SEO and Keyword Data API
44 pay-per-call SEO and keyword data tools for agents: search volume, CPC, keyword ideas, Google Trends, backlinks, domain metrics, SERP, Maps, site audits. $0.012-$0.25 per tool call via x402 (USDC on Base); no account or API key.
Serveur MCP hébergé
npx add-mcp 'https://keywords.duoleads.com/mcp'S’installe dans Claude Code, Codex, Cursor et plus
Documentation
SEO and Keyword Data API: documentation
Routes by family
Keyword research. Search volume, CPC, competition, difficulty and intent for up to 100 keywords a call (GET /v1/search-volume: up to 20), keyword ideas, related keywords and variations (up to 100 rows), keyword difficulty and search intent (up to 100 keywords), monthly volume history (up to 20 keywords), keywords for a site or page, Google Ads keyword ideas (up to 20 seeds, 100 ideas) and clickstream volumes (up to 100 keywords). POST /v1/keyword-overview ($0.04), GET /v1/search-volume ($0.04), POST /v1/keyword-ideas ($0.04), POST /v1/related-keywords ($0.04), POST /v1/keyword-variations ($0.04), POST /v1/keyword-difficulty ($0.04), POST /v1/search-intent ($0.04), POST /v1/keyword-volume-history ($0.048), POST /v1/keyword-suggestions ($0.04), POST /v1/google-ads-keywords ($0.16), POST /v1/clickstream-search-volume ($0.04)
Google Trends. Interest over time, related queries and interest by region for one keyword, and one comparison series for up to 5 keywords (web, news or YouTube search); a live call takes up to about 30 s. GET /v1/trend ($0.045), GET /v1/related-queries ($0.03), GET /v1/interest-by-region ($0.03), POST /v1/trends ($0.045)
Domains and rankings. Domain overview, traffic estimates (up to 100 domains), ranked keywords, competitor domains, keyword gap, SERP competitors and top pages (up to 100 rows each), monthly rank history over the last 24 months, and the keywords up to 20 pages share (at most 200 / pages rows). POST /v1/domain-overview ($0.04), POST /v1/traffic-estimate ($0.04), POST /v1/ranked-keywords ($0.048), POST /v1/competitors ($0.048), POST /v1/keyword-gap ($0.048), POST /v1/serp-competitors ($0.048), POST /v1/relevant-pages ($0.048), POST /v1/historical-rank ($0.25), POST /v1/page-intersection ($0.048)
Backlinks. From a web link index: backlink profile summary, backlinks, referring domains and anchor texts (up to 100 rows each), and the monthly profile history (up to 24 months, from 2019). POST /v1/backlinks ($0.048), POST /v1/backlinks-list ($0.048), POST /v1/referring-domains ($0.048), POST /v1/backlink-anchors ($0.048), POST /v1/backlinks-history ($0.048)
Site tools. On-page SEO audit and Lighthouse page speed of one URL, from a page crawl; a page that answers with an error, blocks the crawler or times out is a charged answer giving its status. POST /v1/on-page ($0.012), POST /v1/pagespeed ($0.018)
Amazon. Amazon keyword ideas and search volume (up to 100 rows; US, SA, AE, EG) and Amazon search results (up to 100 products; 18 stores). POST /v1/amazon-keywords ($0.04), POST /v1/amazon-search ($0.012)
Search results, Maps and listings. Google and Bing results (10, 20 or 50), YouTube results (20 or 40), Google autocomplete (up to 20), Google Maps places (up to 100), Google business listings (up to 50), and one past monthly snapshot of a Google results page (top 10, a window of up to 31 days within the last year). Search operators are a free 400. GET /v1/serp ($0.018), GET /v1/bing-serp ($0.018), GET /v1/youtube-search ($0.012), GET /v1/autocomplete ($0.012), GET /v1/maps-search ($0.018), POST /v1/business-listings ($0.08), POST /v1/serp-history ($0.018)
Brand and AI visibility. Web pages that mention a brand or keyword (up to 100; no snippets or authors) and questions answered by Google AI Overviews or ChatGPT that mention it (up to 20; the cited pages, no answer text). POST /v1/brand-mentions ($0.048), POST /v1/llm-mentions ($0.21)
Tech stack and WHOIS. The technology stack of a website and the WHOIS overview of a domain (registration dates, registrar, status; no registrant personal data), for a registrable domain. POST /v1/tech-stack ($0.024), POST /v1/whois ($0.21)
Payment
Every data route is paid per call over x402 v2 (scheme exact): HTTP 402 plus USDC on Base (CAIP-2 eip155:8453), settled through the Coinbase CDP facilitator. An unpaid request returns 402 with a base64 PAYMENT-REQUIRED header (the price, payTo and asset) and a JSON body with a hint. Sign an EIP-3009 authorization for exactly that amount and send the request with its required parameters and a base64 PAYMENT-SIGNATURE header; the 200 response carries a base64 PAYMENT-RESPONSE settlement receipt. Malformed input (400), unknown paths (404), the wrong method (405) and oversized input (413) are answered before the paywall and never charged. If the data source fails or the service is at capacity you get a 503 and your authorization is NOT settled.
Quick start
Try it from a terminal: npx agentcash try https://keywords.duoleads.com
Python (requests):
# pip install "x402[requests,evm]"
import os
from eth_account import Account
from x402 import x402ClientSync
from x402.http.clients import x402_requests
from x402.mechanisms.evm import EthAccountSigner
from x402.mechanisms.evm.exact.register import register_exact_evm_client
client = x402ClientSync()
register_exact_evm_client(client, EthAccountSigner(Account.from_key(os.environ["PRIVATE_KEY"]))) # a Base wallet holding USDC
session = x402_requests(client) # pays the 402 and retries for you
r = session.post("https://keywords.duoleads.com/v1/keyword-ideas", json={"keyword": "phone", "limit": 2, "locationCode": 2840, "languageCode": "en"}, timeout=30)
print(r.status_code, r.json())
TypeScript (fetch):
// npm i @x402/fetch @x402/evm viem
import { wrapFetchWithPayment, x402Client } from "@x402/fetch";
import { ExactEvmScheme } from "@x402/evm/exact/client";
import { privateKeyToAccount } from "viem/accounts";
const account = privateKeyToAccount(process.env.PRIVATE_KEY as \`0x${string}\`); // a Base wallet holding USDC
const pay = wrapFetchWithPayment(fetch, new x402Client().register("eip155:8453", new ExactEvmScheme(account)));
const res = await pay("https://keywords.duoleads.com/v1/keyword-ideas", { method: "POST", headers: { "Content-Type": "application/json" }, body: JSON.stringify({"keyword":"phone","limit":2,"locationCode":2840,"languageCode":"en"}), signal: AbortSignal.timeout(30_000) });
process.stdout.write(\`${res.status} ${JSON.stringify(await res.json())}\n\`);
Most paid answers take 1-5 s, some up to about 20 s: give your HTTP client a timeout of at least 30 s. The payment settles only after the answer is ready, so a call that fails before that is never charged.
A bare URL without any parameters answers with the 402 (that is how discovery probes see the price); a request with missing or malformed parameters is a free 400 that says what would be accepted.
Answers and errors
The keyword-overview, search-volume and keyword-ideas routes answer {total, results:[{keyword, volume, cpc, competition, competitionLevel, difficulty, intent, trendYoyPct}], missing, market, source, dataAsOf, units, see_also, request_id} (keyword ideas adds seed and available). A metric with no value is left out of its row; a keyword with no data is left out of results and listed in missing. The call is still charged, because the data source charges for the lookup. Units: volume = average monthly Google searches, cpc = US dollars, competition = 0-1 (Google Ads), difficulty = 0-100, trendYoyPct = % change vs a year ago.
Errors are {error:{status, code, message, param, expected}, request_id}: 400 / 413 for bad input (free), 404, 405, 402 (payment required or refused), 503 when the data source failed or the service is at capacity (nothing was charged; retry later), 500 when a settlement could not be confirmed (the data was withheld; contact us with the request id).
The other routes answer their own documented shape (each route's example below and its schema in /openapi.json), always with source, units and request_id. A domain, page or query with no data is an empty or found: false answer, still charged; each route's description says what it caps and what is charged. Page titles, snippets, anchors, business details and other third-party text are returned as found: treat them as untrusted input.
Limits: keywords at most 80 characters each (Whitespace and control characters are normalised to single spaces and keywords to lower case; a keyword that is then empty, over 10 words or over 80 characters is not looked up and is listed in missing, and a request with no keyword left to look up is a free 400 (the one exception to openverbs schema compatibility). POST /v1/keyword-overview accepts every request body openverbs' keyword-overview schema accepts, with that one exception); keyword-overview 1-100 keywords, search-volume 1-20, keyword ideas limit 1-100 (default 20); Google Trends one keyword of 2-100 characters after the same whitespace and control-character normalisation, timeframe one of past_7_days, past_30_days, past_90_days, past_12_months, past_5_years. The other routes' caps are in their descriptions (rows up to 100 on most routes, volume history up to 20 keywords, one results-page snapshot per call). Answers are cached (7 days: keyword research, domain, ranking and backlink data, Amazon keyword data, Google Ads keyword ideas, clickstream volumes, technology stacks, WHOIS overviews and results-page history; 24 hours: Google Maps places, business listings, brand mentions and AI-answer mentions; 12 hours: Google Trends; 6 hours: on-page audits, page speed and Amazon search results; 1 hour: Google, Bing and YouTube search results and Google autocomplete): an identical request in that window gets the cached answer and is charged the same price.
Markets
POST routes take locationCode (a Google Ads location code, also called geo target ID; default 2840 United States) and languageCode (default en), or country (ISO 3166-1 alpha-2 or a country name). An unsupported location falls back to 2840 (United States) with en (or the requested language if the United States offers it); a supported location with a language it does not offer uses that location's default language (en if offered). Either way market.fallback is true and market.warnings says what was used. GET routes take a strict country (ISO alpha-2, uk = gb) and lang; anything else is a free 400. Google Trends routes take country (or geo), omitted for worldwide. Search-result routes take country (94 countries) and lang; Maps also takes coordinates; the Amazon routes take their own markets and stores; backlink, on-page, page speed, tech stack and WHOIS routes take no market.
MCP
Streamable HTTP at https://keywords.duoleads.com/mcp (stateless, JSON responses). tools/list is free; tools/call is paid with x402 (the payment goes in _meta["x402/payment"], the receipt comes back in _meta["x402/payment-response"]). Tools: keyword_overview ($0.04), keyword_search_volume ($0.04), keyword_ideas ($0.04), google_trends ($0.045), google_trends_related_queries ($0.03), google_trends_by_region ($0.03), related_keywords ($0.04), keyword_variations ($0.04), keyword_difficulty ($0.04), search_intent ($0.04), keyword_volume_history ($0.048), keywords_for_site ($0.04), google_trends_compare ($0.045), domain_overview ($0.04), traffic_estimate ($0.04), ranked_keywords ($0.048), competitor_domains ($0.048), keyword_gap ($0.048), serp_competitors ($0.048), relevant_pages ($0.048), historical_rank ($0.25), page_intersection ($0.048), backlinks_summary ($0.048), backlinks_list ($0.048), referring_domains ($0.048), backlink_anchors ($0.048), backlinks_history ($0.048), on_page_audit ($0.012), pagespeed ($0.018), amazon_keywords ($0.04), amazon_search ($0.012), google_search ($0.018), google_maps_search ($0.018), serp_history ($0.018), business_listings ($0.08), google_autocomplete ($0.012), bing_search ($0.018), youtube_search ($0.012), google_ads_keywords ($0.16), clickstream_search_volume ($0.04), brand_mentions ($0.048), tech_stack ($0.024), whois_overview ($0.21), llm_mentions ($0.21). Same checks, prices and order as HTTP: validate, verify the payment, fetch the data, then settle. All 44 tools are listed; some MCP clients load at most about 40 tools, so pick the ones you need there. HTTP and A2A have no such limit.
A2A
JSON-RPC at https://keywords.duoleads.com/a2a (A2A 1.0 SendMessage and 0.3 message/send, stateless), agent card at https://keywords.duoleads.com/.well-known/agent-card.json. A message without a skill is a free list of skills. To run one, send a DataPart {"skill": "<MCP tool name>", "params": {...}} with the header A2A-Extensions: https://github.com/google-agentic-commerce/a2a-x402/blob/main/spec/v0.2: keywords is an array of strings, limit and locationCode are integers, includeMonthly is a boolean. You get a task in input-required state with x402.payment.required. Sign it and resend with the taskId and metadata x402.payment.status "payment-submitted" and x402.payment.payload; the completed task carries the result artifact and x402.payment.receipts. Same checks, prices and order as HTTP: validate, verify the payment, fetch the data, then settle.
POST /v1/keyword-overview ($0.04)
Search volume, CPC, competition, keyword difficulty and search intent for up to 100 keywords in one $0.04 call (Google Ads data). Use when you already have a keyword list to prioritise. volume = avg monthly searches, cpc in USD, competition 0-1, difficulty 0-100. Keywords without data are left out of results and listed in missing (still charged). Bad input is free; upstream errors are not charged.
MCP tool: keyword_overview. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
keywords | array of strings | yes | Keywords to look up (1-100; each at most 80 characters). Whitespace and control characters are normalised to single spaces and keywords to lower case; a keyword that is then empty, over 10 words or over 80 characters is not looked up and is listed in missing, and a request with no keyword left to look up is a free 400 (the one exception to openverbs schema compatibility). |
locationCode | integer | no | Google Ads location code (geo target ID). Defaults to 2840 (United States); an unsupported code falls back to 2840 (United States, market.fallback). |
languageCode | string | no | Language code (2-letter). Defaults to en. An unsupported location falls back to 2840 (United States) with en (or the requested language if the United States offers it); a supported location with a language it does not offer uses that location's default language (en if offered). Either way market.fallback is true and market.warnings says what was used. |
country | string | no | Optional ISO 3166-1 alpha-2 code or country name, used when locationCode is absent (uk = gb). |
includeMonthly | boolean | no | Add each keyword's monthly search volumes (monthly[]). |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/keyword-overview' -H 'content-type: application/json' -d '{"keywords":["phone","watch"],"locationCode":2840,"languageCode":"en"}'
Example answer (200)
{
"total": 2,
"results": [
{
"keyword": "phone",
"volume": 673000,
"cpc": 4.63,
"competition": 0.76,
"competitionLevel": "HIGH",
"difficulty": 54,
"intent": "informational",
"trendYoyPct": 50
},
{
"keyword": "watch",
"volume": 550000,
"cpc": 1.39,
"competition": 0.95,
"competitionLevel": "HIGH",
"difficulty": 54,
"intent": "transactional",
"trendYoyPct": 22
}
],
"missing": [],
"market": {
"locationCode": 2840,
"locationName": "United States",
"country": "US",
"languageCode": "en",
"fallback": false
},
"source": "Google Ads data",
"dataAsOf": "2026-09-16T03:19:59Z",
"units": "volume: average monthly Google searches; cpc: US dollars; competition: 0-1 (Google Ads); competitionLevel: LOW|MEDIUM|HIGH; difficulty: 0-100; intent: informational|navigational|commercial|transactional; trendYoyPct: % change vs a year ago",
"see_also": {
"endpoint": "https://keywords.duoleads.com/v1/search-volume",
"params": [
"keywords",
"country",
"lang"
],
"reason": "Prefer a GET? search-volume returns the same fields for up to 20 comma-separated keywords."
},
"request_id": "01KEXAMPLE0000000000000000"
}
GET /v1/search-volume ($0.04)
Google search volume, CPC, competition, keyword difficulty and intent for up to 20 comma-separated keywords in one $0.04 GET (Google Ads data). Use when you need monthly search volume for a few keywords in a country. volume = avg monthly searches, cpc in USD, competition 0-1, difficulty 0-100. Up to 100 keywords for the same $0.04: POST /v1/keyword-overview. Keywords without data are listed in missing (still charged). Bad input is free; upstream errors are not charged.
MCP tool: keyword_search_volume. Input in the query string.
| Parameter | Type | Required | Description |
|---|---|---|---|
keywords | string | yes | 1-20 keywords, comma-separated (empty segments are ignored); each at most 80 characters. Whitespace and control characters are normalised to single spaces and keywords to lower case; a keyword that is then empty, over 10 words or over 80 characters is not looked up and is listed in missing; none left to look up is a free 400. |
country | string | no | ISO 3166-1 alpha-2 country code of a supported location (us, gb, de,...; uk = gb). Default us. |
lang | string | no | Language code offered for the country (en, es, de,...; en-US = en). Default: en when offered, else the country's main language. |
includeMonthly | string | no | true adds each keyword's monthly search volumes. One of: true, false. |
Example request
curl -i 'https://keywords.duoleads.com/v1/search-volume?keywords=phone%2Cwatch&country=us&lang=en'
Example answer (200)
{
"total": 2,
"results": [
{
"keyword": "phone",
"volume": 673000,
"cpc": 4.63,
"competition": 0.76,
"competitionLevel": "HIGH",
"difficulty": 54,
"intent": "informational",
"trendYoyPct": 50
},
{
"keyword": "watch",
"volume": 550000,
"cpc": 1.39,
"competition": 0.95,
"competitionLevel": "HIGH",
"difficulty": 54,
"intent": "transactional",
"trendYoyPct": 22
}
],
"missing": [],
"market": {
"locationCode": 2840,
"locationName": "United States",
"country": "US",
"languageCode": "en",
"fallback": false
},
"source": "Google Ads data",
"dataAsOf": "2026-09-16T03:19:59Z",
"units": "volume: average monthly Google searches; cpc: US dollars; competition: 0-1 (Google Ads); competitionLevel: LOW|MEDIUM|HIGH; difficulty: 0-100; intent: informational|navigational|commercial|transactional; trendYoyPct: % change vs a year ago",
"see_also": {
"endpoint": "https://keywords.duoleads.com/v1/keyword-overview",
"params": [
"keywords",
"locationCode",
"languageCode"
],
"reason": "More than 20 keywords? POST up to 100 to keyword-overview: the same fields in one $0.04 call."
},
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/keyword-ideas ($0.04)
Keyword ideas for a seed keyword, each with Google search volume, CPC, competition, keyword difficulty and search intent; up to 100 ideas in one $0.04 call (Google Ads data). Use when you need related keywords to target. volume = avg monthly searches, cpc in USD, competition 0-1, difficulty 0-100. No ideas returns empty results (still charged). Bad input is free; upstream errors are not charged.
MCP tool: keyword_ideas. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
keyword | string | yes | Seed keyword to expand (at most 80 characters). Whitespace and control characters are normalised to single spaces and the seed to lower case; only a seed of 1 to 10 words and at most 80 characters can be expanded, so a seed that is then empty or longer is a free 400. |
limit | integer | no | Maximum ideas to return. Defaults to 20. |
locationCode | integer | no | Google Ads location code (geo target ID). Defaults to 2840 (United States); an unsupported code falls back to 2840 (United States, market.fallback). |
languageCode | string | no | Language code (2-letter). Defaults to en. An unsupported location falls back to 2840 (United States) with en (or the requested language if the United States offers it); a supported location with a language it does not offer uses that location's default language (en if offered). Either way market.fallback is true and market.warnings says what was used. |
country | string | no | Optional ISO 3166-1 alpha-2 code or country name, used when locationCode is absent (uk = gb). |
includeMonthly | boolean | no | Add each idea's monthly search volumes (monthly[]). |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/keyword-ideas' -H 'content-type: application/json' -d '{"keyword":"phone","limit":2,"locationCode":2840,"languageCode":"en"}'
Example answer (200)
{
"seed": "phone",
"total": 2,
"results": [
{
"keyword": "nothing phone",
"volume": 165000,
"cpc": 0.31,
"competition": 0.92,
"competitionLevel": "HIGH",
"difficulty": 31,
"intent": "informational",
"trendYoyPct": 174
},
{
"keyword": "find phone",
"volume": 14800,
"cpc": 2.14,
"competition": 0.1,
"competitionLevel": "LOW",
"difficulty": 57,
"intent": "informational",
"trendYoyPct": 22
}
],
"available": 425645,
"market": {
"locationCode": 2840,
"locationName": "United States",
"country": "US",
"languageCode": "en",
"fallback": false
},
"source": "Google Ads data",
"dataAsOf": "2026-09-13T09:55:39Z",
"units": "volume: average monthly Google searches; cpc: US dollars; competition: 0-1 (Google Ads); competitionLevel: LOW|MEDIUM|HIGH; difficulty: 0-100; intent: informational|navigational|commercial|transactional; trendYoyPct: % change vs a year ago",
"see_also": {
"endpoint": "https://keywords.duoleads.com/v1/trend",
"params": [
"keyword",
"country",
"timeframe"
],
"reason": "Want the seasonality of an idea? Call trend for its Google Trends interest over time."
},
"request_id": "01KEXAMPLE0000000000000000"
}
GET /v1/trend ($0.045)
Google Trends interest over time for one keyword: a dated 0-100 series for a country or worldwide, past 7 days to 5 years, in one $0.045 GET. Use when you need search-interest trends, seasonality or demand direction for a keyword. Daily points for short windows, weekly for 12 months. A live call takes up to about 30 s. Bad input is free; upstream errors are not charged. Too little Google data gives an empty answer, which is still charged.
MCP tool: google_trends. Input in the query string.
| Parameter | Type | Required | Description |
|---|---|---|---|
keyword | string | yes | One keyword or phrase (2-100 characters, no commas). |
country | string | no | ISO 3166-1 alpha-2 country code (us, gb, de,...; uk = gb). Omit for worldwide. |
timeframe | string | no | Time window. Default past_12_months. One of: past_7_days, past_30_days, past_90_days, past_12_months, past_5_years. |
Example request
curl -i 'https://keywords.duoleads.com/v1/trend?keyword=seo+tools&country=us&timeframe=past_12_months'
Example answer (200)
{
"keyword": "seo tools",
"country": "US",
"queried_at": "2026-10-11T19:24:11Z",
"timeframe": "past_12_months",
"granularity": "week",
"series": [
{
"date": "2025-10-05",
"value": 62
},
{
"date": "2025-10-12",
"value": 45
},
{
"date": "2025-10-19",
"value": 41
},
{
"date": "2026-10-11",
"value": 11,
"partial": true
}
],
"market": {
"country": "US",
"locationCode": 2840,
"locationName": "United States"
},
"source": "Google Trends",
"dataAsOf": "2026-10-11T19:24:11Z",
"units": "value: Google Trends interest, 0-100 relative to the peak of the period (100 = peak, 0 = too little data); partial: the period is not complete yet",
"see_also": {
"endpoint": "https://keywords.duoleads.com/v1/related-queries",
"params": [
"keyword",
"country",
"timeframe"
],
"reason": "Want the top and rising searches related to this keyword? Call related-queries with the same parameters."
},
"request_id": "01KEXAMPLE0000000000000000"
}
GET /v1/related-queries ($0.03)
Google Trends related queries for one keyword: the top related searches (0-100) and the rising ones (% growth), for a country or worldwide, past 7 days to 5 years, in one $0.03 GET. Use when you need query expansion, content ideas or emerging searches around a keyword. A live call takes up to about 30 s. Bad input is free; upstream errors are not charged. Too little Google data gives an empty answer, which is still charged.
MCP tool: google_trends_related_queries. Input in the query string.
| Parameter | Type | Required | Description |
|---|---|---|---|
keyword | string | yes | One keyword or phrase (2-100 characters, no commas). |
country | string | no | ISO 3166-1 alpha-2 country code (us, gb, de,...; uk = gb). Omit for worldwide. |
timeframe | string | no | Time window. Default past_12_months. One of: past_7_days, past_30_days, past_90_days, past_12_months, past_5_years. |
Example request
curl -i 'https://keywords.duoleads.com/v1/related-queries?keyword=seo+api&country=us&timeframe=past_12_months'
Example answer (200)
{
"keyword": "seo api",
"country": "US",
"queried_at": "2022-04-21T18:40:17Z",
"timeframe": "past_12_months",
"top": [
{
"term": "analytics seo api",
"value": 100
},
{
"term": "moz api",
"value": 64
}
],
"rising": [],
"market": {
"country": "US",
"locationCode": 2840,
"locationName": "United States"
},
"source": "Google Trends",
"dataAsOf": "2022-04-21T18:40:17Z",
"units": "top value: 0-100 relative popularity among related queries (100 = most searched); rising value: % increase in searches vs the preceding period",
"see_also": {
"endpoint": "https://keywords.duoleads.com/v1/trend",
"params": [
"keyword",
"country",
"timeframe"
],
"reason": "Need the interest-over-time series for this keyword? Call trend with the same parameters."
},
"request_id": "01KEXAMPLE0000000000000000"
}
GET /v1/interest-by-region ($0.03)
Google Trends interest by region for one keyword: 0-100 interest per subregion of a country (or per country worldwide), past 7 days to 5 years, in one $0.03 GET. Use when you need where a keyword is searched most, for geo-targeting or local market research. A live call takes up to about 30 s. Bad input is free; upstream errors are not charged. Too little Google data gives an empty answer, which is still charged.
MCP tool: google_trends_by_region. Input in the query string.
| Parameter | Type | Required | Description |
|---|---|---|---|
keyword | string | yes | One keyword or phrase (2-100 characters, no commas). |
country | string | no | ISO 3166-1 alpha-2 country code (us, gb, de,...; uk = gb). Omit for worldwide. |
timeframe | string | no | Time window. Default past_12_months. One of: past_7_days, past_30_days, past_90_days, past_12_months, past_5_years. |
Example request
curl -i 'https://keywords.duoleads.com/v1/interest-by-region?keyword=seo+api&country=us&timeframe=past_12_months'
Example answer (200)
{
"keyword": "seo api",
"country": "US",
"queried_at": "2022-04-21T18:40:17Z",
"timeframe": "past_12_months",
"resolution": "region",
"regions": [
{
"region": "New York",
"code": "US-NY",
"value": 100
},
{
"region": "California",
"code": "US-CA",
"value": 37
},
{
"region": "Oregon",
"code": "US-OR",
"value": 31
},
{
"region": "Texas",
"code": "US-TX",
"value": 17
}
],
"regionsWithoutData": 3,
"market": {
"country": "US",
"locationCode": 2840,
"locationName": "United States"
},
"source": "Google Trends",
"dataAsOf": "2022-04-21T18:40:17Z",
"units": "value: Google Trends interest, 0-100 relative to the region with the highest interest; regions with too little data are left out (regionsWithoutData)",
"see_also": {
"endpoint": "https://keywords.duoleads.com/v1/trend",
"params": [
"keyword",
"country",
"timeframe"
],
"reason": "Need how interest changed over time in one country? Call trend with that country."
},
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/related-keywords ($0.04)
Related keywords for a seed from Google's 'searches related to' graph, to a depth of 0-4, each with search volume, CPC, competition, keyword difficulty and search intent; up to 100 in one $0.04 call (Google Ads data). Use when you need topic clusters around a keyword. Each row has its depth and the further related searches. No matches returns empty results (still charged). Bad input is free; upstream errors are not charged.
MCP tool: related_keywords. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
keyword | string | yes | Seed keyword (at most 80 characters; whitespace, control characters and case are normalised; a seed then empty, over 10 words or over 80 characters is a free 400). |
depth | integer | no | How many levels of the related-searches graph to traverse (0-4). Defaults to 1. |
limit | integer | no | Maximum keywords to return (1-100). Defaults to 20. |
locationCode | integer | no | Google Ads location code (geo target ID). Defaults to 2840 (United States); an unsupported code falls back to 2840 (United States, market.fallback). |
languageCode | string | no | Language code (2-letter). Defaults to en. An unsupported location falls back to 2840 (United States) with en (or the requested language if the United States offers it); a supported location with a language it does not offer uses that location's default language (en if offered). Either way market.fallback is true and market.warnings says what was used. |
country | string | no | Optional ISO 3166-1 alpha-2 code or country name, used when locationCode is absent (uk = gb). |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/related-keywords' -H 'content-type: application/json' -d '{"keyword":"running shoes","depth":1,"limit":2,"locationCode":2840,"languageCode":"en"}'
Example answer (200)
{
"seed": "running shoes",
"depth": 1,
"total": 2,
"results": [
{
"keyword": "running shoes",
"volume": 201000,
"cpc": 1.12,
"competition": 1,
"competitionLevel": "HIGH",
"difficulty": 61,
"intent": "commercial",
"trendYoyPct": 22,
"depth": 0,
"relatedKeywords": [
"best running shoes",
"running shoes for women"
]
},
{
"keyword": "best running shoes",
"volume": 60500,
"cpc": 0.85,
"competition": 1,
"competitionLevel": "HIGH",
"difficulty": 58,
"intent": "commercial",
"trendYoyPct": 0,
"depth": 1,
"relatedKeywords": [
"best running shoes for men"
]
}
],
"available": 9,
"market": {
"locationCode": 2840,
"locationName": "United States",
"country": "US",
"languageCode": "en",
"fallback": false
},
"source": "Google Ads data",
"dataAsOf": "2026-10-19T17:25:00Z",
"units": "volume: average monthly Google searches; cpc: US dollars; competition: 0-1 (Google Ads); competitionLevel: LOW|MEDIUM|HIGH; difficulty: 0-100; intent: informational|navigational|commercial|transactional; trendYoyPct: % change vs a year ago; depth: steps from the seed",
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/keyword-variations ($0.04)
Keyword variations: long-tail Google searches that contain a seed phrase, each with search volume, CPC, competition, keyword difficulty and search intent; up to 100 in one $0.04 call (Google Ads data). Use when you need phrase-match ideas for a keyword, sorted by volume. No matches returns empty results (still charged). Bad input is free; upstream errors are not charged.
MCP tool: keyword_variations. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
keyword | string | yes | Seed phrase; every result contains it (at most 80 characters; normalised like keyword-ideas). |
limit | integer | no | Maximum variations to return (1-100). Defaults to 20. |
locationCode | integer | no | Google Ads location code (geo target ID). Defaults to 2840 (United States); an unsupported code falls back to 2840 (United States, market.fallback). |
languageCode | string | no | Language code (2-letter). Defaults to en. An unsupported location falls back to 2840 (United States) with en (or the requested language if the United States offers it); a supported location with a language it does not offer uses that location's default language (en if offered). Either way market.fallback is true and market.warnings says what was used. |
country | string | no | Optional ISO 3166-1 alpha-2 code or country name, used when locationCode is absent (uk = gb). |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/keyword-variations' -H 'content-type: application/json' -d '{"keyword":"running shoes","limit":2,"locationCode":2840,"languageCode":"en"}'
Example answer (200)
{
"seed": "running shoes",
"total": 2,
"results": [
{
"keyword": "best running shoes",
"volume": 60500,
"cpc": 0.85,
"competition": 1,
"competitionLevel": "HIGH",
"difficulty": 58,
"intent": "commercial",
"trendYoyPct": 0
},
{
"keyword": "running shoes for women",
"volume": 33100,
"cpc": 0.95,
"competition": 1,
"competitionLevel": "HIGH",
"difficulty": 47,
"intent": "transactional",
"trendYoyPct": 22
}
],
"available": 1088,
"market": {
"locationCode": 2840,
"locationName": "United States",
"country": "US",
"languageCode": "en",
"fallback": false
},
"source": "Google Ads data",
"dataAsOf": "2026-10-17T08:27:20Z",
"units": "volume: average monthly Google searches; cpc: US dollars; competition: 0-1 (Google Ads); competitionLevel: LOW|MEDIUM|HIGH; difficulty: 0-100; intent: informational|navigational|commercial|transactional; trendYoyPct: % change vs a year ago",
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/keyword-difficulty ($0.04)
Keyword difficulty (0-100: how hard it is to rank in Google's top 10) for up to 100 keywords in one $0.04 call. Use when you need to prioritise keywords by ranking difficulty. Keywords are normalised to lower case; over 100 keywords is a free 400. Keywords without a score are listed in missing (still charged). Bad input is free; upstream errors are not charged.
MCP tool: keyword_difficulty. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
keywords | array of strings | yes | Keywords to score (1-100; each at most 80 characters, normalised to lower case; one that is then empty, over 10 words or over 80 characters is listed in missing). |
locationCode | integer | no | Google Ads location code (geo target ID). Defaults to 2840 (United States); an unsupported code falls back to 2840 (United States, market.fallback). |
languageCode | string | no | Language code (2-letter). Defaults to en. An unsupported location falls back to 2840 (United States) with en (or the requested language if the United States offers it); a supported location with a language it does not offer uses that location's default language (en if offered). Either way market.fallback is true and market.warnings says what was used. |
country | string | no | Optional ISO 3166-1 alpha-2 code or country name, used when locationCode is absent (uk = gb). |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/keyword-difficulty' -H 'content-type: application/json' -d '{"keywords":["running shoes","trail running shoes"],"locationCode":2840,"languageCode":"en"}'
Example answer (200)
{
"total": 2,
"results": [
{
"keyword": "running shoes",
"difficulty": 61
},
{
"keyword": "trail running shoes",
"difficulty": 38
}
],
"missing": [],
"market": {
"locationCode": 2840,
"locationName": "United States",
"country": "US",
"languageCode": "en",
"fallback": false
},
"source": "Google Ads data",
"units": "difficulty: 0-100, how hard it is to rank in Google's top 10 for the keyword",
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/search-intent ($0.04)
Search intent of up to 100 keywords in one $0.04 call: the main intent (informational, navigational, commercial or transactional) with its probability, plus secondary intents. Use when you need to map keywords to funnel stage or content type. Any language; no location needed. Over 100 keywords is a free 400. Keywords not classified are listed in missing (still charged). Bad input is free; upstream errors are not charged.
MCP tool: search_intent. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
keywords | array of strings | yes | Keywords to classify (1-100; each at most 80 characters, normalised to lower case; one that is then empty, over 10 words or over 80 characters is listed in missing). |
languageCode | string | no | Accepted for compatibility and ignored: the classifier detects the language itself. |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/search-intent' -H 'content-type: application/json' -d '{"keywords":["buy running shoes","how to clean running shoes"]}'
Example answer (200)
{
"total": 2,
"results": [
{
"keyword": "buy running shoes",
"intent": "transactional",
"probability": 0.92,
"secondaryIntents": [
{
"label": "commercial",
"probability": 0.31
}
]
},
{
"keyword": "how to clean running shoes",
"intent": "informational",
"probability": 0.99
}
],
"missing": [],
"source": "Google search results (intent model)",
"units": "intent: informational|navigational|commercial|transactional; probability: 0-1 (model confidence)",
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/keyword-volume-history ($0.048)
Monthly Google search volume history for up to 20 keywords in one $0.048 call: every month the data has (typically several years), oldest first, plus current volume, CPC and competition (Google Ads data). Use when you need seasonality or long-term demand trends. Over 20 keywords is a free 400. Keywords without data are listed in missing (still charged). Bad input is free; upstream errors are not charged.
MCP tool: keyword_volume_history. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
keywords | array of strings | yes | Keywords to look up (1-20; each at most 80 characters, normalised to lower case; one that is then empty, over 10 words or over 80 characters is listed in missing). |
locationCode | integer | no | Google Ads location code (geo target ID). Defaults to 2840 (United States); an unsupported code falls back to 2840 (United States, market.fallback). |
languageCode | string | no | Language code (2-letter). Defaults to en. An unsupported location falls back to 2840 (United States) with en (or the requested language if the United States offers it); a supported location with a language it does not offer uses that location's default language (en if offered). Either way market.fallback is true and market.warnings says what was used. |
country | string | no | Optional ISO 3166-1 alpha-2 code or country name, used when locationCode is absent (uk = gb). |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/keyword-volume-history' -H 'content-type: application/json' -d '{"keywords":["running shoes","trail running shoes"],"locationCode":2840,"languageCode":"en"}'
Example answer (200)
{
"total": 2,
"results": [
{
"keyword": "running shoes",
"volume": 201000,
"cpc": 1.12,
"competition": 1,
"competitionLevel": "HIGH",
"history": [
{
"date": "2026-08",
"volume": 246000
},
{
"date": "2026-09",
"volume": 201000
}
]
},
{
"keyword": "trail running shoes",
"volume": 49500,
"cpc": 0.98,
"competition": 1,
"competitionLevel": "HIGH",
"history": [
{
"date": "2026-08",
"volume": 60500
},
{
"date": "2026-09",
"volume": 49500
}
]
}
],
"missing": [],
"market": {
"locationCode": 2840,
"locationName": "United States",
"country": "US",
"languageCode": "en",
"fallback": false
},
"source": "Google Ads data",
"dataAsOf": "2026-10-19T17:25:00Z",
"units": "history: average Google searches per month (date YYYY-MM), oldest first; volume: average monthly searches (last 12 months); cpc: US dollars; competition: 0-1 (Google Ads)",
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/keyword-suggestions ($0.04)
Keyword suggestions for a website or one page: the Google searches relevant to it, each with search volume, CPC, competition, keyword difficulty and search intent; up to 100 in one $0.04 call (Google Ads data). Use when you need keyword ideas from a domain (www. is dropped) or a page URL. A site with no data returns empty results (still charged). Bad input is free; upstream errors are not charged.
MCP tool: keywords_for_site. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
target | string | yes | targetType site: a bare domain like example.com (no scheme or path; a leading www. is dropped). targetType page: an absolute http(s) page URL. |
targetType | string | no | Suggest keywords for the whole site (default) or only the target page. One of: site, page. |
limit | integer | no | Maximum keywords to return (1-100). Defaults to 20. |
locationCode | integer | no | Google Ads location code (geo target ID). Defaults to 2840 (United States); an unsupported code falls back to 2840 (United States, market.fallback). |
languageCode | string | no | Language code (2-letter). Defaults to en. An unsupported location falls back to 2840 (United States) with en (or the requested language if the United States offers it); a supported location with a language it does not offer uses that location's default language (en if offered). Either way market.fallback is true and market.warnings says what was used. |
country | string | no | Optional ISO 3166-1 alpha-2 code or country name, used when locationCode is absent (uk = gb). |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/keyword-suggestions' -H 'content-type: application/json' -d '{"target":"example.com","targetType":"site","limit":2,"locationCode":2840,"languageCode":"en"}'
Example answer (200)
{
"target": "example.com",
"targetType": "site",
"total": 2,
"results": [
{
"keyword": "example domain",
"volume": 2400,
"cpc": 0.42,
"competition": 0.03,
"competitionLevel": "LOW",
"difficulty": 12,
"intent": "informational",
"trendYoyPct": 0
},
{
"keyword": "example website",
"volume": 1300,
"cpc": 0.65,
"competition": 0.05,
"competitionLevel": "LOW",
"difficulty": 21,
"intent": "navigational",
"trendYoyPct": -18
}
],
"available": 140,
"market": {
"locationCode": 2840,
"locationName": "United States",
"country": "US",
"languageCode": "en",
"fallback": false
},
"source": "Google Ads data",
"dataAsOf": "2026-10-15T07:48:51Z",
"units": "volume: average monthly Google searches; cpc: US dollars; competition: 0-1 (Google Ads); competitionLevel: LOW|MEDIUM|HIGH; difficulty: 0-100; intent: informational|navigational|commercial|transactional; trendYoyPct: % change vs a year ago",
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/trends ($0.045)
Google Trends comparison of up to 5 keywords: one relative 0-100 interest series for all of them (same scale), with each keyword's average and latest value; worldwide or one country, 7 days to 5 years, web, news or YouTube search, in one $0.045 call. Use when you need to compare demand or seasonality between terms. A live call takes up to about 30 s. Too little data gives found false (still charged). Bad input is free; upstream errors are not charged.
MCP tool: google_trends_compare. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
keywords | array of strings | yes | 1-5 terms to compare, e.g. ["running shoes", "trail running shoes"] (each 2-100 characters, no commas). |
location | string | no | Market: a country name ("United States") or ISO 3166-1 alpha-2 code (us, gb, de; uk = gb). Omit for worldwide. |
timeframe | string | no | Time window: 7d, 30d, 90d, 12m or 5y (default 12m); the past_* names of /v1/trend work too. One of: 7d, 30d, 90d, 12m, 5y, past_7_days, past_30_days, past_90_days, past_12_months, past_5_years. |
type | string | no | Google property: web (default), news or youtube. One of: web, news, youtube. |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/trends' -H 'content-type: application/json' -d '{"keywords":["running shoes","trail running shoes"],"location":"United States","timeframe":"12m","type":"web"}'
Example answer (200)
{
"keywords": [
"running shoes",
"trail running shoes"
],
"found": true,
"timeframe": "past_12_months",
"type": "web",
"granularity": "week",
"interest": [
{
"date": "2025-10-05",
"values": [
72,
21
]
},
{
"date": "2025-10-12",
"values": [
70,
20
]
},
{
"date": "2026-10-11",
"values": [
81,
25
],
"partial": true
}
],
"averages": [
74.3,
22
],
"latest": [
81,
25
],
"market": {
"country": "US",
"locationCode": 2840,
"locationName": "United States"
},
"source": "Google Trends",
"dataAsOf": "2026-10-11T19:24:11Z",
"units": "values: Google Trends interest, 0-100 relative to the highest point of all compared keywords in the period, in keywords order; partial: the period is not complete yet; averages / latest: per keyword",
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/domain-overview ($0.04)
Domain overview: a domain's Google SEO footprint in one market: organic and paid ranking keywords, estimated monthly traffic (etv) and its ad value, and keywords in position 1, top 3, top 10 and top 100, in one $0.04 call. Use when you need a quick SEO size-up of a website or competitor (bare domain; www. is dropped). A domain with no data gives found false (still charged). Bad input is free; upstream errors are not charged.
MCP tool: domain_overview. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
target | string | yes | Bare domain like example.com (no scheme, path or port; a leading www. is dropped). |
locationCode | integer | no | Google Ads location code (geo target ID). Defaults to 2840 (United States); an unsupported code falls back to 2840 (United States, market.fallback). |
languageCode | string | no | Language code (2-letter). Defaults to en. An unsupported location falls back to 2840 (United States) with en (or the requested language if the United States offers it); a supported location with a language it does not offer uses that location's default language (en if offered). Either way market.fallback is true and market.warnings says what was used. |
country | string | no | Optional ISO 3166-1 alpha-2 code or country name, used when locationCode is absent (uk = gb). |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/domain-overview' -H 'content-type: application/json' -d '{"target":"example.com","locationCode":2840,"languageCode":"en"}'
Example answer (200)
{
"target": "example.com",
"found": true,
"organic": {
"keywords": 8703,
"etv": 15860.74,
"trafficCostUsd": 89125.12,
"pos1": 40,
"top3": 182,
"top10": 1279,
"top100": 8703
},
"paid": {
"keywords": 0,
"etv": 0,
"trafficCostUsd": 0,
"pos1": 0,
"top3": 0,
"top10": 0,
"top100": 0
},
"market": {
"locationCode": 2840,
"locationName": "United States",
"country": "US",
"languageCode": "en",
"fallback": false
},
"source": "Google search results (estimates)",
"units": "keywords: ranking keywords; etv: estimated monthly organic (or paid) visits from Google; trafficCostUsd: estimated monthly cost of that traffic as ads, US dollars; pos1/top3/top10/top100: keywords ranking at position 1 / 1-3 / 1-10 / 1-100",
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/traffic-estimate ($0.04)
Traffic estimate for up to 100 domains in one $0.04 call: each domain's estimated monthly organic and paid visits from Google (etv) and its count of ranking keywords. Use when you need to compare the search traffic of many websites at once (bare domains; www. is dropped). A domain with no data gives found false (still charged). Bad input is free; upstream errors are not charged.
MCP tool: traffic_estimate. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
targets | array of strings | yes | Bare domains (1-100) like example.com; no scheme or path; a leading www. is dropped. |
locationCode | integer | no | Google Ads location code (geo target ID). Defaults to 2840 (United States); an unsupported code falls back to 2840 (United States, market.fallback). |
languageCode | string | no | Language code (2-letter). Defaults to en. An unsupported location falls back to 2840 (United States) with en (or the requested language if the United States offers it); a supported location with a language it does not offer uses that location's default language (en if offered). Either way market.fallback is true and market.warnings says what was used. |
country | string | no | Optional ISO 3166-1 alpha-2 code or country name, used when locationCode is absent (uk = gb). |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/traffic-estimate' -H 'content-type: application/json' -d '{"targets":["example.com","example.org"],"locationCode":2840,"languageCode":"en"}'
Example answer (200)
{
"total": 2,
"results": [
{
"target": "example.com",
"found": true,
"organicEtv": 28068.67,
"organicKeywords": 31554,
"paidEtv": 80.66,
"paidKeywords": 14
},
{
"target": "example.org",
"found": true,
"organicEtv": 1210.4,
"organicKeywords": 2210,
"paidEtv": 0,
"paidKeywords": 0
}
],
"market": {
"locationCode": 2840,
"locationName": "United States",
"country": "US",
"languageCode": "en",
"fallback": false
},
"source": "Google search results (estimates)",
"units": "organicEtv / paidEtv: estimated monthly visits from Google organic / paid results; organicKeywords / paidKeywords: ranking keywords",
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/ranked-keywords ($0.048)
Ranked keywords: the Google keywords a domain ranks for, each with its position, ranking URL, estimated traffic and the keyword's search volume, CPC, difficulty and intent; up to 100 in one $0.048 call. Use when you need a site's or competitor's organic keywords (bare domain; www. is dropped). A domain with no data returns empty results (still charged). Bad input is free; upstream errors are not charged.
MCP tool: ranked_keywords. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
target | string | yes | Bare domain like example.com (no scheme, path or port; a leading www. is dropped). |
limit | integer | no | Maximum keywords to return (1-100). Defaults to 20. |
locationCode | integer | no | Google Ads location code (geo target ID). Defaults to 2840 (United States); an unsupported code falls back to 2840 (United States, market.fallback). |
languageCode | string | no | Language code (2-letter). Defaults to en. An unsupported location falls back to 2840 (United States) with en (or the requested language if the United States offers it); a supported location with a language it does not offer uses that location's default language (en if offered). Either way market.fallback is true and market.warnings says what was used. |
country | string | no | Optional ISO 3166-1 alpha-2 code or country name, used when locationCode is absent (uk = gb). |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/ranked-keywords' -H 'content-type: application/json' -d '{"target":"example.com","limit":2,"locationCode":2840,"languageCode":"en"}'
Example answer (200)
{
"target": "example.com",
"total": 2,
"results": [
{
"keyword": "example domain",
"position": 1,
"rankAbsolute": 1,
"type": "organic",
"url": "https://example.com/",
"etv": 312.4,
"volume": 2400,
"cpc": 0.42,
"competition": 0.03,
"competitionLevel": "LOW",
"difficulty": 12,
"intent": "informational",
"trendYoyPct": 0
},
{
"keyword": "example website",
"position": 3,
"rankAbsolute": 4,
"type": "organic",
"url": "https://example.com/about",
"etv": 51.2,
"volume": 1300,
"cpc": 0.65,
"competition": 0.05,
"competitionLevel": "LOW",
"difficulty": 21,
"intent": "navigational",
"trendYoyPct": -18
}
],
"available": 8806,
"market": {
"locationCode": 2840,
"locationName": "United States",
"country": "US",
"languageCode": "en",
"fallback": false
},
"source": "Google search results (estimates) and Google Ads data",
"dataAsOf": "2026-10-19T17:25:00Z",
"units": "position: rank among Google's organic results; rankAbsolute: rank among all result elements; etv: estimated monthly visits from this ranking; volume: average monthly Google searches; cpc: US dollars; competition: 0-1 (Google Ads); competitionLevel: LOW|MEDIUM|HIGH; difficulty: 0-100; intent: informational|navigational|commercial|transactional; trendYoyPct: % change vs a year ago",
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/competitors ($0.048)
Competitor domains: the websites competing with a domain in Google organic search, each with the keywords both rank for, the average position on them, and the competitor's total ranking keywords and estimated traffic; up to 100 in one $0.048 call. Use when you need to find a site's real SEO competitors (bare domain; www. is dropped). No data returns empty results (still charged). Bad input is free; upstream errors are not charged.
MCP tool: competitor_domains. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
target | string | yes | Bare domain like example.com (no scheme, path or port; a leading www. is dropped). |
limit | integer | no | Maximum competitors to return (1-100). Defaults to 20. |
locationCode | integer | no | Google Ads location code (geo target ID). Defaults to 2840 (United States); an unsupported code falls back to 2840 (United States, market.fallback). |
languageCode | string | no | Language code (2-letter). Defaults to en. An unsupported location falls back to 2840 (United States) with en (or the requested language if the United States offers it); a supported location with a language it does not offer uses that location's default language (en if offered). Either way market.fallback is true and market.warnings says what was used. |
country | string | no | Optional ISO 3166-1 alpha-2 code or country name, used when locationCode is absent (uk = gb). |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/competitors' -H 'content-type: application/json' -d '{"target":"example.com","limit":2,"locationCode":2840,"languageCode":"en"}'
Example answer (200)
{
"target": "example.com",
"total": 2,
"results": [
{
"domain": "example.org",
"commonKeywords": 115,
"avgPosition": 13.84,
"keywords": 151186,
"etv": 3055537.62
},
{
"domain": "example.net",
"commonKeywords": 64,
"avgPosition": 21.5,
"keywords": 20410,
"etv": 88120.5
}
],
"available": 993,
"market": {
"locationCode": 2840,
"locationName": "United States",
"country": "US",
"languageCode": "en",
"fallback": false
},
"source": "Google search results (estimates)",
"units": "commonKeywords: keywords both domains rank for; avgPosition: the competitor's average Google position on them; keywords / etv: the competitor's ranking keywords and estimated monthly organic visits",
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/keyword-gap ($0.048)
Keyword gap: the Google keywords a competitor ranks for and your domain does not, each with the competitor's position, URL and estimated traffic plus search volume, CPC, difficulty and intent; up to 100 in one $0.048 call. Use when you need content gaps and keyword opportunities (bare domains; www. is dropped). No gap returns empty results (still charged). Bad input is free; upstream errors are not charged.
MCP tool: keyword_gap. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
domain | string | yes | Your domain, e.g. example.com (no scheme or path; a leading www. is dropped). |
competitor | string | yes | The competitor domain to compare against, e.g. example.org. |
limit | integer | no | Maximum gap keywords to return (1-200; above 100 is lowered to 100 with a warning). Defaults to 20. |
locationCode | integer | no | Google Ads location code (geo target ID). Defaults to 2840 (United States); an unsupported code falls back to 2840 (United States, market.fallback). |
languageCode | string | no | Language code (2-letter). Defaults to en. An unsupported location falls back to 2840 (United States) with en (or the requested language if the United States offers it); a supported location with a language it does not offer uses that location's default language (en if offered). Either way market.fallback is true and market.warnings says what was used. |
country | string | no | Optional ISO 3166-1 alpha-2 code or country name, used when locationCode is absent (uk = gb). |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/keyword-gap' -H 'content-type: application/json' -d '{"domain":"example.com","competitor":"example.org","limit":2,"locationCode":2840,"languageCode":"en"}'
Example answer (200)
{
"domain": "example.com",
"competitor": "example.org",
"total": 2,
"results": [
{
"keyword": "example nonprofit",
"competitorRank": 2,
"competitorRankAbsolute": 4,
"competitorUrl": "https://example.org/about",
"competitorEtv": 2.4,
"volume": 40,
"competition": 0.06,
"competitionLevel": "LOW",
"difficulty": 0,
"intent": "informational",
"trendYoyPct": -67
},
{
"keyword": "example organization",
"competitorRank": 5,
"competitorRankAbsolute": 7,
"competitorUrl": "https://example.org/",
"competitorEtv": 1.1,
"volume": 30,
"difficulty": 3,
"intent": "navigational"
}
],
"available": 147,
"market": {
"locationCode": 2840,
"locationName": "United States",
"country": "US",
"languageCode": "en",
"fallback": false
},
"source": "Google search results (estimates) and Google Ads data",
"dataAsOf": "2026-10-19T17:25:00Z",
"units": "competitorRank: the competitor's rank among Google's organic results; competitorEtv: its estimated monthly visits from that ranking; volume: average monthly Google searches; cpc: US dollars; competition: 0-1 (Google Ads); competitionLevel: LOW|MEDIUM|HIGH; difficulty: 0-100; intent: informational|navigational|commercial|transactional; trendYoyPct: % change vs a year ago",
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/serp-competitors ($0.048)
SERP competitors: the domains that rank in Google for a set of up to 100 keywords, each with average and median position, visibility, estimated traffic and how many of the keywords it ranks for; up to 100 domains in one $0.048 call. Use when you need who owns the search results of a topic. No data returns empty results (still charged). Bad input is free; upstream errors are not charged.
MCP tool: serp_competitors. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
keywords | array of strings | yes | Keywords whose Google results to analyse (1-100; each at most 80 characters, normalised to lower case; one that is then empty, over 10 words or over 80 characters is listed in missing). |
limit | integer | no | Maximum domains to return (1-100). Defaults to 20. |
locationCode | integer | no | Google Ads location code (geo target ID). Defaults to 2840 (United States); an unsupported code falls back to 2840 (United States, market.fallback). |
languageCode | string | no | Language code (2-letter). Defaults to en. An unsupported location falls back to 2840 (United States) with en (or the requested language if the United States offers it); a supported location with a language it does not offer uses that location's default language (en if offered). Either way market.fallback is true and market.warnings says what was used. |
country | string | no | Optional ISO 3166-1 alpha-2 code or country name, used when locationCode is absent (uk = gb). |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/serp-competitors' -H 'content-type: application/json' -d '{"keywords":["running shoes","trail running shoes"],"limit":2,"locationCode":2840,"languageCode":"en"}'
Example answer (200)
{
"keywords": [
"running shoes",
"trail running shoes"
],
"total": 2,
"results": [
{
"domain": "example.com",
"avgPosition": 1.5,
"medianPosition": 1.5,
"rating": 188,
"etv": 21840,
"keywordsCount": 2,
"visibility": 0.92,
"relevantSerpItems": 2
},
{
"domain": "example.org",
"avgPosition": 4,
"medianPosition": 4,
"rating": 92,
"etv": 5120.5,
"keywordsCount": 1,
"visibility": 0.41,
"relevantSerpItems": 1
}
],
"available": 86,
"missing": [],
"market": {
"locationCode": 2840,
"locationName": "United States",
"country": "US",
"languageCode": "en",
"fallback": false
},
"source": "Google search results (estimates)",
"units": "avgPosition / medianPosition: Google position across the keywords; visibility: 0-1 share of the keywords' results; etv: estimated monthly visits from these keywords; keywordsCount: keywords the domain ranks for; rating: provider ranking score",
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/relevant-pages ($0.048)
Relevant pages: a domain's pages that get the most Google search traffic, each with its organic and paid ranking keywords, estimated monthly traffic, ad value and top-3/10/100 counts; up to 100 pages in one $0.048 call. Use when you need a site's top SEO pages (bare domain; www. is dropped). A domain with no data returns empty results (still charged). Bad input is free; upstream errors are not charged.
MCP tool: relevant_pages. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
target | string | yes | Bare domain like example.com (no scheme, path or port; a leading www. is dropped). |
limit | integer | no | Maximum pages to return (1-100). Defaults to 20. |
locationCode | integer | no | Google Ads location code (geo target ID). Defaults to 2840 (United States); an unsupported code falls back to 2840 (United States, market.fallback). |
languageCode | string | no | Language code (2-letter). Defaults to en. An unsupported location falls back to 2840 (United States) with en (or the requested language if the United States offers it); a supported location with a language it does not offer uses that location's default language (en if offered). Either way market.fallback is true and market.warnings says what was used. |
country | string | no | Optional ISO 3166-1 alpha-2 code or country name, used when locationCode is absent (uk = gb). |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/relevant-pages' -H 'content-type: application/json' -d '{"target":"example.com","limit":2,"locationCode":2840,"languageCode":"en"}'
Example answer (200)
{
"target": "example.com",
"total": 2,
"results": [
{
"page": "https://example.com/",
"organic": {
"keywords": 606,
"etv": 226.37,
"trafficCostUsd": 701.5,
"pos1": 2,
"top3": 9,
"top10": 58,
"top100": 606
}
},
{
"page": "https://example.com/about",
"organic": {
"keywords": 212,
"etv": 80.1,
"trafficCostUsd": 190.25,
"pos1": 0,
"top3": 3,
"top10": 17,
"top100": 212
}
}
],
"available": 79,
"market": {
"locationCode": 2840,
"locationName": "United States",
"country": "US",
"languageCode": "en",
"fallback": false
},
"source": "Google search results (estimates)",
"units": "keywords: ranking keywords; etv: estimated monthly organic (or paid) visits from Google; trafficCostUsd: estimated monthly cost of that traffic as ads, US dollars; pos1/top3/top10/top100: keywords ranking at position 1 / 1-3 / 1-10 / 1-100",
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/historical-rank ($0.25)
Historical rank: a domain's monthly Google visibility over the last 24 months: organic and paid ranking keywords, estimated traffic (etv), its ad value and position 1 / top 3 / top 10 / top 100 counts per month, in one $0.25 call. Use when you need a site's SEO trajectory, traffic drops or recoveries (bare domain; www. is dropped). A domain with no data returns an empty series (still charged). Bad input is free; upstream errors are not charged.
MCP tool: historical_rank. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
target | string | yes | Bare domain like example.com (no scheme, path or port; a leading www. is dropped). |
locationCode | integer | no | Google Ads location code (geo target ID). Defaults to 2840 (United States); an unsupported code falls back to 2840 (United States, market.fallback). |
languageCode | string | no | Language code (2-letter). Defaults to en. An unsupported location falls back to 2840 (United States) with en (or the requested language if the United States offers it); a supported location with a language it does not offer uses that location's default language (en if offered). Either way market.fallback is true and market.warnings says what was used. |
country | string | no | Optional ISO 3166-1 alpha-2 code or country name, used when locationCode is absent (uk = gb). |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/historical-rank' -H 'content-type: application/json' -d '{"target":"example.com","locationCode":2840,"languageCode":"en"}'
Example answer (200)
{
"target": "example.com",
"dateFrom": "2024-12-01",
"count": 2,
"series": [
{
"year": 2026,
"month": 9,
"organic": {
"keywords": 9310,
"etv": 30210.5,
"trafficCostUsd": 180110.2,
"pos1": 45,
"top3": 190,
"top10": 1302,
"top100": 9310
}
},
{
"year": 2026,
"month": 10,
"organic": {
"keywords": 9528,
"etv": 31868.48,
"trafficCostUsd": 193646.92,
"pos1": 48,
"top3": 196,
"top10": 1340,
"top100": 9528
}
}
],
"market": {
"locationCode": 2840,
"locationName": "United States",
"country": "US",
"languageCode": "en",
"fallback": false
},
"source": "Google search results (estimates)",
"units": "keywords: ranking keywords; etv: estimated monthly organic (or paid) visits from Google; trafficCostUsd: estimated monthly cost of that traffic as ads, US dollars; pos1/top3/top10/top100: keywords ranking at position 1 / 1-3 / 1-10 / 1-100; one entry per month, oldest first",
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/page-intersection ($0.048)
Page intersection: the Google keywords that up to 20 pages all rank for (minus keywords of up to 10 excluded pages), with each page's position, URL and estimated traffic plus search volume, CPC, difficulty and intent; up to 100 keywords in one $0.048 call. Use when you need the shared keywords of competing pages. Rows are capped at 200 / pages (2 pages: 100, 20 pages: 10). No shared keywords returns empty results (still charged). Bad input is free; upstream errors are not charged.
MCP tool: page_intersection. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
pages | array of strings | yes | Page URLs whose shared ranking keywords to find (1-20), e.g. ["https://example.com/guide", "https://example.org/guide"\]. |
excludePages | array of strings | no | Optional page URLs (1-10) to exclude: only keywords the pages rank for and these do not. |
limit | integer | no | Maximum keywords to return (1-200; lowered with a warning to 100, and to 200 / pages for more than 2 pages: 4 pages 50, 10 pages 20, 20 pages 10). Defaults to 20. |
locationCode | integer | no | Google Ads location code (geo target ID). Defaults to 2840 (United States); an unsupported code falls back to 2840 (United States, market.fallback). |
languageCode | string | no | Language code (2-letter). Defaults to en. An unsupported location falls back to 2840 (United States) with en (or the requested language if the United States offers it); a supported location with a language it does not offer uses that location's default language (en if offered). Either way market.fallback is true and market.warnings says what was used. |
country | string | no | Optional ISO 3166-1 alpha-2 code or country name, used when locationCode is absent (uk = gb). |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/page-intersection' -H 'content-type: application/json' -d '{"pages":["https://example.com/guide","https://example.org/guide"],"limit":2,"locationCode":2840,"languageCode":"en"}'
Example answer (200)
{
"pages": [
"https://example.com/guide",
"https://example.org/guide"
],
"total": 2,
"results": [
{
"keyword": "example guide",
"ranks": [
{
"page": 1,
"position": 2,
"rankAbsolute": 3,
"type": "organic",
"url": "https://example.com/guide",
"etv": 231
},
{
"page": 2,
"position": 5,
"rankAbsolute": 7,
"type": "organic",
"url": "https://example.org/guide",
"etv": 40.5
}
],
"volume": 1900,
"cpc": 1.2,
"competition": 0.02,
"competitionLevel": "LOW",
"difficulty": 41,
"intent": "informational",
"trendYoyPct": 22
},
{
"keyword": "example guide pdf",
"ranks": [
{
"page": 1,
"position": 4,
"rankAbsolute": 6,
"type": "organic",
"url": "https://example.com/guide",
"etv": 12.3
},
{
"page": 2,
"position": 9,
"rankAbsolute": 12,
"type": "organic",
"url": "https://example.org/guide",
"etv": 1.1
}
],
"volume": 320,
"difficulty": 18,
"intent": "informational"
}
],
"available": 147,
"market": {
"locationCode": 2840,
"locationName": "United States",
"country": "US",
"languageCode": "en",
"fallback": false
},
"source": "Google search results (estimates) and Google Ads data",
"dataAsOf": "2026-10-19T17:25:00Z",
"units": "ranks[].page: 1-based index into pages; position: rank among Google's organic results; etv: estimated monthly visits from that ranking; volume: average monthly Google searches; cpc: US dollars; competition: 0-1 (Google Ads); competitionLevel: LOW|MEDIUM|HIGH; difficulty: 0-100; intent: informational|navigational|commercial|transactional; trendYoyPct: % change vs a year ago",
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/backlinks ($0.048)
Backlink profile of a domain: domain rank (0-1000), backlinks, referring domains, main domains, IPs and subnets, nofollow and broken links, spam score, first seen, and breakdowns by link type, TLD and country, from a web link index, in one $0.048 call. Use when you need a site's link authority at a glance. A domain with no links is an answer with zeros or no counts, which is still charged. Bad input is a free 400; a data source failure is not charged.
MCP tool: backlinks_summary. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
target | string | yes | A domain like example.com (subdomains allowed). A site-root URL such as https://www.example.com/ is reduced to the domain; a path or port is refused. |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/backlinks' -H 'content-type: application/json' -d '{"target":"example.com"}'
Example answer (200)
{
"target": "example.com",
"domainRank": 512,
"backlinks": 28941,
"referringDomains": 3429,
"referringMainDomains": 3269,
"referringPages": 28256,
"referringIps": 2633,
"referringSubnets": 1495,
"nofollowReferringDomains": 312,
"brokenBacklinks": 215,
"brokenPages": 15,
"spamScore": 16,
"firstSeen": "2019-01-15T23:54:14Z",
"linkTypes": {
"anchor": 18723,
"image": 7230,
"redirect": 2985,
"canonical": 3
},
"topTlds": {
"com": 12682,
"de": 8292,
"net": 1551
},
"topCountries": {
"unknown": 15828,
"US": 8342,
"DE": 2651
},
"source": "web link index",
"units": "domainRank = 0-1000 (provider scale); counts = links, domains, pages, IPs or subnets; spamScore = 0-100; firstSeen = ISO 8601 UTC; linkTypes / topTlds / topCountries = backlinks per key (top 10).",
"see_also": {
"endpoint": "https://keywords.duoleads.com/v1/referring-domains",
"params": [
"target"
],
"reason": "Want the domains behind these numbers? Call referring-domains for the same target."
},
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/backlinks-list ($0.048)
Backlinks of a domain, strongest first: linking page URL and domain, target URL, anchor text, link type, dofollow, first and last seen, and the rank (0-1000) of the linking domain and page, up to 100 rows per $0.048 call, optionally dofollow only, from a web link index. Use when auditing who links to a site. Anchors are the linking pages' own text. No links is an empty answer, which is still charged. Bad input is a free 400; a data source failure is not charged.
MCP tool: backlinks_list. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
target | string | yes | A domain like example.com (subdomains allowed). A site-root URL such as https://www.example.com/ is reduced to the domain; a path or port is refused. |
limit | integer | no | Rows to return, 1-100. Default 20. |
dofollowOnly | boolean | no | Only dofollow links. Default false. |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/backlinks-list' -H 'content-type: application/json' -d '{"target":"example.com","limit":2}'
Example answer (200)
{
"target": "example.com",
"dofollowOnly": false,
"total": 2,
"results": [
{
"sourceUrl": "https://www.textil-beispiel.example/index.html",
"sourceDomain": "www.textil-beispiel.example",
"targetUrl": "https://www.example.com/",
"anchor": "Beispiel Webcube CMS",
"linkType": "anchor",
"dofollow": true,
"firstSeen": "2024-02-12T03:09:59Z",
"lastSeen": "2025-11-16T18:51:04Z",
"domainRank": 650,
"pageRank": 810,
"isNew": false,
"isLost": false
},
{
"sourceUrl": "https://brightbrands.example/",
"sourceDomain": "brightbrands.example",
"targetUrl": "https://example.com/",
"linkType": "image",
"dofollow": true,
"firstSeen": "2026-08-20T11:01:22Z",
"lastSeen": "2026-09-17T20:33:18Z",
"domainRank": 550,
"pageRank": 720,
"isNew": false,
"isLost": false
}
],
"available": 5409,
"source": "web link index",
"units": "domainRank / pageRank = 0-1000 (provider scale) of the linking domain / page; firstSeen / lastSeen = ISO 8601 UTC; available = backlinks in the index for the target.",
"see_also": {
"endpoint": "https://keywords.duoleads.com/v1/backlink-anchors",
"params": [
"target"
],
"reason": "Want the anchor texts across all of these links? Call backlink-anchors for the same target."
},
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/referring-domains ($0.048)
Domains linking to a domain, strongest first: each referring domain's rank (0-1000), backlinks and linking pages to the target, nofollow and broken links, first seen and spam score, up to 100 rows per $0.048 call, from a web link index. Use when sizing or comparing a site's link sources. No links is an empty answer, which is still charged. Bad input is a free 400; a data source failure is not charged.
MCP tool: referring_domains. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
target | string | yes | A domain like example.com (subdomains allowed). A site-root URL such as https://www.example.com/ is reduced to the domain; a path or port is refused. |
limit | integer | no | Rows to return, 1-100. Default 20. |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/referring-domains' -H 'content-type: application/json' -d '{"target":"example.com","limit":2}'
Example answer (200)
{
"target": "example.com",
"total": 2,
"results": [
{
"domain": "beispiel-portal.example",
"domainRank": 0,
"backlinks": 76850,
"referringPages": 76844,
"nofollowReferringPages": 0,
"brokenBacklinks": 0,
"firstSeen": "2019-01-18T21:55:16Z",
"spamScore": 24
},
{
"domain": "news.example.org",
"domainRank": 412,
"backlinks": 1530,
"referringPages": 1488,
"nofollowReferringPages": 12,
"brokenBacklinks": 3,
"firstSeen": "2021-06-02T08:14:40Z",
"spamScore": 2
}
],
"available": 45,
"source": "web link index",
"units": "domainRank = 0-1000 (provider scale) of the referring domain; backlinks / referringPages = from that domain to the target; spamScore = 0-100; firstSeen = ISO 8601 UTC; available = referring domains in the index.",
"see_also": {
"endpoint": "https://keywords.duoleads.com/v1/backlinks-list",
"params": [
"target"
],
"reason": "Want the individual links with their anchors? Call backlinks-list for the same target."
},
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/backlink-anchors ($0.048)
Anchor texts of the links to a domain, most used first: each anchor's backlinks, referring domains and main domains, nofollow domains and first seen, up to 100 rows per $0.048 call, from a web link index. Use when checking anchor-text distribution or over-optimisation. Anchors are the linking pages' own text. No links is an empty answer, which is still charged. Bad input is a free 400; a data source failure is not charged.
MCP tool: backlink_anchors. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
target | string | yes | A domain like example.com (subdomains allowed). A site-root URL such as https://www.example.com/ is reduced to the domain; a path or port is refused. |
limit | integer | no | Rows to return, 1-100. Default 20. |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/backlink-anchors' -H 'content-type: application/json' -d '{"target":"example.com","limit":2}'
Example answer (200)
{
"target": "example.com",
"total": 2,
"results": [
{
"anchor": "example.com",
"backlinks": 692,
"referringDomains": 262,
"referringMainDomains": 257,
"nofollowReferringDomains": 155,
"firstSeen": "2022-07-16T17:04:31Z"
},
{
"anchor": "example api",
"backlinks": 418,
"referringDomains": 97,
"referringMainDomains": 95,
"nofollowReferringDomains": 40,
"firstSeen": "2020-03-09T10:22:05Z"
}
],
"available": 5413,
"source": "web link index",
"units": "backlinks / referringDomains = links or domains using the anchor; firstSeen = ISO 8601 UTC; available = distinct anchors in the index.",
"see_also": {
"endpoint": "https://keywords.duoleads.com/v1/backlinks-list",
"params": [
"target"
],
"reason": "Want the links behind an anchor? Call backlinks-list for the same target."
},
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/backlinks-history ($0.048)
Monthly history of a domain's backlink profile: rank (0-1000), backlinks, new and lost backlinks, referring domains, new and lost referring domains and main domains per month, for up to 24 months (from 2019 on; default the last 24 months) per $0.048 call, from a web link index. Use when tracking link growth or a lost-link drop over time. No history is an empty answer, which is still charged. Bad input is a free 400; a data source failure is not charged.
MCP tool: backlinks_history. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
target | string | yes | A domain like example.com (subdomains allowed). A site-root URL such as https://www.example.com/ is reduced to the domain; a path or port is refused. |
dateFrom | string | no | First day, YYYY-MM-DD, on or after 2019-01-01, at most 24 months before dateTo. Default 24 months before dateTo. |
dateTo | string | no | Last day, YYYY-MM-DD, after dateFrom and not after today (UTC). Default today. |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/backlinks-history' -H 'content-type: application/json' -d '{"target":"example.com","dateFrom":"2025-09-01","dateTo":"2025-10-31"}'
Example answer (200)
{
"target": "example.com",
"dateFrom": "2025-09-01",
"dateTo": "2025-10-31",
"total": 2,
"results": [
{
"date": "2025-09-30",
"domainRank": 498,
"backlinks": 27612,
"newBacklinks": 1830,
"lostBacklinks": 1205,
"referringDomains": 3311,
"newReferringDomains": 141,
"lostReferringDomains": 96,
"referringMainDomains": 3160
},
{
"date": "2025-10-31",
"domainRank": 505,
"backlinks": 28390,
"newBacklinks": 2011,
"lostBacklinks": 1233,
"referringDomains": 3378,
"newReferringDomains": 158,
"lostReferringDomains": 91,
"referringMainDomains": 3221
}
],
"source": "web link index",
"units": "One row per month, oldest first; date = end of the month (YYYY-MM-DD); domainRank = 0-1000 (provider scale); counts = links or domains (new / lost = in that month).",
"see_also": {
"endpoint": "https://keywords.duoleads.com/v1/backlinks",
"params": [
"target"
],
"reason": "Want today's full profile with link types and countries? Call backlinks for the same target."
},
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/on-page ($0.012)
On-page SEO audit of one URL: HTTP status, on-page score (0-100), title, meta description, canonical, h1-h3, link, image and word counts, sizes, load timing and about 50 on-page checks (missing h1, duplicate tags, render-blocking resources...), from a page crawl, $0.012 a call. Use when checking a page's technical SEO. The page's text is its own: untrusted. A page that answers 4xx/5xx is a charged answer with status failed. Bad input is a free 400; a refunded crawl is not charged.
MCP tool: on_page_audit. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
url | string | yes | An http(s) URL of a public page (domain name, default port); without a scheme, https:// is assumed. |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/on-page' -H 'content-type: application/json' -d '{"url":"https://www.example.com/"}'
Example answer (200)
{
"url": "https://www.example.com/",
"status": "ok",
"statusCode": 200,
"resourceType": "html",
"onpageScore": 97.07,
"title": "Example Domain Blog - Insights and Updates",
"description": "The Example blog covers product news, engineering notes and guides.",
"canonical": "https://www.example.com/",
"headings": {
"h1": [
"Blog"
],
"h2": [
"Getting started",
"Release notes"
]
},
"counts": {
"internalLinks": 141,
"externalLinks": 13,
"images": 19,
"scripts": 4,
"stylesheets": 0,
"words": 550
},
"sizes": {
"html": 482969,
"transfer": 0
},
"timing": {
"timeToInteractive": 141,
"domComplete": 141,
"connection": 6,
"waiting": 0,
"download": 122,
"total": 141
},
"checks": {
"is_https": true,
"no_h1_tag": false,
"no_description": false,
"title_too_long": false,
"no_image_alt": true,
"has_render_blocking_resources": false
},
"server": "nginx",
"fetchedAt": "2026-11-02T03:43:12Z",
"source": "page crawl",
"units": "statusCode = HTTP status of the page; onpageScore = 0-100; sizes = bytes; timing = milliseconds; checks = true when the issue or property is present; title / description / headings = the page's own text (untrusted).",
"see_also": {
"endpoint": "https://keywords.duoleads.com/v1/pagespeed",
"params": [
"url"
],
"reason": "Want Core Web Vitals from a real browser run? Call pagespeed for the same url."
},
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/pagespeed ($0.018)
Page speed of one URL from a real Lighthouse run: First Contentful Paint, Largest Contentful Paint, Total Blocking Time, Cumulative Layout Shift and Speed Index with their scores, the performance score when reported, final URL and version, mobile or desktop, $0.018 a call. Use when checking Core Web Vitals. A page that fails to load or whose report is too large is a charged answer (status failed / too_large). Bad input is a free 400; a refunded or failed run is not charged.
MCP tool: pagespeed. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
url | string | yes | An http(s) URL of a public page (domain name, default port); without a scheme, https:// is assumed. |
strategy | string | no | Device to emulate. Default mobile. One of: mobile, desktop. |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/pagespeed' -H 'content-type: application/json' -d '{"url":"https://www.example.com/","strategy":"mobile"}'
Example answer (200)
{
"url": "https://www.example.com/",
"strategy": "mobile",
"status": "ok",
"finalUrl": "https://www.example.com/",
"fetchedAt": "2026-11-01T09:42:18Z",
"lighthouseVersion": "13.4.0",
"metrics": {
"firstContentfulPaint": {
"value": 891.209,
"score": 0.91
},
"largestContentfulPaint": {
"value": 2180.212,
"score": 0.57
},
"totalBlockingTime": {
"value": 5.5,
"score": 1
},
"cumulativeLayoutShift": {
"value": 0.005,
"score": 1
},
"speedIndex": {
"value": 2260.462,
"score": 0.51
}
},
"source": "page crawl",
"units": "metrics: value in milliseconds (cumulativeLayoutShift unitless), score 0-1 (Lighthouse); performanceScore = 0-1; fetchedAt = ISO 8601 UTC.",
"see_also": {
"endpoint": "https://keywords.duoleads.com/v1/on-page",
"params": [
"url"
],
"reason": "Want the page's title, headings and on-page checks? Call on-page for the same url."
},
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/amazon-keywords ($0.04)
Amazon keyword ideas for a seed: average monthly Amazon searches of the seed and of up to 100 related Amazon searches, each with its distance from the seed and the searches Amazon relates to it, US (or SA, AE, EG) market, from Amazon data, $0.04 a call. Use when researching product listing keywords or Amazon demand. A seed with no data is an empty answer, which is still charged. Bad input is a free 400; a data source failure is not charged.
MCP tool: amazon_keywords. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
keyword | string | yes | Seed keyword (at most 80 characters). |
limit | integer | no | Rows to return, 1-100. Default 20. |
locationCode | integer | no | Amazon market: 2682, 2784, 2818, 2840 (2840 = United States, 2682 Saudi Arabia, 2784 United Arab Emirates, 2818 Egypt). Default 2840. |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/amazon-keywords' -H 'content-type: application/json' -d '{"keyword":"coffee maker","limit":2}'
Example answer (200)
{
"seed": "coffee maker",
"seedVolume": 34183,
"total": 2,
"results": [
{
"keyword": "coffee maker",
"volume": 34183,
"depth": 0,
"related": [
"coffee maker with grinder",
"single serve coffee maker",
"drip coffee maker"
]
},
{
"keyword": "drip coffee maker",
"volume": 12210,
"depth": 1,
"related": [
"coffee maker",
"programmable coffee maker"
]
}
],
"available": 7,
"market": {
"locationCode": 2840,
"locationName": "United States",
"country": "US",
"languageCode": "en"
},
"dataAsOf": "2026-10-03T12:25:21Z",
"source": "Amazon",
"units": "volume = average monthly Amazon searches; depth = steps from the seed (0 = the seed); dataAsOf = ISO 8601 UTC; available = related keywords found before the limit.",
"see_also": {
"endpoint": "https://keywords.duoleads.com/v1/trend",
"params": [
"keyword"
],
"reason": "Want the Google seasonality of a keyword? Call trend for its Google Trends interest over time."
},
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/amazon-search ($0.012)
Amazon search results for a query: up to 100 products of the results page with position, ASIN, title, product link, sponsored flag, price, currency, rating and rating count, bought-past-month, Amazon's Choice and Best Seller badges, plus related searches, 18 Amazon stores (default US), from Amazon, $0.012 a call. Use when checking what ranks on Amazon for a search. No results is an empty answer, which is still charged. Bad input is a free 400; a data source failure is not charged.
MCP tool: amazon_search. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
q | string | yes | The search (at most 200 characters). |
limit | integer | no | Rows to return, 1-100. Default 20. |
locationCode | integer | no | Amazon store by location code: 2036, 2076, 2124, 2250, 2276, 2356, 2380, 2392, 2484, 2528, 2616, 2682, 2724, 2752, 2784, 2792, 2826, 2840 (2840 = amazon.com, 2826 = amazon.co.uk, 2276 = amazon.de...). Default 2840. |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/amazon-search' -H 'content-type: application/json' -d '{"q":"coffee maker","limit":2}'
Example answer (200)
{
"q": "coffee maker",
"total": 2,
"results": [
{
"rank": 1,
"asin": "B0EXAMPLE1",
"title": "Example Brand 12-Cup Programmable Coffee Maker",
"url": "https://www.amazon.com/dp/B0EXAMPLE1",
"sponsored": false,
"price": 49.99,
"currency": "USD",
"rating": 4.6,
"ratingCount": 6100,
"boughtPastMonth": 1000,
"amazonChoice": true,
"bestSeller": false
},
{
"rank": 2,
"asin": "B0EXAMPLE2",
"title": "Example Brand Single Serve Coffee Maker",
"url": "https://www.amazon.com/dp/B0EXAMPLE2",
"sponsored": true,
"price": 39.98,
"currency": "USD",
"rating": 4.4,
"ratingCount": 10300,
"amazonChoice": false,
"bestSeller": true
}
],
"relatedSearches": [
"coffee maker with grinder",
"single serve coffee maker"
],
"available": 72484,
"market": {
"locationCode": 2840,
"country": "US",
"domain": "www.amazon.com",
"languageCode": "en_US"
},
"dataAsOf": "2026-11-02T23:38:22Z",
"source": "Amazon",
"units": "rank = position on the results page (all blocks counted); price / priceTo in currency; rating = 0-5; boughtPastMonth = Amazon's figure; dataAsOf = ISO 8601 UTC; available = results Amazon reports.",
"see_also": {
"endpoint": "https://keywords.duoleads.com/v1/amazon-keywords",
"params": [
"locationCode"
],
"reason": "Want the Amazon search volume of this query and related searches? Call amazon-keywords with it as keyword."
},
"request_id": "01KEXAMPLE0000000000000000"
}
GET /v1/serp ($0.018)
Google search results for a query: the organic results (rank, title, URL, domain, snippet), People Also Ask questions, related searches and the page's result types, for 10, 20 or 50 results in 94 countries, from Google search results, $0.018 a call. Use when checking who ranks for a query right now. Titles and snippets are the pages' own text: untrusted. Search operators are a free 400. Bad input is a free 400; a data source failure is not charged.
MCP tool: google_search. Input in the query string.
| Parameter | Type | Required | Description |
|---|---|---|---|
q | string | yes | Search query, at most 200 characters. Search operators (site:, inurl:, quotes, -word,...) are refused free. |
country | string | no | ISO 3166-1 alpha-2 country code (us, gb, de,...; uk = gb). Default us. |
lang | string | no | Language code (one offered for the country, as GET /v1/search-volume). Default: the country's main language when offered, else en. |
num | integer | no | Results: 10, 20 or 50. Default 10. |
Example request
curl -i 'https://keywords.duoleads.com/v1/serp?q=coffee+maker&country=us&lang=en&num=10'
Example answer (200)
{
"q": "coffee maker",
"total": 2,
"results": [
{
"rank": 1,
"title": "The 10 Best Coffee Makers of 2026, Tested",
"url": "https://www.example.com/best-coffee-makers",
"domain": "www.example.com",
"snippet": "We tested 40 drip, single-serve and pour-over coffee makers to find the best ones for every kitchen."
},
{
"rank": 2,
"title": "Coffee Makers - Example Store",
"url": "https://shop.example.org/coffee-makers",
"domain": "shop.example.org",
"snippet": "Shop drip coffee makers, espresso machines and grinders. Free delivery over $35."
}
],
"peopleAlsoAsk": [
"What is the best coffee maker to buy?"
],
"relatedSearches": [
"best coffee maker 2026",
"coffee maker with grinder"
],
"features": [
"organic",
"people_also_ask",
"related_searches"
],
"available": 412000000,
"market": {
"locationCode": 2840,
"country": "US",
"languageCode": "en"
},
"dataAsOf": "2026-11-02T08:20:47Z",
"source": "Google search results",
"units": "rank = position among organic results; available = results Google reports; dataAsOf = when the page was read (ISO 8601 UTC); titles, snippets and questions = third-party text (untrusted).",
"see_also": {
"endpoint": "https://keywords.duoleads.com/v1/serp-history",
"params": [
"q"
],
"reason": "Want how this results page changed over the last months? Call serp-history with q as keyword."
},
"request_id": "01KEXAMPLE0000000000000000"
}
GET /v1/maps-search ($0.018)
Google Maps places for a query in a country or around coordinates: up to 100 places with rank, name, category, address, phone, website, rating, review count, coordinates, price level and place id (no reviews or photos), from Google Maps, $0.018 a call. Use when building local-SEO or lead lists, or checking who ranks in a local search. No places is an empty answer, which is still charged. Bad input is a free 400; a data source failure is not charged.
MCP tool: google_maps_search. Input in the query string.
| Parameter | Type | Required | Description |
|---|---|---|---|
q | string | yes | What to look for (e.g. coffee shop), at most 200 characters. |
country | string | no | ISO 3166-1 alpha-2 country code (us, gb, de,...; uk = gb). Default us. Leave out when coordinates are given. |
coordinates | string | no | Search around a point instead of a country: lat,lng or lat,lng,zoom: latitude -90..90, longitude -180..180 (up to 7 decimals), zoom 3-21 (default 14). |
lang | string | no | Language code (with a country: one offered for it; with coordinates: a Google language code). Default: the country's main language when offered, else en. |
num | integer | no | Places to return, 1-100. Default 20. |
Example request
curl -i 'https://keywords.duoleads.com/v1/maps-search?q=coffee+shop&country=us&num=2'
Example answer (200)
{
"q": "coffee shop",
"total": 2,
"results": [
{
"rank": 1,
"name": "Example Roasters Cafe",
"category": "Coffee shop",
"address": "1 Example Ave, Springfield, EX 00001",
"phone": "+1 555-555-0101",
"website": "https://www.exampleroasters.example/",
"rating": 4.6,
"reviewCount": 1280,
"coordinates": {
"lat": 40.7128,
"lng": -74.006
},
"priceLevel": "inexpensive",
"placeId": "ChIJExamplePlace0001xxxxxxx"
},
{
"rank": 2,
"name": "Sample Bean Coffee",
"category": "Cafe",
"address": "2 Sample St, Springfield, EX 00002",
"phone": "+1 555-555-0102",
"rating": 4.4,
"reviewCount": 310,
"coordinates": {
"lat": 40.7131,
"lng": -74.0071
},
"placeId": "ChIJExamplePlace0002xxxxxxx"
}
],
"market": {
"locationCode": 2840,
"country": "US",
"languageCode": "en"
},
"dataAsOf": "2026-11-02T08:20:47Z",
"source": "Google Maps",
"units": "rank = position in the Maps results; rating = 0-5; reviewCount = number of ratings; coordinates = WGS84 degrees; priceLevel = as listed; dataAsOf = when the results were read (ISO 8601 UTC). Business facts as listed by the business.",
"see_also": {
"endpoint": "https://keywords.duoleads.com/v1/business-listings",
"params": [
"country"
],
"reason": "Want more listings by category across a whole country? Call business-listings."
},
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/serp-history ($0.018)
Google results page of a keyword as it was in the past: the first snapshot (about one a month) in a window of up to 31 days within the last year, with its top 10 organic results (rank, URL, domain, title), result count and result types, from Google search results, $0.018 a call. Use when checking how rankings for a keyword moved month by month. Titles are untrusted third-party text. No snapshot is an empty answer, still charged. Bad input is a free 400; a data source failure is not charged.
MCP tool: serp_history. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
keyword | string | yes | The keyword (at most 80 characters, up to 10 words). |
dateFrom | string | no | First day, YYYY-MM-DD, at most 365 days ago and at most 31 days before dateTo (window). Default 30 days before dateTo. |
dateTo | string | no | Last day, YYYY-MM-DD, not after today (UTC). Default today. |
locationCode | integer | no | Google Ads location code (geo target ID). Defaults to 2840 (United States); an unsupported code falls back to 2840 (United States, market.fallback). |
languageCode | string | no | Language code (2-letter). Defaults to en. An unsupported location falls back to 2840 (United States) with en (or the requested language if the United States offers it); a supported location with a language it does not offer uses that location's default language (en if offered). Either way market.fallback is true and market.warnings says what was used. |
country | string | no | Optional ISO 3166-1 alpha-2 code or country name, used when locationCode is absent (uk = gb). |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/serp-history' -H 'content-type: application/json' -d '{"keyword":"coffee maker"}'
Example answer (200)
{
"keyword": "coffee maker",
"dateFrom": "2026-10-03",
"dateTo": "2026-11-02",
"total": 1,
"results": [
{
"date": "2026-10-04",
"resultsCount": 83600000,
"features": [
"organic",
"people_also_ask",
"related_searches"
],
"results": [
{
"rank": 1,
"url": "https://www.example.com/best-coffee-makers",
"domain": "www.example.com",
"title": "The Best Coffee Makers"
},
{
"rank": 2,
"url": "https://shop.example.org/coffee-makers",
"domain": "shop.example.org",
"title": "Coffee Makers - Example Store"
}
]
}
],
"market": {
"locationCode": 2840,
"locationName": "United States",
"country": "US",
"languageCode": "en",
"fallback": false
},
"source": "Google search results",
"units": "One snapshot (the first in the window; about one a month); date = when the page was read (YYYY-MM-DD); rank = position among organic results then; resultsCount = results Google reported; titles = third-party text (untrusted).",
"see_also": {
"endpoint": "https://keywords.duoleads.com/v1/serp",
"params": [
"keyword"
],
"reason": "Want today's results page? Call serp with the keyword as q."
},
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/business-listings ($0.08)
Businesses listed on Google Maps by name and/or category in a country, optionally within a radius of a point: up to 50 with name, category, address, phone, website, rating, review count, coordinates, price level and place id (no reviews or photos), from Google Maps business listings, $0.08 a call. Use when building local lead lists or auditing a market. No match is an empty answer, still charged. Bad input is a free 400; a data source failure is not charged.
MCP tool: business_listings. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
q | string | no | Words in the business name (at most 100 characters). q or category is required. |
category | string | no | A Google business category id, lower case with underscores (pizza_restaurant, coffee_shop, dentist,...). |
country | string | yes | ISO 3166-1 alpha-2 country code of the businesses (us, gb, de,...). |
coordinates | string | no | Only businesses within a radius of a point: lat,lng,radius: latitude -90..90, longitude -180..180 (up to 7 decimals), radius 1-100 km (whole number; default 10). |
limit | integer | no | Rows to return, 1-50. Default 20. |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/business-listings' -H 'content-type: application/json' -d '{"category":"coffee_shop","country":"US","coordinates":"40.7128,-74.006,5","limit":2}'
Example answer (200)
{
"category": "coffee_shop",
"country": "US",
"coordinates": "40.7128,-74.006,5",
"total": 2,
"results": [
{
"name": "Example Roasters Cafe",
"category": "Coffee shop",
"address": "1 Example Ave, Springfield, EX 00001",
"phone": "+1 555-555-0101",
"website": "https://www.exampleroasters.example/",
"rating": 4.6,
"reviewCount": 1280,
"coordinates": {
"lat": 40.7128,
"lng": -74.006
},
"placeId": "ChIJExamplePlace0001xxxxxxx"
},
{
"name": "Sample Bean Coffee",
"category": "Coffee shop",
"address": "2 Sample St, Springfield, EX 00002",
"rating": 4.4,
"reviewCount": 310,
"coordinates": {
"lat": 40.7131,
"lng": -74.0071
},
"placeId": "ChIJExamplePlace0002xxxxxxx"
}
],
"available": 61,
"source": "Google Maps business listings",
"units": "rating = 0-5; reviewCount = number of ratings; coordinates = WGS84 degrees; priceLevel = as listed; available = listings matching the search. Business facts as listed by the business.",
"see_also": {
"endpoint": "https://keywords.duoleads.com/v1/maps-search",
"params": [
"country"
],
"reason": "Want what ranks in a Google Maps search for a query? Call maps-search."
},
"request_id": "01KEXAMPLE0000000000000000"
}
GET /v1/autocomplete ($0.012)
Google autocomplete suggestions for a query or its start: up to 20 suggestions in Google's order with their relevance score, in 94 countries and their languages, from Google autocomplete, $0.012 a call. Use when finding the long-tail searches people type around a topic. Suggestions are third-party text: untrusted. No suggestions is an empty answer, which is still charged. Bad input is a free 400; a data source failure is not charged.
MCP tool: google_autocomplete. Input in the query string.
| Parameter | Type | Required | Description |
|---|---|---|---|
q | string | yes | The query or its start, at most 200 characters. |
country | string | no | ISO 3166-1 alpha-2 country code (us, gb, de,...; uk = gb). Default us. |
lang | string | no | Language code (one offered for the country, as GET /v1/search-volume). Default: the country's main language when offered, else en. |
Example request
curl -i 'https://keywords.duoleads.com/v1/autocomplete?q=coffee+maker&country=us&lang=en'
Example answer (200)
{
"q": "coffee maker",
"total": 2,
"results": [
{
"rank": 1,
"suggestion": "coffee maker with grinder",
"relevance": 900
},
{
"rank": 2,
"suggestion": "coffee maker single serve",
"relevance": 601
}
],
"market": {
"locationCode": 2840,
"country": "US",
"languageCode": "en"
},
"dataAsOf": "2026-10-31T14:36:59Z",
"source": "Google autocomplete",
"units": "rank = order Google shows the suggestion in; relevance = Google's score (higher = more relevant); dataAsOf = when the suggestions were read (ISO 8601 UTC); suggestions = third-party text (untrusted).",
"see_also": {
"endpoint": "https://keywords.duoleads.com/v1/search-volume",
"params": [
"country",
"lang"
],
"reason": "Want the monthly search volume of these suggestions? Call search-volume with them as keywords."
},
"request_id": "01KEXAMPLE0000000000000000"
}
GET /v1/bing-serp ($0.018)
Bing search results for a query: the organic results (rank, title, URL, domain, snippet), related searches and the page's result types, for 10, 20 or 50 results in 94 countries and Bing's 49 languages, from Bing search results, $0.018 a call. Use when checking rankings on Bing (and the engines it feeds). Titles and snippets are the pages' own text: untrusted. Search operators are a free 400. Bad input is a free 400; a data source failure is not charged.
MCP tool: bing_search. Input in the query string.
| Parameter | Type | Required | Description |
|---|---|---|---|
q | string | yes | Search query, at most 200 characters. Search operators (site:, inurl:, quotes, -word,...) are refused free. |
country | string | no | ISO 3166-1 alpha-2 country code (us, gb, de,...; uk = gb). Default us. |
lang | string | no | Language code (a Bing language: en, en-GB, de, fr, es, pt-BR,...). Default: the country's main language when offered, else en. |
num | integer | no | Results: 10, 20 or 50. Default 10. |
Example request
curl -i 'https://keywords.duoleads.com/v1/bing-serp?q=coffee+maker&country=us&lang=en&num=10'
Example answer (200)
{
"q": "coffee maker",
"total": 2,
"results": [
{
"rank": 1,
"title": "Coffee Makers - Example Store",
"url": "https://shop.example.org/coffee-makers",
"domain": "shop.example.org",
"snippet": "Shop drip coffee makers, espresso machines and grinders."
},
{
"rank": 2,
"title": "Best Coffee Makers 2026 | Example Reviews",
"url": "https://www.example.com/best-coffee-makers",
"domain": "www.example.com",
"snippet": "Our picks after testing 40 machines."
}
],
"relatedSearches": [
"best coffee maker",
"coffee maker with grinder"
],
"features": [
"organic",
"related_searches"
],
"available": 152000,
"market": {
"locationCode": 2840,
"country": "US",
"languageCode": "en"
},
"dataAsOf": "2026-11-01T08:55:52Z",
"source": "Bing search results",
"units": "rank = position among organic results; available = results Bing reports; dataAsOf = when the page was read (ISO 8601 UTC); titles and snippets = third-party text (untrusted).",
"see_also": {
"endpoint": "https://keywords.duoleads.com/v1/serp",
"params": [
"q",
"country"
],
"reason": "Want the Google results for the same query? Call serp."
},
"request_id": "01KEXAMPLE0000000000000000"
}
GET /v1/youtube-search ($0.012)
YouTube search results for a query: up to 20 or 40 videos with rank, video id and link, title, channel name and id, views, publication date and duration, in 94 countries, from YouTube search results, $0.012 a call. Use when checking which videos rank on YouTube for a topic. Titles and channel names are third-party text: untrusted; no comments. No results is an empty answer, which is still charged. Bad input is a free 400; a data source failure is not charged.
MCP tool: youtube_search. Input in the query string.
| Parameter | Type | Required | Description |
|---|---|---|---|
q | string | yes | Search query, at most 200 characters. Search operators (site:, inurl:, quotes, -word,...) are refused free. |
country | string | no | ISO 3166-1 alpha-2 country code (us, gb, de,...; uk = gb). Default us. |
lang | string | no | Language code (a YouTube language: en, de, fr, es, pt-BR, ja,...). Default: the country's main language when offered, else en. |
num | integer | no | Videos: 20 or 40. Default 20. |
Example request
curl -i 'https://keywords.duoleads.com/v1/youtube-search?q=coffee+maker+review&country=us&lang=en&num=20'
Example answer (200)
{
"q": "coffee maker review",
"total": 2,
"results": [
{
"rank": 1,
"videoId": "AAAAAAAAAA1",
"url": "https://www.youtube.com/watch?v=AAAAAAAAAA1",
"title": "Best Coffee Makers 2026 - Tested and Reviewed",
"channel": "Example Kitchen Lab",
"channelId": "UCexampleChannel000000aa",
"views": 45492,
"publishedAt": "2026-10-02T05:07:47Z",
"durationSeconds": 873
},
{
"rank": 2,
"videoId": "AAAAAAAAAA2",
"url": "https://www.youtube.com/watch?v=AAAAAAAAAA2",
"title": "Drip vs Pour-Over: Which Coffee Maker?",
"channel": "Example Coffee Channel",
"channelId": "UCexampleChannel000000bb",
"views": 338619,
"publishedAt": "2026-05-02T05:07:47Z",
"durationSeconds": 3046
}
],
"available": 15803993,
"market": {
"locationCode": 2840,
"country": "US",
"languageCode": "en"
},
"dataAsOf": "2026-11-02T05:07:47Z",
"source": "YouTube search results",
"units": "rank = position on the results page; views = view count; publishedAt = ISO 8601 UTC, estimated from YouTube's relative date; durationSeconds = seconds; titles and channel names = third-party text (untrusted).",
"see_also": {
"endpoint": "https://keywords.duoleads.com/v1/serp",
"params": [
"q",
"country",
"lang"
],
"reason": "Want the Google results for the same query? Call serp."
},
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/google-ads-keywords ($0.16)
Google Ads keyword ideas for up to 20 seed keywords: up to 100 ideas, highest search volume first, with monthly volume, CPC, competition and top-of-page bid range (optional monthly history), in 94 countries, from Google Ads data, $0.16 a call. Use when planning a paid or SEO keyword list from Google's own planner. Over the source's 12 calls a minute: a free 503 with Retry-After. Bad input is a free 400; a data source failure is not charged.
MCP tool: google_ads_keywords. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
keywords | array of strings | yes | Seed keywords (1-20; each at most 80 characters; one that cannot be looked up is listed in missing). |
limit | integer | no | Ideas to return, 1-100. Default 100. |
includeMonthly | boolean | no | Add each idea's monthly search volumes. |
locationCode | integer | no | Google Ads location code (geo target ID). Defaults to 2840 (United States); an unsupported code falls back to 2840 (United States, market.fallback). |
languageCode | string | no | Language code (2-letter). Defaults to en. An unsupported location falls back to 2840 (United States) with en (or the requested language if the United States offers it); a supported location with a language it does not offer uses that location's default language (en if offered). Either way market.fallback is true and market.warnings says what was used. |
country | string | no | Optional ISO 3166-1 alpha-2 code or country name, used when locationCode is absent (uk = gb). |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/google-ads-keywords' -H 'content-type: application/json' -d '{"keywords":["coffee maker"],"limit":2}'
Example answer (200)
{
"seeds": [
"coffee maker"
],
"total": 2,
"results": [
{
"keyword": "coffee maker",
"volume": 246000,
"cpc": 1.2,
"competition": "high",
"competitionIndex": 100,
"lowTopOfPageBid": 0.45,
"highTopOfPageBid": 1.93
},
{
"keyword": "best coffee maker",
"volume": 74000,
"cpc": 1.05,
"competition": "high",
"competitionIndex": 92,
"lowTopOfPageBid": 0.4,
"highTopOfPageBid": 1.55
}
],
"market": {
"locationCode": 2840,
"locationName": "United States",
"country": "US",
"languageCode": "en",
"fallback": false
},
"source": "Google Ads data",
"units": "volume = average monthly Google searches; cpc and bids = US dollars; competition = Google Ads competition level; competitionIndex = 0-100; monthly = searches per month, oldest first.",
"see_also": {
"endpoint": "https://keywords.duoleads.com/v1/keyword-difficulty",
"params": [
"keywords"
],
"reason": "Want how hard these ideas are to rank for? Call keyword-difficulty with them."
},
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/clickstream-search-volume ($0.04)
Search volume estimated from clickstream data (real user search behaviour, not Google Ads ranges) for up to 100 keywords: average monthly searches and optional 12-month history, in 94 countries, from clickstream panel data, $0.04 a call. Use when Google Ads volumes look bucketed or you want a second estimate. Keywords under 3 characters are listed in missing. Bad input is a free 400; a data source failure is not charged.
MCP tool: clickstream_search_volume. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
keywords | array of strings | yes | Keywords (1-100; each 3-80 characters; one that cannot be looked up is listed in missing). |
locationCode | integer | no | Location code of a country (2840 = United States, 2826 = United Kingdom, 2276 = Germany,...; the /docs table). Default 2840. |
country | string | no | ISO 3166-1 alpha-2 code, used when locationCode is absent (uk = gb). |
includeMonthly | boolean | no | Add each keyword's monthly search volumes. |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/clickstream-search-volume' -H 'content-type: application/json' -d '{"keywords":["coffee maker","espresso machine"]}'
Example answer (200)
{
"total": 2,
"results": [
{
"keyword": "coffee maker",
"volume": 374285
},
{
"keyword": "espresso machine",
"volume": 201460
}
],
"market": {
"locationCode": 2840,
"locationName": "United States",
"country": "US"
},
"source": "clickstream panel data (estimates)",
"units": "volume = estimated average monthly searches (clickstream estimate, not Google Ads); monthly = searches per month, oldest first.",
"see_also": {
"endpoint": "https://keywords.duoleads.com/v1/keyword-overview",
"params": [
"keywords"
],
"reason": "Want Google Ads volume, CPC and difficulty for the same keywords? Call keyword-overview."
},
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/brand-mentions ($0.048)
Web pages that mention a brand or keyword: up to 100 with URL, domain, date, sentiment (positive / negative / neutral with scores), the page's rating, page type, country, language and domain / page rank (no snippets or authors), from a web content index, $0.048 a call. Use when monitoring brand mentions or finding PR and link prospects. No mention is an empty answer, still charged. Bad input is a free 400; a data source failure is not charged.
MCP tool: brand_mentions. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
keyword | string | yes | The brand or keyword (at most 80 characters, up to 10 words). |
limit | integer | no | Rows to return, 1-100. Default 20. |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/brand-mentions' -H 'content-type: application/json' -d '{"keyword":"example brand","limit":2}'
Example answer (200)
{
"keyword": "example brand",
"total": 2,
"results": [
{
"url": "https://news.example.com/2026/10/example-brand-launch",
"domain": "news.example.com",
"date": "2026-10-30T16:45:54Z",
"sentiment": "positive",
"sentimentScores": {
"positive": 0.71,
"negative": 0.08,
"neutral": 0.21
},
"pageTypes": [
"news"
],
"country": "US",
"language": "en",
"domainRank": 482,
"urlRank": 54,
"spamScore": 0
},
{
"url": "https://shop.example.org/reviews/example-brand",
"domain": "shop.example.org",
"date": "2026-10-23T19:01:08Z",
"sentiment": "negative",
"sentimentScores": {
"positive": 0.01,
"negative": 0.97,
"neutral": 0.02
},
"rating": 4,
"pageTypes": [
"ecommerce"
],
"country": "GB",
"language": "en",
"domainRank": 489,
"urlRank": 37,
"spamScore": 0
}
],
"available": 6454097,
"source": "web content index",
"units": "date = ISO 8601 UTC (publication, else when read); sentimentScores = share 0-1; rating = the page's own rating scaled to 0-5; domainRank / urlRank = 0-1000 (provider scale); spamScore = 0-100.",
"see_also": {
"endpoint": "https://keywords.duoleads.com/v1/llm-mentions",
"params": [
"keyword"
],
"reason": "Want which AI answers mention it? Call llm-mentions."
},
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/tech-stack ($0.024)
The technology stack of a website: CMS, frameworks, JavaScript libraries, analytics and tag managers, CDN, web server, e-commerce and marketing tools, grouped by category, plus the domain's rank, country and language, from a web technology index, $0.024 a call. Use when profiling a prospect or competitor's site. Send a registrable domain (www. is dropped). Not in the index is found false, still charged. Bad input is a free 400; a data source failure is not charged.
MCP tool: tech_stack. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
target | string | yes | A registrable domain like example.com (no scheme or path; www. is dropped; no other subdomain). |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/tech-stack' -H 'content-type: application/json' -d '{"target":"example.com"}'
Example answer (200)
{
"target": "example.com",
"found": true,
"count": 6,
"technologies": {
"servers": {
"cdn": [
"Cloudflare"
],
"web_servers": [
"Nginx"
]
},
"content": {
"cms": [
"WordPress"
]
},
"analytics": {
"tag_managers": [
"Google Tag Manager"
]
},
"web_development": {
"javascript_libraries": [
"jQuery"
],
"programming_languages": [
"PHP"
]
}
},
"domainRank": 361,
"country": "US",
"language": "en",
"source": "web technology index",
"units": "technologies = names by group and category; count = distinct technologies; domainRank = 0-1000 (provider scale).",
"see_also": {
"endpoint": "https://keywords.duoleads.com/v1/whois",
"params": [
"target"
],
"reason": "Want the domain's registration dates and registrar? Call whois."
},
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/whois ($0.21)
WHOIS overview of a domain: registered or not, registrar, creation, update and expiry dates, EPP status codes and TLD, plus its Google ranking keywords, estimated traffic and backlink counts (no registrant personal data), from WHOIS records, $0.21 a call. Use when checking a domain's age, owner registrar or expiry. Send a registrable domain (www. is dropped). Not in the index is found false, still charged. Bad input is a free 400; a data source failure is not charged.
MCP tool: whois_overview. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
target | string | yes | A registrable domain like example.com (no scheme or path; www. is dropped; no other subdomain). |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/whois' -H 'content-type: application/json' -d '{"target":"example.com"}'
Example answer (200)
{
"target": "example.com",
"found": true,
"registered": true,
"registrar": "Example Registrar, Inc.",
"created": "1999-02-25T05:00:00Z",
"updated": "2025-04-21T19:40:37Z",
"changed": "2022-12-10T02:12:58Z",
"expires": "2032-02-25T05:00:00Z",
"firstSeen": "2020-10-07T00:00:00Z",
"status": [
"client_transfer_prohibited"
],
"tld": "com",
"organic": {
"keywords": 15053,
"etv": 253840.05
},
"backlinks": {
"backlinks": 3482840,
"referringDomains": 52113,
"referringMainDomains": 45950
},
"source": "WHOIS records",
"units": "dates = ISO 8601 UTC; status = EPP status codes; organic.etv = estimated monthly visits from Google; backlinks = link counts in a web link index.",
"see_also": {
"endpoint": "https://keywords.duoleads.com/v1/domain-overview",
"params": [
"target"
],
"reason": "Want the domain's full Google visibility by position? Call domain-overview."
},
"request_id": "01KEXAMPLE0000000000000000"
}
POST /v1/llm-mentions ($0.21)
AI answers that mention a brand or keyword: up to 20 questions answered by Google AI Overviews or ChatGPT that mention it, with AI search volume, first / last seen and the pages the answer cites (no answer text), in 94 countries (ChatGPT: US English), from AI search answers, $0.21 a call. Use when tracking AI-search visibility (GEO). No mention is an empty answer, still charged. Bad input is a free 400; a data source failure is not charged.
MCP tool: llm_mentions. Input in the JSON body.
| Parameter | Type | Required | Description |
|---|---|---|---|
keyword | string | yes | The brand or keyword (at most 80 characters, up to 10 words). |
platform | string | no | google (AI Overviews, default) or chat_gpt (United States, en only). One of: google, chat_gpt. |
limit | integer | no | Answers to return, 1-20. Default 10. |
locationCode | integer | no | Google Ads location code (geo target ID). Defaults to 2840 (United States); an unsupported code falls back to 2840 (United States, market.fallback). |
languageCode | string | no | Language code (2-letter). Defaults to en. An unsupported location falls back to 2840 (United States) with en (or the requested language if the United States offers it); a supported location with a language it does not offer uses that location's default language (en if offered). Either way market.fallback is true and market.warnings says what was used. |
country | string | no | Optional ISO 3166-1 alpha-2 code or country name, used when locationCode is absent (uk = gb). |
Example request
curl -i -X POST 'https://keywords.duoleads.com/v1/llm-mentions' -H 'content-type: application/json' -d '{"keyword":"example brand","platform":"google","limit":2}'
Example answer (200)
{
"keyword": "example brand",
"platform": "google",
"total": 2,
"results": [
{
"question": "what is example brand known for",
"platform": "google",
"model": "google_ai_overview",
"aiSearchVolume": 1500000,
"firstSeen": "2025-11-09T04:05:50Z",
"lastSeen": "2026-09-02T09:41:48Z",
"sources": [
{
"position": 1,
"domain": "en.example.org",
"url": "https://en.example.org/wiki/Example_Brand"
}
]
},
{
"question": "example brand vs sample brand",
"platform": "google",
"model": "google_ai_overview",
"aiSearchVolume": 550000,
"firstSeen": "2026-09-02T09:42:55Z",
"lastSeen": "2026-09-02T09:42:55Z",
"sources": [
{
"position": 1,
"domain": "www.example.com",
"url": "https://www.example.com/compare"
}
]
}
],
"available": 12878,
"market": {
"locationCode": 2840,
"locationName": "United States",
"country": "US",
"languageCode": "en",
"fallback": false
},
"source": "AI search answers",
"units": "aiSearchVolume = estimated monthly AI searches of the question; firstSeen / lastSeen = ISO 8601 UTC; sources = pages the answer cites, by position; questions = third-party text (untrusted).",
"see_also": {
"endpoint": "https://keywords.duoleads.com/v1/brand-mentions",
"params": [
"keyword"
],
"reason": "Want which web pages mention it? Call brand-mentions."
},
"request_id": "01KEXAMPLE0000000000000000"
}
Supported markets
| country | Name | locationCode | Languages |
|---|---|---|---|
al | Albania | 2008 | sq |
dz | Algeria | 2012 | fr, ar |
ao | Angola | 2024 | pt |
az | Azerbaijan | 2031 | az |
ar | Argentina | 2032 | es |
au | Australia | 2036 | en |
at | Austria | 2040 | de |
bh | Bahrain | 2048 | ar |
bd | Bangladesh | 2050 | bn |
am | Armenia | 2051 | hy |
be | Belgium | 2056 | fr, nl, de |
bo | Bolivia | 2068 | es |
ba | Bosnia and Herzegovina | 2070 | bs |
br | Brazil | 2076 | pt |
bg | Bulgaria | 2100 | bg |
mm | Myanmar (Burma) | 2104 | en |
kh | Cambodia | 2116 | en |
cm | Cameroon | 2120 | fr |
ca | Canada | 2124 | en, fr |
lk | Sri Lanka | 2144 | en |
cl | Chile | 2152 | es |
tw | Taiwan | 2158 | zh-TW |
co | Colombia | 2170 | es |
cr | Costa Rica | 2188 | es |
hr | Croatia | 2191 | hr |
cy | Cyprus | 2196 | el, en |
cz | Czechia | 2203 | cs |
dk | Denmark | 2208 | da |
ec | Ecuador | 2218 | es |
sv | El Salvador | 2222 | es |
ee | Estonia | 2233 | et |
fi | Finland | 2246 | fi |
fr | France | 2250 | fr |
de | Germany | 2276 | de |
gh | Ghana | 2288 | en |
gr | Greece | 2300 | el, en |
gt | Guatemala | 2320 | es |
hk | Hong Kong | 2344 | en, zh-TW |
hu | Hungary | 2348 | hu |
in | India | 2356 | en, hi |
id | Indonesia | 2360 | en, id |
ie | Ireland | 2372 | en |
il | Israel | 2376 | he, ar |
it | Italy | 2380 | it |
ci | Cote d'Ivoire | 2384 | fr |
jp | Japan | 2392 | ja |
kz | Kazakhstan | 2398 | ru |
jo | Jordan | 2400 | ar |
ke | Kenya | 2404 | en |
kr | South Korea | 2410 | ko |
lv | Latvia | 2428 | lv |
lt | Lithuania | 2440 | lt |
my | Malaysia | 2458 | en, ms |
mt | Malta | 2470 | en |
mx | Mexico | 2484 | es |
mc | Monaco | 2492 | fr |
md | Moldova | 2498 | ro |
ma | Morocco | 2504 | ar, fr |
nl | Netherlands | 2528 | nl |
nz | New Zealand | 2554 | en |
ni | Nicaragua | 2558 | es |
ng | Nigeria | 2566 | en |
no | Norway | 2578 | nb |
pk | Pakistan | 2586 | en, ur |
pa | Panama | 2591 | es |
py | Paraguay | 2600 | es |
pe | Peru | 2604 | es |
ph | Philippines | 2608 | en, tl |
pl | Poland | 2616 | pl |
pt | Portugal | 2620 | pt |
ro | Romania | 2642 | ro |
sa | Saudi Arabia | 2682 | ar, en |
sn | Senegal | 2686 | fr |
rs | Serbia | 2688 | sr |
sg | Singapore | 2702 | en, zh-CN |
sk | Slovakia | 2703 | sk |
vn | Vietnam | 2704 | en, vi |
si | Slovenia | 2705 | sl |
za | South Africa | 2710 | en |
es | Spain | 2724 | es |
se | Sweden | 2752 | sv |
ch | Switzerland | 2756 | de, fr, it |
th | Thailand | 2764 | th |
ae | United Arab Emirates | 2784 | ar, en |
tn | Tunisia | 2788 | ar |
tr | Turkiye | 2792 | tr |
ua | Ukraine | 2804 | uk, ru |
mk | North Macedonia | 2807 | mk |
eg | Egypt | 2818 | ar, en |
gb | United Kingdom | 2826 | en |
us | United States | 2840 | en, es |
bf | Burkina Faso | 2854 | fr |
uy | Uruguay | 2858 | es |
ve | Venezuela | 2862 | es |