Google Autocomplete Scraper avatar

Google Autocomplete Scraper

Pricing

from $1.99 / 1,000 search results

Go to Apify Store
Google Autocomplete Scraper

Google Autocomplete Scraper

Scrapes Google Autocomplete suggestions for any seed query. Extracts the full set of autocomplete fields: suggestion text + HTML, vertical (web/people/products/news/...), highlighted flag, completion, replacement, locale, client, and more.

Pricing

from $1.99 / 1,000 search results

Rating

0.0

(0)

Developer

Search API

Search API

Maintained by Community

Actor stats

0

Bookmarked

2

Total users

1

Monthly active users

9 days ago

Last modified

Share

Collect live Google autocomplete suggestions for one seed or a batch of seeds. The Actor uses Google's public structured autocomplete response first and opens a real Google textbox/listbox session only when every structured request fails.

Input

  • query: one seed phrase.
  • queries: multiple seed phrases; keywords remains a backward-compatible alias.
  • mode: exact, questions, prepositions, comparisons, alphabet, or all.
  • maxItems and maxItemsPerQuery: global and fair per-seed limits.
  • language and country: Google hl and gl localization.
  • minLength, maxLength, and containsKeyword: optional filters. Prefix the substring with ! to exclude matches.
  • outputFormat: flat suggestion records or one nested tree per seed.
  • maxConcurrency, retry, and timeout inputs: bounded performance controls.
  • useBrowserFallback: enable the actual autocomplete textbox/listbox fallback.
  • proxyConfiguration: optional Residential or custom proxy for the browser fallback.

The public autocomplete endpoint normally works directly and does not require a proxy. Apify's GOOGLE_SERP group is designed for HTTP Google Search/Shopping result requests; it cannot tunnel the Playwright fallback and is rejected there. Use Residential or custom proxy URLs only if the browser fallback needs them.

Output

Flat records include a stable ID, seed/query/request positions, suggestion and source text, Google relevance/type/subtype metadata when returned, expansion provenance, text statistics, localization, public search/source URLs, retrieval method, connection/profile information, timestamps, and normalized search metadata. Optional values Google does not return are omitted—never filled with nulls or invented data.

Tree mode returns one record per seed with the same clean flat records nested in suggestions.

The run summary is saved as OUTPUT and includes record counts, request failures, runtime, per-query counts, and dataset audits for duplicates, empty values, and malformed URLs.

Safety and limitations

  • CAPTCHA and unusual-traffic pages fail closed and are never saved.
  • The Actor does not solve CAPTCHAs or bypass authentication/access controls.
  • Google controls suggestion availability, ordering, and metadata; some seeds legitimately return an empty dataset.
  • Structured responses are status/content-type checked and malformed JSON/JSONP is rejected safely.