：批量获取 Google、YouTube、Amazon、Bing、Google 购物和 DuckDuckGo 的搜索联想词，支持疑问词与 A–Z 扩展，适用于长尾关键词研究。支持任意国家和语言，无需 API 密钥，按建议词付费。

🇸🇬 Pengikis Autolengkap Google: dapatkan cadangan kata kunci pukal daripada autolengkap Google, YouTube, Amazon, Bing dan DuckDuckGo, dengan pengembangan soalan dan A–Z untuk penyelidikan kata kunci ekor panjang. Tiada kunci API.

🤖 NEW: Connect This To Your AI!

Use this actor from Claude, ChatGPT, Cursor or any MCP client through the IntelScrape MCP Server, or call it directly over the Apify API.

🏆 Featured Bots by IntelScrape

• Skip Trace Pro
• TruePeopleSearch Scraper
• TikTok Email Scraper
• Amazon Product Reviews Scraper
• Website Contact Scraper
• Skip Trace MCP Server

⚡ Works with n8n, Make, Zapier, Google Sheets and the Apify API.

Bulk keyword ideas from Google, YouTube, Amazon, Bing, Google Shopping and DuckDuckGo autocomplete, deduplicated and labeled, in one run.

Honest coverage & billing: Autocomplete reflects popular queries, not search volume; suggestions can change daily. You pay only for results actually delivered, and a built-in cost guard stops the run at your max-charge limit.

🌍 Built for teams in the US, China and Singapore, and anyone who needs this data in a spreadsheet or pipeline.

⭐ Found it useful? Please leave a review; it helps others find it.

⚖️ Lawful business use only. Collect public data only, and comply with the source site's terms, copyright and data-protection laws (GDPR, CCPA, China PIPL, Singapore PDPA).

Examples below are illustrative; names and values may be fictional or out of date.

What is Google Autocomplete Scraper?

Autocomplete is the closest thing to a live feed of what people type into a search box. Every suggestion Google, YouTube or Amazon shows is a query that real users search often enough to be predicted. Google Autocomplete Scraper turns a list of seed keywords into hundreds of real long-tail keywords per seed, from up to six engines at once, with the question and A–Z expansion that SEO tools charge a monthly subscription for.

What does it do?

• Six autocomplete sources: Google Search, YouTube, Amazon, Bing, Google Shopping and DuckDuckGo; pick any combination
• Question & modifier expansion (on by default): queries how / what / why / when / where / who / which / can / is / are / does / best / vs / for / without / near me variants of each seed
• A–Z expansion: "seed a" … "seed z" for the deepest long-tail list
• Any market: set country and language (e.g. DE + de, BR + pt, JP + ja); Amazon supports US, UK, DE, FR, CA, IN, IT and ES
• Clean output: one row per unique suggestion per seed and engine, with its position, the exact query that produced it, word count, and question flag
• Self-healing: if a source rate-limits, the actor retries and switches to Apify Proxy automatically

Why choose us over cheaper scrapers?

• Honest output: limits are documented, missing data is labeled, never invented
• Clean, consistent schema ready for Excel, Sheets, BI tools or an LLM
• No account risk: no logins, no cookies, no API keys
• Pay per result with a built-in cost guard; no subscription, failed items are never charged
• Fast and light: no browser, no residential proxy, so runs finish in seconds and stay cheap
• Maintained by IntelScrape and fixed fast when the source changes

Who is it for?

• SEO content planning: find the long-tail and question keywords to target with blog posts, FAQs and landing pages
• YouTube creators: discover what viewers actually search on YouTube before choosing titles and tags
• Amazon sellers & PPC: harvest Amazon search terms for listings, backend keywords and Sponsored Products campaigns
• PPC managers: build keyword and negative-keyword lists for Google and Bing Ads
• Market and product research: see how people describe a product category in each country
• AI agents and automation: feed keyword ideas into n8n, Make, Zapier or an LLM pipeline

How to use it

In Apify Console (no code)

1. Click Try for free / Start. The form is prefilled with a working example.
2. Change the inputs (see examples below).
3. Click Start, then export the results as JSON, CSV, Excel, XML or HTML, or open them in Google Sheets.
4. Optional: Schedule the actor to run hourly, daily or weekly.

Via API

curl -X POST "https://api.apify.com/v2/acts/intelscrape~google-autocomplete-scraper/run-sync-get-dataset-items?token=YOUR_APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"keywords": ["coffee maker"], "engines": ["google", "youtube", "bing", "amazon"], "questionExpansion": true, "country": "US", "language": "en"}'

Or use the official Apify JavaScript and Python clients:

from apify_client import ApifyClient
client = ApifyClient("YOUR_APIFY_TOKEN")
run = client.actor("intelscrape/google-autocomplete-scraper").call(run_input={"keywords": ["coffee maker"], "engines": ["google", "youtube", "bing", "amazon"], "questionExpansion": true, "country": "US", "language": "en"})
for item in client.dataset(run["default_dataset_id"]).iterate_items():
    print(item)

With AI agents (MCP)

Add Apify's MCP server (https://mcp.apify.com) or the IntelScrape MCP Server to Claude Desktop, Cursor, ChatGPT or any MCP client, then ask in plain language, e.g. "Run intelscrape/google-autocomplete-scraper with the default input and summarise the results."

Input examples

One keyword, four engines (default)

{
  "keywords": ["coffee maker"],
  "engines": ["google", "youtube", "bing", "amazon"],
  "questionExpansion": true,
  "country": "US",
  "language": "en"
}

Deep long-tail with A–Z expansion

{
  "keywords": ["running shoes", "trail running"],
  "engines": ["google"],
  "questionExpansion": true,
  "alphabetExpansion": true
}

Amazon Germany only

{
  "keywords": ["kaffeemaschine"],
  "engines": ["amazon"],
  "country": "DE",
  "language": "de"
}

Cap cost on a big list

{
  "keywords": ["crm software", "email marketing", "seo tools"],
  "engines": ["google", "bing"],
  "maxSuggestionsPerSeed": 50
}

Output fields

Field	Description
seedKeyword	The seed keyword you entered
engine	google, youtube, amazon, bing, google-shopping or duckduckgo
suggestion	The autocomplete suggestion
sourceQuery	The exact text sent to the engine (seed or expanded variant)
expansionType	seed, question/modifier or alphabet
position	Position of the suggestion in that dropdown (1 = top)
wordCount	Number of words in the suggestion
isQuestion	true when the suggestion starts with a question word
containsSeed	true when the suggestion contains the seed keyword
country / language	Market used for the request
scrapedAt	Time of collection (ISO)

Sample output

{
  "seedKeyword": "coffee maker",
  "engine": "google",
  "suggestion": "how coffee maker works",
  "sourceQuery": "how coffee maker",
  "expansionType": "question/modifier",
  "position": 1,
  "wordCount": 4,
  "isQuestion": true,
  "containsSeed": true,
  "country": "US",
  "language": "en",
  "scrapedAt": "2026-10-21T03:07:18.596Z"
}

Pricing (pay per event)

Event	When it is charged	Price
Actor start	once per run	$0.005
Suggestion	per unique suggestion row	$0.0002

Cost examples

Run	Cost
Default run: 1 keyword × 4 engines with question expansion (~450–500 suggestions)	≈ $0.10
10 keywords, Google only, no expansion (~100 suggestions)	≈ $0.025
100 keywords × Google + YouTube with question expansion (~25,000 suggestions)	≈ $5.00

Platform usage is included in these prices. Set Max charge per run in the run options to cap spend; the actor stops cleanly when the limit is reached and keeps everything collected so far.

FAQ

Do I need a Google, Amazon or Bing API key?

No. The actor reads the same public autocomplete endpoints your browser uses. There is nothing to sign up for.

How many suggestions do I get per keyword?

Each request returns up to 10 suggestions (Amazon up to 11). With question expansion each seed sends 17 requests per engine, so expect roughly 60–170 unique suggestions per seed per engine. A–Z expansion adds 26 more requests.

Why are some engines returning fewer results?

Autocomplete only shows queries with enough search history. Niche or very new seeds get fewer suggestions, and Amazon and YouTube suggest less for non-shopping or non-video topics. You are charged only for rows delivered.

Does it give search volume?

No; autocomplete has no volumes. Send the suggestions to our Keyword Search Volume actor to estimate monthly searches and trends.

Can I use other countries and languages?

Yes. Set country (two letters) and language. Google, YouTube, Bing and DuckDuckGo work in almost every market. Amazon covers US, UK, DE, FR, CA, IN, IT and ES; other countries fall back to Amazon US.

Are duplicates removed?

Yes, within each seed and engine. The same phrase can appear for two engines, which is useful to see where a query is popular.

Is the order meaningful?

position is the rank inside the dropdown for that exact query, which roughly reflects popularity. It is personalised less than normal search because no cookies or login are used.

Can I schedule it?

Yes. Schedule weekly runs and compare datasets to catch new trending suggestions early.

Honest limits

• Autocomplete reflects popular queries, not search volume; suggestions can change daily.
• Engines may filter sensitive, adult or trademark-related suggestions.
• Very large runs can be slowed by rate limits; the actor retries and can switch to Apify Proxy.
• DuckDuckGo suggestions are based on Bing data, so expect overlap with Bing.

Related IntelScrape bots

• Keyword Search Volume: estimated monthly Google searches & trends
• AI Search Visibility Tracker: brand visibility in AI answers
• Google News Scraper: headlines & summaries
• Apple App Store Scraper: iOS app details, ratings & charts
• Website Contact Scraper: emails, phones & socials from sites
• Skip Trace MCP Server: IntelScrape tools for AI agents

Legal

This actor collects publicly available data only. You are responsible for using the results lawfully and in line with the source's terms and applicable privacy laws (GDPR, CCPA, PIPL, PDPA). Process personal data only with a legitimate purpose. IntelScrape is not affiliated with, endorsed by or sponsored by the data source; all trademarks belong to their owners.