Keyword Autocomplete Intelligence — Google & Bing Suggest, A–Z Long-Tail

Keyword research costs $68–788/month at KeywordTool.io and $29–1,499/month at Ahrefs — for data that comes from public autocomplete routes. This Actor turns a list of seed keywords into hundreds of structured long-tails: A–Z + digit expansion, custom modifiers, multi-language (hl/gl), across Google, Bing and YouTube autocomplete. You pay only for the keywords written. No API key, no login, no browser, no LLM.

Not affiliated with, sponsored by, or endorsed by Google or Microsoft. All data comes from public autocomplete endpoints and remains subject to those providers' terms.

Who it's for:

• SEO / content teams: harvest the long-tail around a topic ("crm software" → "crm software for nonprofits", "crm software vs spreadsheet") in one run, straight to Sheets or your pipeline via Apify integrations.
• PPC / marketplace sellers: real buyer phrasing per country and language, not a scraped guess.
• Market researchers: what people actually type — including localized demand (hl=pt&gl=br returns "melhor crm para whatsapp").

How it works

You provide seed keywords. Default mode is expand: each seed becomes the seed itself plus seed a…seed z and seed 0…seed 9 (plus any custom modifiers, applied as prefix and suffix) — ~37 queries per seed, yielding ~260+ unique long-tails (measured). suggest mode fetches one direct query per seed instead. Results are deduplicated per source, and every run writes one free RUN_SUMMARY record — so even a run where every seed returns nothing still yields a non-empty dataset confirming coverage.

You are only charged for keywords actually written — duplicates, filtered variants and anything past the cap cost you nothing.

Input

Copy-paste ready (long-tail harvest, Google + Bing):

{
  "seeds": ["crm software", "email marketing"],
  "mode": "expand",
  "sources": ["google", "bing"]
}

Field	Type	Default	Description
seeds	array	required	1–100 seed keywords
mode	string	expand	expand (A–Z long-tail harvest) or suggest (direct only)
sources	array	["google"]	google, bing, youtube. Prefill ships Google + Bing for redundancy — see below
language	string	en	hl code (en, pt, de…)
country	string	us	gl code (us, br…)
modifiers	array	[]	Extra terms as prefix AND suffix (e.g. ["how","best","vs"]); max 20
maxResults	integer	5000	Cap on charged keywords; the free summary doesn't count
requestDelayMs	integer	250	Polite pacing before each request (floor 100 ms, deliberate)

Why the prefill includes Bing. These are public, undocumented routes. If one provider changes its response shape, the Actor flags API_CONTRACT_CHANGED for that source and keeps going on the other — a run stays partial-but-useful instead of failing whole. Google is the product default (omit sources); Bing in the prefill is the daily-run safety net.

Output

One record per keyword (recordType: "KEYWORD"):

{
  "recordType": "KEYWORD",
  "keyword": "crm software for nonprofits",
  "seed": "crm software",
  "variantQuery": "crm software f",
  "position": 3,
  "source": "google",
  "language": "en",
  "country": "us",
  "detectedAt": "2026-07-30T13:00:00.000Z"
}

Plus one free RUN_SUMMARY per run (statusCounts, keywordsWritten, duplicatesDiscarded, and a per-seed×source status array).

Schedule it (recommended — cloud, not your desktop)

This Actor is built to run on a schedule. Use Apify's own Schedules, not a local scheduler — configure it once and it runs in the cloud whether or not your machine is on.

1. Save your input as a Task (Console → the Actor → Create task), e.g. your niche's seed list.
2. Console → Schedules → Create schedule, add the Task, set the cron (e.g. weekly Monday 6am → 0 6 * * 1).
3. Point downstream (Sheets, webhook, your DB) at the dataset via Apify integrations.

Weekly is a good cadence for tracking how the long-tail around your topics shifts over time.

Honest limits

• Public but undocumented routes. Google/Bing can change the response shape without notice; the Actor detects it (API_CONTRACT_CHANGED) and reports rather than emitting garbage. Running Google + Bing together means one shape change never empties your run.
• Autocomplete returns ~10 suggestions per query — depth comes from expansion (A–Z, digits, modifiers, languages), not from a single call.
• Suggestions are demand signals, not volumes. This Actor does not provide search volume (no public route exposes it honestly); pair it with a volume source if you need that.
• Non-English hl responses arrive in ISO-8859-1; the Actor decodes by the response charset, so accents come through clean (imobiliária, not mojibake).
• Compliance: suggestqueries.google.com publishes no robots.txt (404) and google.com/robots.txt does not list /complete (verified 2026-07-30) — recorded as fact, not as permission; the providers' general terms still apply. Pacing is polite and identified; a source that walls us off (redirect/403) becomes a controlled SOURCE_BLOCKED, never bypassed.

Ops notes

• STATS key in the run's key-value store: per-seed×source summary, HTTP counters, dedup counts, field-completeness health check, warnings. ERRORS on failure.
• Cost drivers: requests made (≈37 per seed per source in expand) + keywords written. A 2-seed expand over Google runs in well under a minute at 512 MB.

FAQ

Do I need a Google or Bing API key? No. The Actor reads public autocomplete endpoints without a key and without logging in.

Does it return search volume? No. Autocomplete exposes demand signals — what people actually type — not volumes, and no public route exposes volume honestly. Pair this Actor with a volume source if you need that number.

What exactly am I charged for? Per keyword written to the dataset (Pay Per Event) — duplicates, filtered variants and anything past the cap cost you nothing, and the RUN_SUMMARY record is free. The Pricing tab on this page is always the authoritative source for current rates and for any per-run fee.

Can I schedule it? Yes — that is the intended use. See "Schedule it" above.

Is this affiliated with Google or Microsoft? No. This is an unofficial community Actor; all data comes from public autocomplete endpoints and remains subject to those providers' terms.