YouTube Autocomplete Scraper: keyword research, A to Z

YouTube keyword research with autocomplete: YouTube's own search suggestions for up to 500 seed keywords per run, per country and language, with optional a-to-z and question-word expansion for hundreds of long-tail video keywords per seed. Ranked, deduplicated. Pay per keyword.

Why this scraper

• Only delivered results are charged. Inputs that fail come back as clear error records at no cost.
• The endpoint the YouTube search box itself uses, half a kilobyte per call: a seed with question-word expansion costs eleven small requests. Measured on the platform: three seeds, 109 suggestions, for fifteen hundredths of a cent.
• YouTube's own suggestions in YouTube's own order, not the Google web mix: what people type into YouTube in a given country and language. A-to-z expansion, ten question patterns and your own prefixes or suffixes turn one seed into hundreds of long-tail video keywords, de-duplicated and ranked in the order they surfaced. Suggestions that do not carry the seed are dropped, so an expansion never delivers unrelated rows.

Who this is for

Paste seed keywords in keywords (up to 500 per run), set country and language, switch on expandAlphabet or expandQuestions or add customPrefixes (use * for the seed position), and set maxSuggestionsPerKeyword (default 200). Every row carries the seed, the exact query that surfaced the suggestion, the rank, the suggestion, country and language.

Who this is not for

YouTube returns at most about fourteen suggestions per query, so the numbers come from expansion, not from one call. The question patterns are English; for other languages set your own via customPrefixes. YouTube completes almost any prefix, so a seed nobody searches for still yields a few rows built on that seed. There are no search volumes here: pair the output with the Google Trends actors below for demand.

Input fields

Field	Type	Required or default	What it does
keywords	list of text	required	One seed keyword per row (up to 500). A seed without suggestions comes back as a free error record.
country	text	US	Two-letter country code that sets the market (gl), e.g. US, GB, NL, DE.
language	text	en	Interface language (hl), e.g. en, nl, de. Suggestions follow the language people type in that market.
expandAlphabet	true/false	false	Also query '
expandQuestions	true/false	false	Also query 'how
customPrefixes	list of text	empty	Your own expansions, one per row. Use * for the seed position, e.g. 'beste ' or ' tutorial'. Without * the text is used as a suffix.
maxSuggestionsPerKeyword	number	200	Cost ceiling per seed across all expansions.

Input example

{
    "keywords": [
        "how to edit video",
        "air fryer recipe"
    ],
    "country": "US",
    "language": "en",
    "expandAlphabet": false,
    "expandQuestions": false,
    "customPrefixes": [],
    "maxSuggestionsPerKeyword": 200
}

Output example

Field	Type	What it holds	Example
keyword	text	The seed you asked for; every suggestion row traces back to one seed.	how to edit video
query	text	The string that was really sent to YouTube: the seed itself, or the seed widened with a letter or a question word.	how to edit video
rank	number	The place of the suggestion in YouTube's list, counting from 1; the first is what most people pick.	1
suggestion	text	The completion YouTube offers, which is what people actually type there.	how to edit videos for youtube
country	text	The storefront the suggestions were read from; they differ per country.	US
language	text	The interface language the suggestions were read in.	en

Error codes: INVALID_KEYWORD, NO_SUGGESTIONS, BLOCKED.

One delivered row looks like this:

{
  "keyword": "air fryer",
  "query": "air fryer",
  "rank": 1,
  "suggestion": "air fryer recipes",
  "country": "US",
  "language": "en",
  "status": "ok"
}

Related actors from steadydata

• google-autocomplete: the same expansion on Google web search
• youtube-search: the videos that rank for the suggested keywords
• google-trends-compare: which suggestion is searched for more

Pricing

Pay per event: one suggestion-listed event per delivered result. No charge for inputs that fail, no separate platform-usage surcharge.

Free Apify plan: this actor delivers up to 25 rows per run for accounts on the Apify free plan, and then stops with a message. That limit is set by us, not by Apify. It exists so the actor keeps paying for itself for the people who do pay. Any paid Apify plan runs it at full size, billed per delivered row, with failed rows never charged.

Reviews: if this actor saves you time, a short review on this page is the one thing that helps most. Ratings are what other buyers look at first, and we have no other way to ask.

Use with AI agents

AI assistants such as Claude, ChatGPT or Cursor can run this actor through Apify's own MCP server. Ask for the actor steadydata/youtube-autocomplete by its full name, so the assistant does not pick a different one, and start from this minimal input:

{
  "keywords": [
    "how to edit video",
    "air fryer recipe"
  ]
}

The assistant can read the price per event from the Pricing section above before the first paid run. Failed input comes back as an error row, as described above.

FAQ

How is this different from the Google Autocomplete actor? Same expansion logic, different source: this one asks YouTube's search box, so the suggestions are video searches.

Can I get search volume? Not from autocomplete; YouTube publishes none. Use Google Trends Compare on the suggestions that matter.

Which country and language should I set? The market you make videos for: country sets the storefront (gl), language the interface language (hl).

How many rows per seed? Around ten for the seed alone, up to a few hundred with a-to-z and question expansion, capped by maxSuggestionsPerKeyword.

Do I pay for a seed without suggestions? No, that is a free error row.

Is personal data collected? No. Only the search suggestions YouTube shows for a prefix; nothing about who searched.

What do the fields mean? query is the exact prefix sent to YouTube (for example air fryer c or how air fryer); rank counts across all expansions in the order the suggestions surfaced.

What happens when the source changes? Sources change from time to time; that is the nature of this work. The actor is monitored daily and fixed fast, and while it is broken you are not charged, because only delivered results cost anything.

Can I call it from my own code, like an API? Yes. Every run is available through the Apify API: start it with one HTTP request, then read the results as JSON, CSV or Excel from the dataset endpoint, or use the official Python and JavaScript clients. The input is the same JSON as in the input example above, so a scheduled script, a Make or Zapier step or an AI agent can call it without any changes.