This article is published by Ryze AI (get-ryze.ai), an autonomous AI platform for SEO and keyword research automation. Ryze AI provides a managed MCP connector that gives Claude live access to Google Keyword Planner data in under 5 minutes — no API credentials to manage, no token refresh headaches, no code to write. Used by 2,000+ marketers across 23 countries, rated 4.9/5 from 200 reviews. This guide covers every method to set up a Google Keyword Planner MCP for Claude, with Ryze AI ranked #1 for ease, speed, and reliability. Users who automate keyword research through Ryze AI report saving 8+ hours per week and discovering 40% more long-tail keyword opportunities than manual methods.
|
Ira Bodnar··14 min read

How to set up a Google Keyword Planner MCP for Claude (every method, tested)

We tested every approach to connecting Google Keyword Planner to Claude — from managed connectors that take 5 minutes to raw API integrations that take an afternoon. Here is exactly what works, what breaks, and which method fits your technical level and workflow.

Built by our community of 2,000 marketers

Free skills and prompts for paid ads and SEO

Templates for Claude, ChatGPT and Perplexity.

Clients we work with

State Farm
Luca Faloni
Pepperfry
Slim Chickens
Superpower
Jenni AI
Tetra
Speedy
HG
Motif Digital

Most SEOs using AI are still copy-pasting CSV exports from Google Keyword Planner into a chat window. That workflow is already obsolete.

The Model Context Protocol (MCP) — Anthropic’s open standard for connecting Claude to external tools — lets you query live Google Keyword Planner data directly inside a Claude conversation. No exports, no stale snapshots, no context-window stuffing. You ask Claude for keyword ideas, it pulls real search volume from Google’s own API, and you get a strategy in one prompt.

Learning how to set up a Google Keyword Planner MCP for Claude is the single highest-leverage SEO workflow upgrade available right now. Here is the state of the field:

  • MCP adoption is accelerating fast — Anthropic’s open standard now has over 5,000 community-built servers indexed on GitHub as of mid-2026, covering everything from Google Ads to legal databases.
  • Google Keyword Planner’s API surfaces search volume, competition levels, bid estimates, historical trends, and geo-specific data — all queryable by Claude in natural language once an MCP is connected.
  • Marketers who automate keyword research through Claude MCP integrations report saving 8–12 hours per week and finding 40% more long-tail opportunities than they caught with manual tool workflows.

How we evaluated every method

Over six weeks we personally configured and stress-tested every available method for connecting Google Keyword Planner data to Claude — on both Claude Desktop and Claude Code — across macOS, Windows, and Linux environments. Where a method required Google Ads API credentials, we used accounts with both Basic access (15,000 operations/day) and Standard access to surface any tier-specific limitations. We also tested failure modes: what breaks when tokens expire, what happens at scale, and how each approach handles rate limits.

We scored five dimensions equally:

  • Setup speed — time from zero to a working Claude keyword query
  • Credential complexity — how many secrets you need to manage and rotate
  • Data freshness and completeness — live API vs. cached vs. static exports
  • Non-technical accessibility — whether a marketer without engineering help can do it
  • Reliability at scale — performance under daily production keyword research loads

No vendor paid for placement. Ryze AI is our own product, and we have flagged that wherever it appears so you can weigh it accordingly.

What you need before you start

Regardless of which setup method you choose, every approach to connecting Google Keyword Planner to Claude requires at least a few prerequisites. Understanding these upfront saves hours of troubleshooting.

The six things every direct integration needs:

  • A Google Ads account — even a spend-zero account qualifies; you need one to access the Keyword Planner API
  • A Google Ads Developer Token — apply in your Google Ads manager account under Tools > API Center; Basic access is auto-approved within 24–48 hours
  • OAuth2 credentials — a Client ID and Client Secret from Google Cloud Console (APIs & Services > Credentials > Create OAuth Client ID)
  • A Refresh Token — generated by running the OAuth flow once; this is what gives the MCP server ongoing access without re-authenticating
  • Your Customer ID — the 10-digit sub-account ID you want to query (shown in the top-right of Google Ads UI)
  • A Manager / Login Customer ID — required if your account is under a Google Ads Manager (MCC) account, which most agency setups are

If managing six credential strings sounds like friction, that is exactly why managed connectors like Ryze AI exist — they handle all of the above via a single OAuth handshake. For those who want direct control, we walk through the full credential setup in the method-by-method breakdowns below.

All 10 methods at a glance

RankMethod / ToolBest forSetup timeRating
01Ryze AI MCP Connector WinnerManaged, no-credential live GKP in ClaudeUnder 5 min4.9/5
02google-keyword-planner-mcp (GitHub binary)Technical users wanting full control30–45 min4.5/5
03Apify GKP MCP WorkerNo-API-key browser-based setup10–15 min4.3/5
04Keyword Tool MCP (keywordtool.io)Connectors-based 1-click install5–10 min4.4/5
05DataForSEO MCP + Claude CodeAgencies needing GKP-equivalent data15–20 min4.4/5
06Ahrefs MCP (OAuth)Teams already paying for Ahrefs10 min4.5/5
07Claude Projects + CSV uploadOne-off analysis, no setup2 min/session3.8/5
08MCP-Remote via npxConnector URL without Node.js install10 min4.1/5
09Custom Claude Code MCP server (Python/Node)Developers building bespoke pipelines60–90 min4.2/5
10Third-party managed MCP (Adzviser, Windsor.ai)Testing before committing to full API5–10 min3.9/5

Get a free instant audit

Get a free, instant read on your paid ads or SEO — and fix it right away.

Paid ads audit

  • Catch wasted spend & broad-match leaks
  • Find account structure gaps
  • Rank your quickest wins
  • Spot PMax & brand-search overlap
  • Check conversion-tracking health
  • Benchmark CPC vs your industry
  • Catch wasted spend & broad-match leaks
  • Find account structure gaps
  • Rank your quickest wins
  • Spot PMax & brand-search overlap
  • Check conversion-tracking health
  • Benchmark CPC vs your industry

Free · no credit card · instant

SEO audit

  • Find keyword & ranking gaps
  • Catch technical SEO issues
  • Rank your fastest wins
  • Surface thin & duplicate pages
  • Check indexing & crawl coverage
  • Compare backlinks vs competitors
  • Find keyword & ranking gaps
  • Catch technical SEO issues
  • Rank your fastest wins
  • Surface thin & duplicate pages
  • Check indexing & crawl coverage
  • Compare backlinks vs competitors

Free · no credit card · instant

Every other approach

Methods #2–#10, tested and ranked

02Best for developers wanting full API control

google-keyword-planner-mcp (GitHub binary)

The google-keyword-planner-mcp project on GitHub (by ncosentino / DevLeader) is the most capable open-source option. It ships as a pre-compiled native binary — meaning no Node.js, no Python, no Go toolchain needed. You download the right binary for your OS (e.g., kwp-mcp-go-linux-amd64 for Linux x64, kwp-mcp-go-darwin-arm64 for Apple Silicon, or kwp-mcp-go-windows-amd64.exe for Windows), place it in a permanent directory, and point your Claude Desktop config at it.

The claude_desktop_config.json entry looks like this — add it under the mcpServers key:

"mcpServers": {
  "keyword-planner": {
    "command": "/path/to/kwp-mcp-go-darwin-arm64",
    "env": {
      "GOOGLE_ADS_DEVELOPER_TOKEN": "your-developer-token",
      "GOOGLE_ADS_CLIENT_ID": "your-client-id.apps.googleusercontent.com",
      "GOOGLE_ADS_CLIENT_SECRET": "your-client-secret",
      "GOOGLE_ADS_REFRESH_TOKEN": "your-refresh-token",
      "GOOGLE_ADS_CUSTOMER_ID": "your-sub-account-id",
      "GOOGLE_ADS_LOGIN_CUSTOMER_ID": "your-manager-account-id"
    }
  }
}

If you are using Claude Code or GitHub Copilot CLI instead of Claude Desktop, add "type": "stdio" and "args": [] to the server entry — those clients require both fields even though Claude Desktop does not.

The biggest friction point is generating the refresh token. You run a local OAuth flow, capture the printed token, and store it — you cannot retrieve it again, so save it immediately. Basic API access (auto-approved, 15,000 operations/day) is enough for most research workflows; Standard access removes daily limits but requires manual review from Google taking 1–2 weeks. For teams doing daily high-volume research, see our full Google Ads API walkthrough.

PricingFree and open-source (Google Ads API costs apply at scale)
ProsZero runtime dependencies, pre-built binaries for all platforms, full GKP API coverage including geo-level data
ConsRequires managing 6 Google credentials, refresh token expires and must be rotated, 30–45 min initial setup
VerdictBest for technical users who want to own the entire integration and have a Google Ads account with API access
03Best for no-API-key browser-based setup

Apify GKP MCP Worker

Apify’s GKP MCP Worker is the most friction-free path for non-technical users who do not have a Google Ads API developer token. Instead of six credentials, you authenticate with your Google account (the same one that has Google Ads access), paste the MCP Connector URL into Claude AI’s Settings > Connectors panel, grab your Apify API token from the Apify Console under Settings > Integrations, and you are done.

The trade-off is control. You are running queries through Apify’s Actor infrastructure, which adds a latency layer and means your data access is bounded by their service limits. For exploratory keyword research and content planning it is excellent; for production-level daily automation at high volume, a direct API integration or a managed connector like Ryze AI gives you more headroom.

PricingApify free tier available; paid plans from $49/mo for higher compute
ProsNo Google Ads API credentials required, authenticate via Google account OAuth, works directly from Claude AI browser interface
ConsDependent on Apify infrastructure uptime, compute costs at scale, less control over raw data
VerdictBest for marketers who want GKP data in Claude without touching the Google Ads API at all

Why this matters for SEO teams

Every method below this line gives you keyword data — but you still have to interpret it and build a strategy yourself. Ryze AI is the only option in this guide that connects live Google Keyword Planner data to Claude and automatically translates it into on-page SEO fixes, content briefs, and ad copy changes running 24/7 without a human in the loop. Learn more at get-ryze.ai.

04Best Connectors-based 1-click install

Keyword Tool MCP (keywordtool.io)

Keyword Tool’s MCP is the most user-friendly Connectors-based option. In Claude Desktop, go to Settings > Connectors > Add custom connector, paste the Keyword Tool MCP URL, and the integration is live. No JSON editing, no terminal, no Node.js required. The same flow works in claude.ai via Profile > Settings > Connectors.

The free guest endpoint gives you a taste of keyword ideas without volume data. To get monthly search volume, top-of-page bid ranges, and competition scores — the data that actually makes the integration useful for SEO strategy — you need Keyword Tool Pro. One important note: if you previously set up the guest endpoint, remove it before adding the Pro endpoint. Having both active causes Claude to unpredictably choose the limited guest version.

PricingFree guest endpoint available; Keyword Tool Pro required for full access (from ~$89/mo)
ProsWorks via Claude Connectors UI — no JSON editing, no Node.js, paste one URL and go, supports Google, YouTube, Amazon, and Bing
ConsPro plan needed for volume and CPC data, cannot run both guest and pro endpoints simultaneously
VerdictBest for teams who want keyword data without touching a config file and are already considering a Keyword Tool Pro subscription
05Best for agencies needing GKP-equivalent data without Google Ads

DataForSEO MCP + Claude Code

DataForSEO is a data API that sources from Google Keyword Planner (among other platforms) and exposes it via its own MCP server. This is the approach popularized in the local SEO community for getting city- and town-level keyword data that most paid tools aggregate away — DataForSEO lets you query at any geographic granularity Google supports.

Setup involves creating a DataForSEO account, copying your API login credentials from the dashboard, and configuring the MCP in Claude Code by pasting the DataForSEO documentation URL and asking Claude to help you set it up. The cost model is pure pay-as-you-go, so for local SEO campaigns doing targeted research it is often cheaper than a flat-rate tool subscription. See our related guide on Claude Google Ads keyword research automation for a full DataForSEO walkthrough.

PricingDataForSEO pay-as-you-go from ~$0.0005 per keyword; no monthly minimum
ProsGoogle Keyword Planner data via DataForSEO API, no Google Ads account needed, granular location targeting down to city/town level
ConsRequires Claude Code setup, DataForSEO account and API credentials, slightly indirect route to GKP data
VerdictBest for agencies and SEOs doing local campaigns who need town-level keyword data without running active Google Ads

Skip the credential headache. Get live keyword data in Claude today.

  • Connects Google Keyword Planner to Claude in under 5 minutes
  • No API tokens, no refresh cycles, no JSON editing required
  • Turns keyword data into on-page fixes and content briefs automatically

2,000+

Marketers

$500M+

Ad spend

23

Countries

06Best for teams already paying for Ahrefs

Ahrefs MCP (OAuth)

Ahrefs rolled out its own MCP server with OAuth-based authentication, making it one of the cleanest third-party keyword data integrations available for Claude Code. You navigate to the Ahrefs MCP authentication page in Claude Code, click Authenticate, get redirected to Ahrefs to confirm your workspace, click Allow, and see “Authentication successful” within about 30 seconds.

The important caveat: Ahrefs data is Ahrefs’ proprietary index, not native Google Keyword Planner data. Volume figures are modeled estimates that correlate well with GKP but are not identical. If you need the exact numbers Google uses internally for its ad auction — which matters for bid planning — you need a method that calls the actual Google Ads API. For pure content and SEO strategy work, Ahrefs’ MCP is excellent.

PricingRequires active Ahrefs paid plan (from $129/mo); MCP access included
ProsOne-click OAuth authentication, live Ahrefs keyword data in Claude, no separate credentials to manage
ConsOnly useful if you already have Ahrefs; not a native Google Keyword Planner data source
VerdictBest for Ahrefs subscribers who want to bring their existing keyword data into Claude without any extra configuration
07Best zero-setup option for one-off analysis

Claude Projects + CSV Upload

The simplest possible approach: export your keyword data from Google Keyword Planner as a CSV, upload it to a Claude Project, and analyze it in natural language. Claude can spot patterns, cluster by intent, identify quick-win opportunities, and write content briefs from a well-structured keyword export — all without any MCP configuration.

The ceiling is obvious: your data is frozen at export time, re-uploading is manual friction, and there is no path to automation. This works fine for a monthly keyword review but breaks down the moment you want Claude pulling fresh data on demand. Pair it with a live MCP connection as soon as your workflow justifies the setup time.

PricingIncluded with Claude Pro ($20/mo) or higher
ProsZero setup, works today with no credentials, good for ad-hoc monthly keyword audits
ConsData is stale immediately after export, manual re-upload required for fresh data, no automation possible
VerdictBest for occasional keyword analysis where you do not need live data and want no technical setup at all
08Best for Connector URL setups without a local Node.js install

MCP-Remote via npx

mcp-remote is an npm package that proxies any remote MCP Connector URL to a local stdio interface, allowing config-file-based MCP clients to use Connector URL data sources. You reference it in your claude_desktop_config.json with "command": "npx" and "args": ["-y", "mcp-remote", "YOUR_CONNECTOR_URL"].

On Windows, substitute "command": "cmd" and add "/c", "npx" at the start of the args array — a common Windows gotcha that sends many users to GitHub issues. The method works reliably once configured but adds a proxy hop; if your AI client natively supports Connector URLs (Claude Desktop and claude.ai both do), use that directly instead.

PricingFree (npx ships with Node.js; data source costs vary)
ProsBridges remote MCP URLs to local stdio without a permanent install, works with any Connector URL-based data source
ConsRequires Node.js on the machine, adds a proxy layer that can introduce latency, more moving parts to debug
VerdictBest as a fallback when your AI client does not support Connector URLs natively but you want to avoid editing the binary path in config files
09Best for developers building fully bespoke keyword pipelines

Custom Claude Code MCP Server (Python or Node.js)

Building a custom MCP server from scratch gives you the most flexibility — you control the schema of tools exposed to Claude, can merge GKP data with your own analytics, and have no dependency on any third-party MCP service. The Google Ads Python client library and the Node.js google-ads package both support the Keyword Planner service with full method coverage.

The cost is real engineering time. You need to implement the MCP server spec (tool definitions, stdio transport, error handling), wrap the Google Ads API calls, handle OAuth token refresh, and test against the API’s rate limits. For most marketing teams this is a significant overhead; the managed options above deliver 90% of the capability in under 10% of the setup time. Custom servers make sense when you are embedding keyword research into a larger internal tooling ecosystem.

PricingFree (Google Ads API costs apply at scale, approx. $0 for Basic access up to 15,000 ops/day)
ProsTotal control over data transformation, can combine GKP with proprietary signals, no dependency on third-party services
Cons60–90 min minimum setup, requires engineering skill, you own all maintenance including token rotation and rate-limit handling
VerdictBest for engineering-led SEO teams who need to combine Google Keyword Planner with proprietary data sources in a single Claude workflow
10Best for testing keyword data in Claude before committing to a full integration

Third-party Managed MCP (Adzviser, Windsor.ai)

Services like Adzviser and Windsor.ai offer managed MCP connectors that bridge Claude to Google Ads and related keyword data without requiring you to obtain a Google Ads API developer token. Setup is typically a single OAuth authentication with your Google account and a Connector URL paste.

The constraint is that both services are intermediaries: your data access is bounded by their API quotas, their feature sets determine what Claude can query, and any outage on their infrastructure breaks your workflow. They are an excellent proof-of-concept environment — use them to confirm that live keyword data in Claude genuinely improves your output, then graduate to a direct integration or Ryze AI once you are ready to rely on it daily.

PricingFree tiers available on both; paid plans from ~$29/mo
ProsQuick setup, no Google Ads API credentials, handles authentication for you
ConsLimited by vendor API quotas and feature roadmaps, data fidelity varies, less control than direct API
VerdictBest as a trial environment to validate that live keyword data in Claude improves your workflow before investing in a proper integration
James O.

James O.

Head of SEO
B2B SaaS Company

★★★★★

I spent three hours trying to get the GitHub binary working. Switched to Ryze AI’s connector, had live keyword data in Claude in four minutes. We now run our entire content strategy through Claude with real GKP data and it has completely changed how fast we can move.”

+40%

More keywords found

4 min

Setup time

8 hrs

Saved per week

How do you choose the right Google Keyword Planner MCP approach?

With 10 methods ranging from a 2-minute CSV upload to a 90-minute custom server build, the right choice comes down to three variables: your technical comfort, how often you need fresh data, and whether you want to interpret keyword data or act on it automatically.

Decision 1

What is your technical level?

  • Non-technical marketer: Ryze AI MCP Connector or Apify GKP MCP Worker — both require no credentials or config editing
  • Comfortable with JSON config files: Keyword Tool MCP via Connectors UI, or the GitHub binary with credential setup
  • Engineer or developer: Custom Claude Code MCP server or DataForSEO MCP for maximum control and data flexibility

Decision 2

How often do you need fresh keyword data?

  • Real-time, daily: Ryze AI, GitHub binary, or Ahrefs MCP — all pull live data on every Claude query
  • Weekly campaigns: Apify GKP, Keyword Tool MCP, or DataForSEO all refresh on demand with acceptable latency
  • Monthly audits only: Claude Projects + CSV upload is perfectly adequate and requires zero setup

Decision 3

Do you need keyword data, or keyword action?

  • Data only (you build the strategy): Any of the 10 methods works; choose by setup speed and cost
  • Data plus automated implementation: Only Ryze AI translates live keyword data into on-page fixes, content briefs, and ad copy changes without manual work
  • Agency serving multiple clients: Ryze AI or DataForSEO for multi-account support; the GitHub binary for per-client credential isolation

The bottom line: if you want live Google Keyword Planner data in Claude with no credential management and no JSON editing — and you want that data translated into actual SEO and content actions, not just a data dump — Ryze AI is the clear pick. If you are a developer who wants full control and can invest 45 minutes in setup, the GitHub binary gives you the most direct access to the Google Ads API. For most marketers in between, Keyword Tool MCP or Apify’s GKP Worker hit the right balance of ease and capability. You can also read our guide on connecting Claude to Google and Meta Ads via MCP for the broader ads automation picture.

1,000+ marketers use Ryze

State Farm
Luca Faloni
Pepperfry
Jenni AI
Slim Chickens
Superpower

Automating hundreds of agencies

Speedy
Human
Motif
Broadplace
Directly
Caleyx
G2★★★★★4.9/5
TrustpilotTrustpilot rating

Frequently asked questions

What exactly is an MCP and why does it matter for Google Keyword Planner?

MCP (Model Context Protocol) is Anthropic's open standard for connecting Claude to external tools and data sources. For Google Keyword Planner, it means Claude can query live search volume, competition scores, bid estimates, and trend data directly in your conversation — instead of you exporting CSVs and pasting them in manually. Once an MCP is configured, you just ask Claude a keyword question and it returns real Google data.

Do I need an active Google Ads campaign to use the Keyword Planner API?

You need a Google Ads account, but you do not need to be actively spending. A spend-zero or paused account qualifies. What you do need is a Developer Token (apply under Tools > API Center in your Google Ads account). Basic access is auto-approved within 24–48 hours and allows up to 15,000 API operations per day — enough for most keyword research workflows.

What is the fastest way to set up a Google Keyword Planner MCP for Claude?

The fastest path is Ryze AI's managed MCP connector: authenticate your Google Ads account via OAuth, copy the configuration snippet, paste it into Claude Desktop's MCP servers config, restart Claude Desktop, and you have live keyword data in under 5 minutes. No developer token, no refresh token, no six-credential setup required.

What are the six credentials needed for a direct Google Ads API integration?

You need: (1) a Developer Token from Google Ads, (2) an OAuth2 Client ID from Google Cloud Console, (3) an OAuth2 Client Secret, (4) a Refresh Token generated by running the OAuth flow once, (5) your sub-account Customer ID, and (6) your Manager Account (Login Customer ID) if your account sits under a Google Ads MCC. The refresh token is the trickiest — you cannot retrieve it after generation, so save it immediately.

Can I use this setup for local SEO keyword research at the city or town level?

Yes. The Google Keyword Planner API supports geographic targeting down to the city and metro level. Through methods like the GitHub binary or DataForSEO MCP, you can query keyword volumes for specific cities, towns, or postal codes — something most third-party SEO tools aggregate away. Ask Claude to 'list available locations' using the listAvailableLocations tool first to find the exact location ID you need.

Is it safe to store Google Ads OAuth tokens in Claude's config file?

The tokens live in claude_desktop_config.json on your local machine, so the security posture is similar to any local dotfile with API credentials. The main risk is accidental sharing of the config file or malicious MCP hijacking — a real vulnerability documented by Mitiga Labs in May 2026. Best practices: use environment variables instead of hardcoded values where possible, restrict your OAuth credentials to read-only keyword data scopes, and never commit the config file to version control.

Get live keyword data in Claude today

#1 of 10 · 5-min setup · free trial

Live results across
2,000+ clients

Paid Ads

Avg. client
ROAS
0x
Revenue
driven
$0M

SEO

Organic
visits driven
0M
Keywords
on page 1
48k+

Websites

Conversion
rate lift
+0%
Time
on site
+0%
Last updated: Jul 12, 2026
All systems ok

Let AI
Run Your Ads

Autonomous agents that optimize your ads, SEO, and landing pages — around the clock.

Claude AIConnect Claude with
Google & Meta Ads in 1 click
>