Google Autocomplete Scraper - Alphabet Soup Keywords

Get hundreds of long-tail keyword ideas from Google Autocomplete (Google Suggest) for any seed keyword, country and language: alphabet soup a-z, 0-9, question words, prepositions and your own modifiers, deduplicated, one row per suggestion.

What does Google Autocomplete Scraper do?

It types your seed keyword into Google's search box the way a person would, adds every letter, number or question word you choose, and collects every suggestion Google shows (up to 15 per query). One seed keyword with alphabet soup and question words returned 549 unique suggestions in 40 seconds in our test run. Requests leave from the country you pick, so a UK run shows what people in the UK search for.

No login, no Google account, no API key and no proxy setup on your side: proxies are included in the price.

Who is it for?

• SEO agencies and content marketers: build keyword lists and blog topic clusters from real searches, including the questions people ask.
• PPC teams: find long-tail and negative keywords per country before you build a Google Ads campaign.
• E-commerce and Amazon sellers: see which features, brands and comparisons buyers type after a product name.
• Product and market researchers: compare what people search for in the US, UK, Germany or any other country.

What data can you get?

Field	Example
suggestion	coffee maker with grinder
keyword	coffee maker (your seed keyword)
modifier	"" for the plain keyword, a to z, 0 to 9, how ..., vs, or your custom modifier
query	coffee maker a (exactly what was sent to Google)
position	1 (place in Google's suggestion list, 1 to 15)
relevance	601 (Google's own relevance score for the suggestion)
suggestionType	QUERY
countryCode, languageCode	us, en
wordCount	4
isQuestion	true when the suggestion starts with who, what, how, why, can, is and similar
containsKeyword	true when the suggestion contains your seed keyword
scrapedAt	2026-10-14T16:16:07Z

Every suggestion appears once per country, even when several modifiers or seed keywords return it, so you never pay twice for the same keyword.

How much does it cost?

You pay only for unique suggestions: $0.80 per 1,000 suggestions on the Free plan, with lower prices on paid Apify plans. Proxies and platform usage are included.

Examples on the Free plan:

• 1 seed keyword with alphabet soup and question words, about 550 suggestions: about $0.44.
• 200 suggestions (the prefilled limit): $0.16.

Set Max suggestions or a maximum cost per run to cap spending.

How to use it

1. Click Try for free.
2. Enter one or more seed keywords, for example coffee maker.
3. Pick the countries (us, gb, de ...) and the modifiers: alphabet soup and question words are on by default; numbers, prepositions and custom modifiers are optional.
4. Click Start and download the keywords as JSON, CSV, Excel or via API.

Input example

{
  "keywords": ["kaffeemaschine", "laufschuhe"],
  "countryCodes": ["de"],
  "expandAlphabet": true,
  "expandQuestions": false,
  "expandNumbers": true,
  "maxItems": 300
}

For German, French, Spanish, Turkish and other languages the alphabet soup also adds the local letters (for example ä, ö, ü). Custom modifiers can place the keyword anywhere with {keyword}, for example best {keyword} for.

Output example

Real rows from our test runs on 2026-10-14 (UK, question words on):

[
  {
    "suggestion": "why standing desk is better",
    "keyword": "standing desk",
    "modifier": "when ...",
    "query": "when standing desk",
    "position": 8,
    "relevance": 553,
    "suggestionType": "QUERY",
    "countryCode": "gb",
    "languageCode": "en",
    "wordCount": 5,
    "isQuestion": true,
    "containsKeyword": true
  },
  {
    "suggestion": "how long should you stand at a standing desk",
    "keyword": "standing desk",
    "modifier": "when ...",
    "query": "when standing desk",
    "position": 10,
    "relevance": 551,
    "suggestionType": "QUERY",
    "countryCode": "gb",
    "languageCode": "en",
    "wordCount": 9,
    "isQuestion": true,
    "containsKeyword": true
  }
]

Germany, plain keyword:

{
  "suggestion": "kaffeemaschine mit mahlwerk",
  "keyword": "kaffeemaschine",
  "modifier": "",
  "query": "kaffeemaschine",
  "position": 1,
  "relevance": 601,
  "suggestionType": "QUERY",
  "countryCode": "de",
  "languageCode": "de",
  "wordCount": 3,
  "isQuestion": false,
  "containsKeyword": true
}

Integrations and API

Run it from your code with the Apify API or the apify-client package, schedule it in Apify, or connect it to Make, n8n, Zapier, Google Sheets or a webhook.

from apify_client import ApifyClient

client = ApifyClient("<YOUR_APIFY_TOKEN>")
run = client.actor("kamerozkan/google-autocomplete-scraper").call(run_input={
    "keywords": ["coffee maker"], "countryCodes": ["us"], "maxItems": 200})
for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item["suggestion"])

FAQ

What is alphabet soup keyword research? You type your keyword followed by every letter (coffee maker a, coffee maker b, ...) and write down what Google suggests. This Actor does it for all letters, digits and question words in one run and removes duplicates.

Why do results differ by country? Google suggests what people in that country search for, based on both the country setting and where the request comes from. Each country in countryCodes is queried from that country.

Do I get search volume? No. Google Autocomplete does not publish search volume. position and relevance show how strongly Google ranks a suggestion; use a volume tool on the final list if you need numbers.

Some suggestions do not contain my keyword. Why? Google sometimes suggests related searches. Filter on containsKeyword if you only want exact variations.

Is it legal to collect this data? It collects publicly available search suggestions only, no personal data. You are responsible for how you use it.

Support and feedback

• Found a bug or missing a field? Open an issue on the Issues tab with your run link. We reply within 48 hours.
• Have an idea for a feature (for example YouTube or Amazon suggestions)? Issues are the fastest way to get it built.