Google Autocomplete API
The Google Autocomplete API for developers
Get the raw suggestion strings Google offers as a query is typed — up to 10 per call, localized by language — at half the cost of a full search: 0.5 credits per query. It returns suggestion text only, with no search-volume or CPC data attached, so pair it with a keyword-volume source rather than expecting one.
{ "searchTime": 0.17, "totalResults": 85000, "page": 1, "organic": [ { "position": 1, "title": "31.10. Connection Pools and Data Sources", "link": "www.postgresql.org/docs/…", "domain": "postgresql.org", "snippet": "For an environment without an application se…" }, { "position": 2, "domain": "medium.com", … }, { "position": 3, "domain": "stackoverflow.blog", … } ], "credits": 1}What does a Google Autocomplete API return, and does it include search volume?
No — and that's worth saying plainly. Searlo's /search/suggestions endpoint returns the same suggestion strings Google shows as you type, up to 10 per call, for 0.5 credits per query ($0.15 per 1,000). The response is raw suggestion text: no search volume, no CPC, no trend score. It's a keyword-expansion input, not a keyword-research suite.
That distinction matters because the phrase "autocomplete API for keyword research" gets used for a real, specific technique — using suggestions as a seed-expansion tool, not as a volume source — and this page is honest about which half of that job the endpoint does.
- Cost per query
- 0.5 credits (=$0.15 per 1,000)— our pricing
- Free to start
- 3,000 credits (~6000 autocomplete calls), no card— our pricing
- Suggestions per call
- Up to 10 (limit, default 10)
- What's NOT in the response
- Search volume, CPC, or a trend score — suggestion text only
- SerpAPI Google Autocomplete
- Sold as a separate engine; $15.00 per 1,000 on the Developer tier applies uniformly across SerpAPI's engines— serpapi.com/pricing
Third-party figures on this page last verified against the sources linked above. Prices change; if you find one of these stale, tell us and we will correct it.
"Google autocomplete api" pulls a real, distinct audience: SerpAPI (serpapi.com/google-autocomplete-api) and DataForSEO (dataforseo.com's Google Autocomplete SERP API, plus a dedicated blog post on using it for keyword research) both sell this as its own product, separately from their general search APIs — and the people searching for it skew toward SEO and content teams building keyword lists, not toward developers who just want search results.
Searlo's /search/suggestions endpoint is a thin, honest match for that: one GET request with a seed query (q), a result cap (limit, up to 10), and a language (hl) returns the literal array of strings Google's own autocomplete would show. It costs 0.5 credits per query — half the rate of a full search — because there is less work behind a single suggestion call than a full result page.
What it deliberately doesn't do is estimate demand. There's no volume number, no CPC, no trend line in the response — just the suggestion text Google offers. That's a real limitation worth stating up front rather than discovering after integrating: this endpoint expands a seed term into more terms; it doesn't tell you which of them matter.
Features
Same suggestions Google shows
The literal autocomplete list Google renders as a query is typed, returned as a JSON array of strings.
Seed-expansion at scale
Loop one query through the alphabet as a suffix and you turn one seed term into hundreds of real long-tail phrases.
Localized by language
hl sets the language Google localizes suggestions to, so results match the market you're targeting.
Half-price queries
0.5 credits per call instead of the usual 1 — the lightest-weight endpoint on the API.
$0.15 per 1,000 queries
Half of Searlo's $0.30 headline rate, because a suggestions call costs half a credit.
Get started in one request
One GET request returns an array of suggestion strings for one seed query. The pattern below — the "alphabet soup" technique — turns one seed into a long list by looping the same call through a suffix per letter.
curl -G "https://api.searlo.tech/api/v1/search/suggestions" \
-H "x-api-key: YOUR_API_KEY" \
--data-urlencode "q=best running shoes for" \
--data-urlencode "limit=10" \
--data-urlencode "hl=en"import string
import requests
API_KEY = "YOUR_API_KEY"
BASE = "https://api.searlo.tech/api/v1/search/suggestions"
def suggestions(q, limit=10, hl="en"):
resp = requests.get(
BASE,
headers={"x-api-key": API_KEY},
params={"q": q, "limit": limit, "hl": hl},
)
return resp.json()["data"]["suggestions"]
# One call.
print(suggestions("best running shoes for"))
# "Alphabet soup" expansion: one seed term becomes up to 260 long-tail
# phrases (26 letters x up to 10 suggestions each), deduplicated.
seed = "best running shoes for"
expanded = set()
for letter in string.ascii_lowercase:
for s in suggestions(f"{seed} {letter}"):
expanded.add(s)
print(f"{len(expanded)} unique long-tail phrases from one seed term")const API_KEY = "YOUR_API_KEY";
const BASE = "https://api.searlo.tech/api/v1/search/suggestions";
async function suggestions(q, limit = 10, hl = "en") {
const params = new URLSearchParams({ q, limit: String(limit), hl });
const resp = await fetch(`${BASE}?${params}`, {
headers: { "x-api-key": API_KEY },
});
const data = await resp.json();
return data.data.suggestions;
}
// "Alphabet soup" expansion: one seed term becomes many long-tail phrases.
const seed = "best running shoes for";
const expanded = new Set();
for (const letter of "abcdefghijklmnopqrstuvwxyz") {
const results = await suggestions(`${seed} ${letter}`);
results.forEach((s) => expanded.add(s));
}
console.log(`${expanded.size} unique long-tail phrases from one seed term`);Searlo vs SerpAPI
SerpAPI and DataForSEO both sell autocomplete as its own product. Searlo prices it as a lighter-weight call within the same API, at half the cost of a full search.
| Capability | Searlo | SerpAPI |
|---|---|---|
| Price per 1,000 queries | $0.15 | $15.00 (Developer tier, uniform across engines) |
| Free credits to start | 3,000 (~6000 calls) | 250/month |
| Suggestions per call | Up to 10 | Not verified |
| Search volume / CPC included | No — suggestion text only | Not verified |
The alphabet-soup technique, honestly
Google's autocomplete only ever returns a handful of suggestions per call — this endpoint caps at 10, matching Google's own UI. The way SEO tooling gets real keyword volume out of that constraint is by looping: take a seed phrase, append every letter of the alphabet (or every two-letter combination — "aa", "ab", ... — for a seed that's already common), and call suggestions once per suffix. Twenty-six calls against a-z can turn one seed into over a hundred distinct long-tail phrases once you deduplicate.
This is real, substantive work an API makes practical — twenty-six manual searches in Google's own search box would take longer than the twenty-six API calls, and the API calls are trivially scriptable into a batch job. It is not, however, a shortcut to search-volume data: the alphabet-soup technique tells you what people type, not how often. Pair its output with a volume or trend source before prioritizing which of the resulting phrases to target.
- Single-letter suffixes (a–z) for a rare seed term
- Two-letter prefixes (aa, ab, ...) for a seed that already returns 10 generic suggestions
- Question-word prefixes ("how", "why", "can", "does") for content-ideation angles instead of alphabet order
- Always deduplicate — the same long-tail phrase surfaces under multiple suffixes
What this endpoint is not
It's not a keyword-research suite, and this page won't pretend otherwise. There is no search-volume figure, no CPC estimate, no difficulty score anywhere in the response — just the suggestion strings themselves, exactly as Google's own autocomplete dropdown would show them. If a project's requirement is "how many people search this per month," this endpoint answers a different question and needs to be paired with a real volume data source.
What it is good at: generating the candidate list a volume tool then scores, surfacing real user phrasing for content ideation, and powering a search-as-you-type UI where Google's own suggestion set is the point.
What people build on the Autocomplete API
- Long-tail keyword expansion via the alphabet-soup technique, feeding a separate volume tool
- Content-ideation lists built from real user phrasing rather than guesses
- Search-as-you-type UI for an internal search box or storefront
- Competitive content-gap research — what people ask around a competitor's brand terms
FAQ
Frequently asked questions
Does the Autocomplete API return search volume or CPC?
No. The response is the raw suggestion strings Google's autocomplete shows — no volume, CPC, or trend score attached to any of them.
What does it cost?
0.5 credits per query — half the rate of a full search — which works out to $0.15 per 1,000 queries.
How many suggestions come back per call?
Up to 10, set by limit (default 10). That matches how many Google's own autocomplete dropdown typically shows.
Can I get results for a specific language?
Yes — hl sets the language Google localizes suggestions to (default en).
What is the "alphabet soup" technique?
Looping a seed query through a-z (or two-letter prefixes for a common seed) as a suffix, one call per letter, then deduplicating the combined results into a much larger long-tail keyword list than one call returns.
Is this the same as Google Keyword Planner?
No. Keyword Planner (and similar tools) attach search-volume and competition data to keywords; this endpoint returns only the suggestion strings themselves. Use this to generate candidates, a volume tool to score them.
Is there a free tier?
Yes — the same 3,000 free credits every Searlo account starts with cover this endpoint too, at 0.5 credits per call.
Why does this cost less than a regular search?
Suggestions is 0.5 credits per query versus 1 credit for web, image, news, scholar and patents search — it's priced as the lightest-weight call on the API.