Google Search API
Ranked pages with titles, links and snippets.
What is this API?
Serqo’s Google Search API retrieves a Google web results page for a query and returns its results as structured JSON. Use the google engine with type: "web"; result positions and optional SERP features describe the page returned for that request.
Common use cases
- Track organic result positions across queries, countries and languages.
- Review snippets and SERP features such as related searches or answer boxes.
- Pass normalized result data to an internal reporting or research workflow.
Building autonomous AI agents or LLM tools? Read our guide on using SERP API with LangChain & AI Agents →
Request example
POST https://api.serqo.io/v1/search
| Field | Value |
|---|---|
| engine | google |
| type | web |
Request headers
Create an API key and store it on your server. Send it in the x-api-key header.
| Header | Value | Description |
|---|---|---|
| Content-Type | application/json | JSON request body. |
| x-api-key | YOUR_API_KEY | API key used to authenticate the request. |
export SERP_API_KEY='YOUR_API_KEY'
curl https://api.serqo.io/v1/search \
-H "x-api-key: $SERP_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"q": "open source observability tools",
"engine": "google",
"type": "web",
"gl": "us",
"hl": "en"
}' Code examples
Use the same API key with a standard HTTP client. These examples require Node.js 22 or later, or Python 3.
const response = await fetch('https://api.serqo.io/v1/search', {
method: 'POST',
headers: {
'x-api-key': process.env.SERP_API_KEY,
'Content-Type': 'application/json',
},
body: JSON.stringify({ q: 'open source observability tools', engine: 'google', type: 'web' }),
});
const result = await response.json();
if (!response.ok) {
console.error(response.status, result.requestId, result.error);
} else {
console.log(result.data.results);
} import json
import os
from urllib.error import HTTPError
from urllib.request import Request, urlopen
request = Request(
'https://api.serqo.io/v1/search',
data=json.dumps({'q': 'open source observability tools', 'engine': 'google', 'type': 'web'}).encode(),
headers={
'x-api-key': os.environ['SERP_API_KEY'],
'Content-Type': 'application/json',
},
)
try:
with urlopen(request) as response:
result = json.load(response)
print(result['data']['results'])
except HTTPError as error:
result = json.load(error)
print(error.code, result['requestId'], result['error']) Parameters
The number of results returned per request depends on the search engine and may vary. Use the pagination information in the response to continue.
| Field | Type | Required | Default | Description |
|---|---|---|---|---|
| q | string | Yes | — | Non-empty search query. |
| engine | string | No | Use google for this API. | |
| type | string | No | web | Use web for this API. |
| gl | string | No | — | Two-letter lowercase country code, for example us. |
| hl | string | No | — | Interface language code, for example en. |
| start | integer | No | — | Offset from 0 to 1000000. Use nextStart from the response. |
| safe | off | active | No | — | Explicit-content filter: off disables it, active enables it. |
| tbs | string | No | — | Google time filter, such as qdr:d, qdr:w or qdr:m. |
Only q is required by the API. Keep engine and type as shown for this search. Unknown parameters return 422 INVALID_PARAM.
Response
HTTP 200. A real API response, trimmed to the first three results for readability. Field availability and result groups vary by the returned page; values can be null.
{
"success": true,
"requestId": "req_3b994246-e7b2-4fbe-b932-0a49c9d5e3d6",
"data": {
"engine": "google",
"type": "web",
"query": {
"q": "open source observability tools",
"gl": "us",
"hl": "en"
},
"searchInformation": {
"totalResults": 123,
"timeTakenMs": 160,
"displayed": "About 123 results"
},
"results": [
{
"position": 1,
"type": "organic",
"title": "Open Source Data Observability Tools: When Free Isn't ...",
"url": "https://montecarlo.ai/blog-data-observability-open-source-tools",
"displayedUrl": "https://montecarlo.ai › blog-data-observability-open-sour...",
"domain": "montecarlo.ai",
"source": "montecarlo.ai",
"date": "Mar 9, 2026",
"snippet": "Mar 9, 2026 — With Monte Carlo , your data team can analyze data at-rest, deploy ML-powered anomaly detection to automatically detect incidents, integrate ... Read more",
"snippetHighlightedWords": [
"Monte Carlo"
],
"thumbnail": null,
"favicon": null,
"duration": null,
"price": null,
"rating": null,
"sitelinks": [
{
"title": "Read more",
"url": "https://www.google.com/goto?url=CAESzQEB6zswFeMmT2D1ovQBWKHRQvogdGw8E8byJiqcx12dE5C6-IAgkF4Byiebi6nfopFZhuJuADXNGSIDxaO5Y_vrnEONjvyxO98TnCdw-Jkpg9NLwQFglBO11Hob5i1hDpAyUpfqG-CYrNTSLl9_CK8juhEUY8eYmgJycAPg7WIWW6-vkUKDFpRg5C7wqqErASYox6Eq7ly4owx7n15uDHevwENJTanA2-22oxzUh1XH1DnuoKMkEIWbInfB31Qcb0Km2Qxj_caQGjPX2jDX"
}
],
"extensions": {
"googleRpos": "0"
}
},
{
"position": 2,
"type": "organic",
"title": "The 11 Best Observability Tools in 2026",
"url": "https://www.dash0.com/comparisons/best-observability-tools",
"displayedUrl": "https://www.dash0.com › Comparisons",
"domain": "dash0.com",
"source": "Dash0",
"date": "Jan 6, 2026",
"snippet": "Jan 6, 2026 — OpenObserve is an open-source observability platform designed for performance and efficiency. Written in Rust, it ships as a single binary ... Read more",
"snippetHighlightedWords": [
"OpenObserve"
],
"thumbnail": null,
"favicon": null,
"duration": null,
"price": null,
"rating": null,
"sitelinks": [
{
"title": "Read more",
"url": "https://www.google.com/goto?url=CAES4gEB6zswFUKVgd3cbv5XUQzW5x89ABN30qEesbRfxFIGdxKP_rlpbeK57JpD4Npi2TzhLx7ayJs8S5Xwrm3whmAsfTK8tG-8972gGVV68a-p6lG26kClQG4IcMBmHx9wOz4XLYlmb7Z6wGCB6VJi0XBNVC857SpxyN_JWVx7Ht6PE-pxH6HPGS9XP-2XRdgjo5Mc9BYfKA7ZJtqnlgI4vmRg7HM0R4i2cPzjZFvvDlQXbp39gf4SMoEuPF3IGHfSo93QsxjnQuj5aBP79wEb_OtRAjvQ1eJ8QYgDuaj_E9JsEGa8"
}
],
"extensions": {
"googleRpos": "1"
}
},
{
"position": 3,
"type": "organic",
"title": "7 Best AI Observability Tools for Engineering Teams",
"url": "https://blog.n8n.io/ai-observability-tools/",
"displayedUrl": "https://blog.n8n.io › ai-observability-tools",
"domain": "blog.n8n.io",
"source": "n8n Blog",
"date": "Aug 6, 2026",
"snippet": "Aug 6, 2026 — Arize Phoenix is an open-source observability platform focused on debugging and improving AI applications. SigNoz and LangFuse are popular open ...",
"snippetHighlightedWords": [
"Arize Phoenix is an open-source observability platform"
],
"thumbnail": null,
"favicon": null,
"duration": null,
"price": null,
"rating": null,
"sitelinks": [],
"extensions": {
"googleRpos": "6"
}
}
],
"features": {
"relatedSearches": [
{
"query": "Open source observability tools free",
"url": "https://www.google.com/search?safe=off&sca_esv=64345335af5ad8d5&hl=en&gl=us&sxsrf=APpeQnvM7CVcRGCpsL6a9rPwZHtxd5bapg:1792732918403&q=Open+source+observability+tools+free&sa=X&ved=2ahUKEwiH3pzGss-XAxWekokEHeK8LfQQ1QJ6BAg3EAE"
},
{
"query": "Open source observability tools github",
"url": "https://www.google.com/search?safe=off&sca_esv=64345335af5ad8d5&hl=en&gl=us&sxsrf=APpeQnvM7CVcRGCpsL6a9rPwZHtxd5bapg:1792732918403&q=Open+source+observability+tools+github&sa=X&ved=2ahUKEwiH3pzGss-XAxWekokEHeK8LfQQ1QJ6BAg9EAE"
},
{
"query": "Best open source observability tools",
"url": "https://www.google.com/search?safe=off&sca_esv=64345335af5ad8d5&hl=en&gl=us&sxsrf=APpeQnvM7CVcRGCpsL6a9rPwZHtxd5bapg:1792732918403&q=Best+open+source+observability+tools&sa=X&ved=2ahUKEwiH3pzGss-XAxWekokEHeK8LfQQ1QJ6BAg8EAE"
},
{
"query": "Observability tools list",
"url": "https://www.google.com/search?safe=off&sca_esv=64345335af5ad8d5&hl=en&gl=us&sxsrf=APpeQnvM7CVcRGCpsL6a9rPwZHtxd5bapg:1792732918403&q=Observability+tools+list&sa=X&ved=2ahUKEwiH3pzGss-XAxWekokEHeK8LfQQ1QJ6BAg7EAE"
},
{
"query": "SigNoz",
"url": "https://www.google.com/search?safe=off&sca_esv=64345335af5ad8d5&hl=en&gl=us&sxsrf=APpeQnvM7CVcRGCpsL6a9rPwZHtxd5bapg:1792732918403&q=SigNoz&sa=X&ved=2ahUKEwiH3pzGss-XAxWekokEHeK8LfQQ1QJ6BAg6EAE"
},
{
"query": "Top 10 observability tools",
"url": "https://www.google.com/search?safe=off&sca_esv=64345335af5ad8d5&hl=en&gl=us&sxsrf=APpeQnvM7CVcRGCpsL6a9rPwZHtxd5bapg:1792732918403&q=Top+10+observability+tools&sa=X&ved=2ahUKEwiH3pzGss-XAxWekokEHeK8LfQQ1QJ6BAg5EAE"
},
{
"query": "OpenObserve",
"url": "https://www.google.com/search?safe=off&sca_esv=64345335af5ad8d5&hl=en&gl=us&sxsrf=APpeQnvM7CVcRGCpsL6a9rPwZHtxd5bapg:1792732918403&q=OpenObserve&sa=X&ved=2ahUKEwiH3pzGss-XAxWekokEHeK8LfQQ1QJ6BAg4EAE"
},
{
"query": "Elastic Observability",
"url": "https://www.google.com/search?safe=off&sca_esv=64345335af5ad8d5&hl=en&gl=us&sxsrf=APpeQnvM7CVcRGCpsL6a9rPwZHtxd5bapg:1792732918403&q=Elastic+Observability&sa=X&ved=2ahUKEwiH3pzGss-XAxWekokEHeK8LfQQ1QJ6BAg2EAE"
},
{
"query": "2",
"url": "https://www.google.com/search?q=open+source+observability+tools&safe=off&sca_esv=64345335af5ad8d5&hl=en&gl=us&sxsrf=APpeQnvM7CVcRGCpsL6a9rPwZHtxd5bapg:1792732918403&ei=9u7aaoeLGJ6lptQP4vm2oQ8&start=10&sa=N&sstk=AS6-VmIfDu1Dr98WUp_qT0VP_EC7x8XKg8A8SOXvYt7fsobqGQn_eqhHNWV8YgjrilITKrMvvSZC8nYZPNAYZ9FGyee2UkKPzSPImw&ved=2ahUKEwiH3pzGss-XAxWekokEHeK8LfQQ8tMDegQIMRAE"
},
{
"query": "3",
"url": "https://www.google.com/search?q=open+source+observability+tools&safe=off&sca_esv=64345335af5ad8d5&hl=en&gl=us&sxsrf=APpeQnvM7CVcRGCpsL6a9rPwZHtxd5bapg:1792732918403&ei=9u7aaoeLGJ6lptQP4vm2oQ8&start=20&sa=N&sstk=AS6-VmIfDu1Dr98WUp_qT0VP_EC7x8XKg8A8SOXvYt7fsobqGQn_eqhHNWV8YgjrilITKrMvvSZC8nYZPNAYZ9FGyee2UkKPzSPImw&ved=2ahUKEwiH3pzGss-XAxWekokEHeK8LfQQ8tMDegQIMRAG"
},
{
"query": "4",
"url": "https://www.google.com/search?q=open+source+observability+tools&safe=off&sca_esv=64345335af5ad8d5&hl=en&gl=us&sxsrf=APpeQnvM7CVcRGCpsL6a9rPwZHtxd5bapg:1792732918403&ei=9u7aaoeLGJ6lptQP4vm2oQ8&start=30&sa=N&sstk=AS6-VmIfDu1Dr98WUp_qT0VP_EC7x8XKg8A8SOXvYt7fsobqGQn_eqhHNWV8YgjrilITKrMvvSZC8nYZPNAYZ9FGyee2UkKPzSPImw&ved=2ahUKEwiH3pzGss-XAxWekokEHeK8LfQQ8tMDegQIMRAI"
},
{
"query": "5",
"url": "https://www.google.com/search?q=open+source+observability+tools&safe=off&sca_esv=64345335af5ad8d5&hl=en&gl=us&sxsrf=APpeQnvM7CVcRGCpsL6a9rPwZHtxd5bapg:1792732918403&ei=9u7aaoeLGJ6lptQP4vm2oQ8&start=40&sa=N&sstk=AS6-VmIfDu1Dr98WUp_qT0VP_EC7x8XKg8A8SOXvYt7fsobqGQn_eqhHNWV8YgjrilITKrMvvSZC8nYZPNAYZ9FGyee2UkKPzSPImw&ved=2ahUKEwiH3pzGss-XAxWekokEHeK8LfQQ8tMDegQIMRAK"
},
{
"query": "6",
"url": "https://www.google.com/search?q=open+source+observability+tools&safe=off&sca_esv=64345335af5ad8d5&hl=en&gl=us&sxsrf=APpeQnvM7CVcRGCpsL6a9rPwZHtxd5bapg:1792732918403&ei=9u7aaoeLGJ6lptQP4vm2oQ8&start=50&sa=N&sstk=AS6-VmIfDu1Dr98WUp_qT0VP_EC7x8XKg8A8SOXvYt7fsobqGQn_eqhHNWV8YgjrilITKrMvvSZC8nYZPNAYZ9FGyee2UkKPzSPImw&ved=2ahUKEwiH3pzGss-XAxWekokEHeK8LfQQ8tMDegQIMRAM"
},
{
"query": "7",
"url": "https://www.google.com/search?q=open+source+observability+tools&safe=off&sca_esv=64345335af5ad8d5&hl=en&gl=us&sxsrf=APpeQnvM7CVcRGCpsL6a9rPwZHtxd5bapg:1792732918403&ei=9u7aaoeLGJ6lptQP4vm2oQ8&start=60&sa=N&sstk=AS6-VmIfDu1Dr98WUp_qT0VP_EC7x8XKg8A8SOXvYt7fsobqGQn_eqhHNWV8YgjrilITKrMvvSZC8nYZPNAYZ9FGyee2UkKPzSPImw&ved=2ahUKEwiH3pzGss-XAxWekokEHeK8LfQQ8tMDegQIMRAO"
},
{
"query": "8",
"url": "https://www.google.com/search?q=open+source+observability+tools&safe=off&sca_esv=64345335af5ad8d5&hl=en&gl=us&sxsrf=APpeQnvM7CVcRGCpsL6a9rPwZHtxd5bapg:1792732918403&ei=9u7aaoeLGJ6lptQP4vm2oQ8&start=70&sa=N&sstk=AS6-VmIfDu1Dr98WUp_qT0VP_EC7x8XKg8A8SOXvYt7fsobqGQn_eqhHNWV8YgjrilITKrMvvSZC8nYZPNAYZ9FGyee2UkKPzSPImw&ved=2ahUKEwiH3pzGss-XAxWekokEHeK8LfQQ8tMDegQIMRAQ"
},
{
"query": "9",
"url": "https://www.google.com/search?q=open+source+observability+tools&safe=off&sca_esv=64345335af5ad8d5&hl=en&gl=us&sxsrf=APpeQnvM7CVcRGCpsL6a9rPwZHtxd5bapg:1792732918403&ei=9u7aaoeLGJ6lptQP4vm2oQ8&start=80&sa=N&sstk=AS6-VmIfDu1Dr98WUp_qT0VP_EC7x8XKg8A8SOXvYt7fsobqGQn_eqhHNWV8YgjrilITKrMvvSZC8nYZPNAYZ9FGyee2UkKPzSPImw&ved=2ahUKEwiH3pzGss-XAxWekokEHeK8LfQQ8tMDegQIMRAS"
}
],
"aiOverview": {
"available": true,
"text": "Open source observability tools are organized around the three core pillars of telemetry: metrics , logs , and traces . Medium · Dotan Horovits (@horovits) +1 Frameworks & Instrumentation OpenTelemetry : The vendor-neutral CNCF framework for collecting metrics, logs, and traces using a unified set of APIs, SDKs, and the OpenTelemetry Collector. OpenTelemetry +1 Metrics Tools Prometheus : A CNCF-graduated time-series database and alerting system built for pull-based metric scraping. Chronosphere +1 VictoriaMetrics : A high-performance, cost-effective, Prometheus-compatible time-series database. Netdata Thanos / Mimir : Scalable, long-term metric storage layers designed to extend Prometheus clusters",
"sources": []
}
},
"pagination": {
"page": 1,
"start": 0,
"hasNext": true,
"nextStart": 10
},
"meta": {
"elapsedMs": 6316
}
},
"meta": {
"cache": "miss"
}
}
Response fields
| Field | Type | Description |
|---|---|---|
| data.query | object | Search query passed to the engine. |
| data.searchInformation | object | Reported result count and search timing, when provided. |
| success | boolean | Whether the request succeeded. True in this example. |
| requestId | string | Request identifier to include when troubleshooting. |
| data.type | string | Search type used for this response. |
| data.engine | string | Engine that returned the results. |
| data.results | array | Main result list. Check feature groups as well when present. |
| data.features | object | Additional result groups; groups not returned by the engine may be absent. |
| data.pagination | object | Current offset and next-page availability. |
| data.meta.elapsedMs | number | Search duration in milliseconds, not a performance guarantee. |
| meta.cache | string | null | hit, miss, or null if cache status is unknown. |
Result fields
Read entries from data.results. Additional groups, when present, are in data.features.
| Field | Type | Description |
|---|---|---|
| data.results[].domain | string | null | Destination domain, when available. |
| data.results[].favicon | string | null | Site icon when provided. |
| data.results[].sitelinks | array | Additional links nested under the result. |
| data.results[].sitelinks[].title | string | Sitelink label. |
| data.results[].sitelinks[].url | string | null | Sitelink destination. |
| data.results[].extensions | object | Other result-specific properties. |
| data.results[].snippetHighlightedWords | string[], optional | Highlighted terms in the snippet, when provided. |
| data.results[].position | integer | Rank on this results page, starting at 1. |
| data.results[].type | string | Result category: organic, news, image, video, shopping, local, forum, book, patent or ad. This can differ from the request type. |
| data.results[].title | string | null | Title shown in the result. |
| data.results[].url | string | null | Destination URL when resolved; it may be null. |
| data.results[].displayedUrl | string | null | URL or domain displayed by the engine; not always a clickable link. |
| data.results[].source | string | null | Source name or displayed domain. |
| data.results[].date | string | null | Date text when extracted; not necessarily an ISO timestamp. |
| data.results[].snippet | string | null | Result summary. |
| data.results[].thumbnail | string | null | Thumbnail URL or inline data URL, when provided. |
| data.results[].duration | string | null | Video duration text when provided. |
| data.results[].price | string | null | Price text when provided. |
| data.results[].rating | number | null | Rating when provided. |
Additional Google web results
Google may also return an AI Overview, related questions, related searches and knowledge panels. These groups are optional and are not included in the abbreviated example above.
| Field | Type | Description |
|---|---|---|
| data.features.peopleAlsoAsk | array | Follow-up questions, answers and source links when provided. |
| data.features.peopleAlsoAsk[].question | string | Suggested question. |
| data.features.peopleAlsoAsk[].answer | string | null | Answer excerpt. |
| data.features.peopleAlsoAsk[].sources | array | Source entries with title and url. |
| data.features.relatedSearches | array | Related search queries. |
| data.features.relatedSearches[].query | string | Suggested query. |
| data.features.relatedSearches[].url | string | null | Link to the related search when provided. |
| data.features.knowledgePanel | object, optional | Knowledge panel with title, subtitle, description, attributes, sources and thumbnail. |
| data.features.answerBox | object, optional | Answer box with title, snippet, url, source and date. |
| data.features.aiOverview | object, optional | Google AI Overview with extracted text and available source links, when present. |
Pagination
- Check
data.pagination.hasNext; stop when it is false. - Use
data.pagination.nextStartasstart. Keep the other search parameters unchanged. - Send the next request. Each page is a separate search request.
| Field | Type | Description |
|---|---|---|
| data.pagination.nextUrl | string | null | Search-engine URL for the next page, when available; use nextStart with this API. |
| data.pagination.requestedPage | integer | Page requested; may differ from the returned page if the engine does not provide it. |
| data.pagination.page | integer | Current page number. |
| data.pagination.start | integer | Current result offset. |
| data.pagination.hasNext | boolean | Whether another page is available. |
| data.pagination.nextStart | integer | null | Offset for the next request, or null when unavailable. |
Errors
{
"success": false,
"requestId": "req_example",
"error": {
"code": "INVALID_PARAM",
"message": "Unknown parameter \"page\"",
"retryable": false
}
} | Field | Type | Description |
|---|---|---|
| error.code | string | Machine-readable error code. |
| error.message | string | Reason the request failed. |
| error.retryable | boolean | Whether retrying is recommended. |
| error.retryAfterMs | number, optional | Minimum wait before retrying, in milliseconds. |
| HTTP status | Meaning | What to do |
|---|---|---|
| 400 | Invalid request or unsupported search type. | Check the query and engine/type combination. |
| 401 | Missing, invalid or revoked API key. | Check the authentication header and API key. |
| 402 | Insufficient credits. | Check your credit balance. |
| 403 | Permission or plan restriction. | Check account access and plan limits. |
| 422 | Invalid parameter. | Correct the parameter identified by error.message. |
| 429 | Request rate or queue limit. | Follow the retry rules below. |
| 5xx | Server or search service failure. | Keep the requestId; retry only if allowed. |
Retry rules
- Retry only when
error.retryableistrue. - If
error.retryAfterMsis present, wait at least that many milliseconds.
Pricing and limits
- Successful searches use credits at the applicable rate.
- Failed searches and timeouts are not charged.
- Cache hits are free at the default rate. Before searching, your available balance must cover the full request cost.
Сравнить цены Google Custom Search API с затратами на SERP API →
See plan pricing and rate-limit and credit details.
Frequently asked questions
How do I request another page of Google results?
Check data.pagination.hasNext, then send data.pagination.nextStart as start with the same query and market settings. Each page is a separate search request.
How are Google Search API requests charged?
Successful searches use the applicable credit rate for the engine, type and country. Failed searches and timeouts release the reservation; cache hits are free at the default rate. See Pricing and limits below for the current rules.