Search results, as JSON.
The SERP API runs a search on Google, Bing or DuckDuckGo from the country you choose and returns organic results, ads, featured snippets and related searches as structured JSON.
What it does
Google, Bing and DuckDuckGo results as structured JSON.
Three engines, one shape
Set
enginetogoogle,bingorduckduckgo. Every answer has the same fields; sections an engine does not have are empty.Every part of the page
Organic results with sitelinks, ads by placement, the featured snippet, the answer box, People also ask and related searches.
Country, language and device
Search from an exit in the
countryyou pick, in the interfacelanguageyou want, as desktop or mobile.Honest parse status
parse_statussays whether everything was read (complete), some of it (partial, with warnings) or nothing (failed).Consent pages and captchas handled
Consent screens, captchas and JavaScript checks are detected and retried on another tier or IP, like any other block.
Sync, async or webhook
Wait for the answer, or send
async: truewith awebhook_urland get a signedjob.finishedevent.
Example
One HTTP call
POST an engine and a query. Positions continue across pages, so page 2 with 10 results per page starts at 11.
Official SDKs are coming soon. Until then, any HTTP client works.
SERP API docscurl https://api.siftzo.com/v1/serp \
-H "X-API-Key: $SIFTZO_API_KEY" \
-H "Content-Type: application/json" \
-d '{"engine": "google", "q": "web scraping api", "country": "us", "language": "en"}'const res = await fetch("https://api.siftzo.com/v1/serp", {
method: "POST",
headers: {
"X-API-Key": process.env.SIFTZO_API_KEY,
"Content-Type": "application/json",
},
body: JSON.stringify({ engine: "google", q: "web scraping api", country: "us", language: "en" }),
signal: AbortSignal.timeout(90_000),
});
const serp = await res.json();
if (!res.ok) throw new Error(serp.error.code);
for (const r of serp.results.organic) {
console.log(r.position, r.title, r.url);
}import os
import requests
res = requests.post(
"https://api.siftzo.com/v1/serp",
headers={"X-API-Key": os.environ["SIFTZO_API_KEY"]},
json={"engine": "google", "q": "web scraping api", "country": "us", "language": "en"},
timeout=90,
)
serp = res.json()
res.raise_for_status()
for r in serp["results"]["organic"]:
print(r["position"], r["title"], r["url"])Reference
Key parameters
enginegoogle | bing | duckduckgorequired- The search engine.
qstringrequired- The search, 1–512 characters.
countryISO 3166-1 alpha-2- Exit country, also passed to the engine as its country.
languagestring- Interface language, for example en or pt-br.
devicedesktop | mobiledefaultdesktop- Mobile searches as Chrome on Android when fetched over HTTP.
pageintegerdefault1- 1–10. DuckDuckGo has page 1 only.
asyncbooleandefaultfalse- Return a job ID at once. Add
webhook_urlfor a signed callback. include_htmlbooleandefaultfalse- Add the raw HTML of the results page.
Every parameter and response field is in the SERP API docs.
Credits
What it costs
| Request | Credits |
|---|---|
| SERP request (any engine, any tier) | 10 |
| Failed request (blocked, timeout, proxy error) | 0 |
A search costs the same whichever engine answers and whichever tier fetched it.
See plansQuestions
Does a search cost more when a browser is needed?
No. A SERP request has one price for every engine and tier.
What if the page cannot be parsed?
You get parse_status: "failed" with warnings that say why. A page that was fetched without a block is billed even if it cannot be parsed; a blocked search costs nothing.
Can I get more than the first page?
Yes, page goes up to 10 on Google and Bing. DuckDuckGo needs a token from the previous page, so only page 1 is available there.
Can I search as a mobile user?
device: "mobile" sends the search as Chrome on Android when the page is fetched over HTTP. When a browser is needed, the search runs as desktop.
More from Siftzo
- Scraper APIAny public page as HTML or Markdown, with proxies, rendering and retries handled.
- Browser APIA fresh cloud Chrome for your Playwright or Puppeteer code, over CDP.
- Crawl APIFollow the links of a site up to 10,000 pages, always within robots.txt.
- Data APIProducts, articles and page metadata as records with fixed fields.
- AI extractionDescribe the data with a JSON Schema or a prompt and get JSON that matches.
- Proxy accessSiftzo's residential network in your own client, billed per MB.
Try the SERP API for free
Create an account, copy your key and send a request. No card needed for the Free plan.