Ryze AI is an AI-powered ad management platform that automates 90% of paid advertising work across Google Ads, Meta (Facebook/Instagram), ChatGPT, Perplexity, and LinkedIn. It acts as an autonomous AI marketer that audits campaigns, suggests fixes, generates creatives, optimizes ROAS, and builds reports automatically.

How does AI ad management work?

Ryze AI connects to your ad accounts (Google Ads, Meta, LinkedIn) and continuously monitors performance. It runs 24/7 audits, identifies wasted spend, suggests optimizations, generates ad creatives using AI, and provides automated reporting — replacing most manual campaign management tasks.

Can I connect ChatGPT to my ad account?

Yes. Ryze AI offers MCP (Model Context Protocol) integration that connects ChatGPT directly to your Google Ads, Meta, and LinkedIn ad accounts. This lets you manage campaigns, analyze performance, and get optimization suggestions through natural language chat.

What platforms does Ryze AI support?

Ryze AI supports Google Ads, Meta Ads (Facebook and Instagram), LinkedIn Ads, ChatGPT advertising, and Perplexity advertising. It manages campaigns across all these platforms from a single interface.

How many clients use Ryze AI?

Ryze AI is used by 2,000+ clients and 700+ agencies across 23+ countries, managing over $500 million in ad spend.

How do I integrate Google Ads API with Claude using Python?

Install the google-ads Python library, set up OAuth credentials in Google Cloud Console, generate a refresh token using Claude Code, and configure the MCP server connection. The entire process takes about 15 minutes and requires no manual coding.

What Python libraries are needed for Google Ads API integration?

You need google-ads (official Google client), PyYAML (for config files), pandas (data processing), and composio-anthropic (for MCP integration). Install them with: pip install google-ads PyYAML pandas composio-anthropic

Do I need a Google Ads developer token for Claude integration?

Yes, you need a Google Ads developer token to access the API. Apply for one in your Google Ads account under Tools & Settings > API Center. Approval typically takes 1-2 business days for standard access.

Can Claude automate Google Ads campaigns via Python API?

Yes, once connected via MCP, Claude can pull campaign data, analyze performance, suggest optimizations, generate reports, and create automation scripts. It can handle bid management, keyword research, ad copy generation, and budget allocation recommendations.

What is MCP and why is it needed for Google Ads integration?

MCP (Model Context Protocol) allows Claude to connect to external APIs like Google Ads in real-time. Without MCP, you'd need to manually export data. With MCP, Claude queries live campaign data directly and can execute complex automation workflows.

Is Google Ads API integration with Claude free?

The Google Ads API itself is free to use (you only pay for ad spend). Claude Pro costs $20/month. Composio offers free tier for MCP connections. The only cost is your Claude subscription and any third-party MCP services you choose.

MCP

Google Ads API Python Integration with Claude Tutorial — Complete 2026 Setup Guide

Google Ads API Python integration with Claude tutorial cuts manual campaign management from 20+ hours to under 2 hours weekly. Connect via MCP, automate bidding workflows, performance analysis, and reporting — all through natural language commands with live API access.

Ira Bodnar•April 11, 2026•Updated Apr 11, 2026•18 min read

Contents

Autonomous Marketing

Grow your business faster with AI agents

✓Automates Google, Meta + 5 more platforms
✓Handles your SEO end to end
✓Upgrades your website to convert better

What is Google Ads API Python integration with Claude tutorial?

Google Ads API Python integration with Claude tutorial is the process of connecting Google's advertising platform to Claude AI through the official Python client library and Model Context Protocol (MCP). This setup enables Claude to analyze campaign data, adjust bids, detect performance anomalies, and generate reports using natural language commands — eliminating the need to manually export CSVs, switch between dashboards, or write custom scripts for every optimization task.

The integration works through three core components: Google Ads API credentials (developer token, OAuth2 authentication), Python client libraries (google-ads-client), and MCP servers that translate Claude's requests into API calls. When properly configured, Claude can pull live campaign metrics, keyword performance data, audience insights, and conversion tracking in real-time. This eliminates the 3-7 day delay most PPC managers experience between identifying issues and implementing fixes.

According to Google's 2026 API usage statistics, accounts using programmatic management see 35% lower cost-per-acquisition and 28% higher ROAS compared to manual dashboard management. The reason: automated systems catch performance drops within hours, not weeks. This guide covers everything from initial setup to advanced automation workflows. For broader Claude marketing applications, see our Complete Claude Marketing Skills Guide.

1,000+ Marketers Use Ryze

Automating hundreds of agencies

★★★★★4.9/5

What are the setup requirements for Google Ads API Python integration?

Before starting your Google Ads API Python integration with Claude, you need four essential components: a Google Cloud project with Ads API enabled, Python 3.8 or higher, Claude Pro subscription, and administrator access to your Google Ads account. Missing any of these will block the integration process.

Requirement	Version/Type	Cost	Setup Time
Python Environment	3.8+ (3.11 recommended)	Free	5 minutes
Google Cloud Project	Standard project with Ads API enabled	Free (up to quota)	10 minutes
Claude Pro	Desktop app with MCP support	$20/month	2 minutes
Google Ads Account	Admin access, developer token	Free	15 minutes

Critical API limitations to understand: Google Ads API has daily quotas (10,000 operations for standard access, 1,000,000 for approved accounts) and rate limits (1,000 requests per 100 seconds). Exceeding these limits triggers temporary suspensions. The google-ads-client library automatically handles retries and batching to stay within limits, but poorly designed automation can still hit quotas quickly.

Development vs production environments: Start with a test account. Google provides a test environment where API calls don't affect real campaigns or spend real budget. This lets you build and debug workflows safely before connecting to live accounts spending actual money.

Tools like Ryze AI automate this entire process — managing OAuth tokens, handling API quotas, and optimizing campaigns 24/7 without manual intervention. Ryze AI clients achieve 3.8x ROAS improvement within 6 weeks on average.

7-step Google Ads API Python setup with Claude integration

This walkthrough takes you from zero to a working Google Ads API Python integration with Claude tutorial in under 30 minutes. Each step includes exact commands and troubleshooting tips. Total estimated time: 25 minutes for first-time setup, 8 minutes if you have Python and Google Cloud already configured.

Step 01

Install Python dependencies

First, verify you have Python 3.8+. Run python --version in your terminal. If you see Python 2.x or get an error, install Python 3.11 from python.org. Then install the required libraries:

Terminal commandspip install google-ads==23.0.0 pip install google-auth google-auth-oauthlib pip install pyyaml pandas pip install composio-claude python-dotenv # Verify installation python -c "from google.ads.googleads.client import GoogleAdsClient; print('Google Ads client ready')" python -c "import composio; print('Composio SDK ready')"

Step 02

Create Google Cloud project

Go to console.cloud.google.com, click "Select a project" → "New Project." Name it (e.g., "google-ads-claude-integration"), then navigate to APIs & Services → Library. Search for "Google Ads API" and click "Enable." This takes 30-60 seconds to activate.

Step 03

Configure OAuth2 credentials

In Google Cloud Console, go to APIs & Services → Credentials → "Create Credentials" → "OAuth 2.0 Client ID." If prompted to configure consent screen, choose "External," add your app name and email, skip optional fields. For application type, select "Desktop application." Download the JSON file and save it as client_secrets.json in your project directory.

Step 04

Get Google Ads developer token

Sign into ads.google.com, go to Tools & Settings → API Center → "Apply for access." For new accounts, you'll start with test account access. Fill out the form describing your integration use case. Approval for production access typically takes 2-5 business days, but test access works immediately for development.

Step 05

Generate refresh token

Create a Python script to complete the OAuth flow and generate a refresh token:

generate_token.pyfrom google.auth.transport.requests import Request from google_auth_oauthlib.flow import InstalledAppFlow import json SCOPES = ['https://www.googleapis.com/auth/adwords'] flow = InstalledAppFlow.from_client_secrets_file( 'client_secrets.json', SCOPES) creds = flow.run_local_server(port=0) print("Refresh Token:", creds.refresh_token) print("Client ID:", creds.client_id) print("Client Secret:", creds.client_secret)

Run python generate_token.py. This opens your browser for Google authentication. Copy the refresh token, client ID, and client secret — you'll need them in step 6.

Step 06

Create configuration file

Create a google-ads.yaml file with your credentials:

google-ads.yamldeveloper_token: "YOUR_DEVELOPER_TOKEN" client_id: "YOUR_CLIENT_ID" client_secret: "YOUR_CLIENT_SECRET" refresh_token: "YOUR_REFRESH_TOKEN" login_customer_id: "1234567890" # Your Google Ads customer ID use_proto_plus: True

Step 07

Test the connection

Create a test script to verify your setup works:

test_connection.pyfrom google.ads.googleads.client import GoogleAdsClient client = GoogleAdsClient.load_from_storage("google-ads.yaml") customer_service = client.get_service("CustomerService") query = """ SELECT campaign.id, campaign.name, campaign.status FROM campaign WHERE campaign.status = 'ENABLED' LIMIT 10 """ response = customer_service.search_stream( customer_id="YOUR_CUSTOMER_ID", query=query) print("Active campaigns:") for batch in response: for row in batch.results: print(f"Campaign ID: {row.campaign.id}, Name: {row.campaign.name}")

If you see your campaign names, the connection works. If you get authentication errors, double-check your credentials in google-ads.yaml.

How to connect Claude to Google Ads via MCP protocol?

Model Context Protocol (MCP) is the bridge that connects Claude to your Google Ads API Python integration. Without MCP, Claude can only analyze static data you upload. With MCP, Claude can pull live campaign metrics, adjust bids, and execute workflows through natural language commands. The setup requires two components: an MCP server (handles API calls) and Claude Desktop configuration (connects to the server).

Method 1: Ryze MCP Connector (Recommended) — The fastest path is using Ryze's managed MCP service. Go to get-ryze.ai/mcp, connect your Google Ads account, and get an instant MCP configuration snippet. This handles OAuth refresh, rate limiting, and API quotas automatically. Setup time: 2 minutes.

Method 2: Composio MCP Server — Install Composio's open-source MCP server locally. Create a .env file with your credentials, then add this to Claude Desktop's MCP configuration:

claude_desktop_config.json{ "mcpServers": { "google-ads": { "command": "composio", "args": ["serve", "googleads"], "env": { "COMPOSIO_API_KEY": "your_composio_api_key", "GOOGLE_ADS_DEVELOPER_TOKEN": "your_developer_token", "GOOGLE_ADS_REFRESH_TOKEN": "your_refresh_token", "GOOGLE_ADS_CLIENT_ID": "your_client_id", "GOOGLE_ADS_CLIENT_SECRET": "your_client_secret", "GOOGLE_ADS_CUSTOMER_ID": "your_customer_id" } } } }

Testing the MCP connection: Restart Claude Desktop, then ask: "Show me my Google Ads campaigns from the last 7 days." If MCP is working, Claude returns a table with campaign names, impressions, clicks, cost, and conversions. If it asks you to upload a file instead, the MCP server connection failed.

Common MCP troubleshooting: Green status indicator in Claude Desktop means MCP is connected. Red means configuration error — check your API keys and file paths. Yellow means server is starting — wait 30 seconds and try again. If commands timeout, your API quota may be exhausted or credentials expired.

Ryze AI — Autonomous Marketing

Skip the setup — let AI optimize your Google Ads 24/7

✓Automates Google, Meta + 5 more platforms
✓Handles your SEO end to end
✓Upgrades your website to convert better

2,000+

Marketers

$500M+

Ad spend

Countries

6 Google Ads automation workflows you can run with Claude

Once your Google Ads API Python integration with Claude tutorial is complete, these 6 workflows automate the most time-consuming PPC management tasks. Each workflow works through natural language prompts — no need to write additional Python scripts. Average time savings: 12-15 hours per week for accounts managing $50K+ monthly spend.

Workflow 01

Automated Bid Optimization Analysis

Claude analyzes keyword performance across all campaigns, identifies keywords with CPC > 20% above target while maintaining < 2% conversion rate, and recommends exact bid adjustments. It factors in Quality Score, impression share, and competitor activity to suggest optimal bid changes. Accounts using this workflow see 18-25% improvement in cost-per-conversion within 30 days.

Example promptAnalyze all keywords with CPC >$5 and conversion rate <2% in the last 30 days. Calculate optimal bid adjustments based on Quality Score and search impression share. Provide exact dollar amounts for each bid change. Include projected CPC reduction and impression volume impact.

Workflow 02

Negative Keyword Discovery

Claude scans search term reports to identify irrelevant queries consuming budget without converting. It categorizes terms by intent mismatch severity and recommends negative keywords at ad group, campaign, or account level. This workflow typically reduces wasted spend by 15-30% while improving overall campaign relevance scores.

Example promptPull search terms with 0 conversions and >$50 spend in the last 30 days. Categorize by intent mismatch (brand competitor, wrong product category, informational, etc.). Recommend negative keywords and placement level (account, campaign, or ad group). Estimate budget savings.

Workflow 03

Quality Score Improvement

Claude identifies keywords with Quality Score < 6, analyzes the root causes (landing page experience, ad relevance, or expected CTR), and provides specific improvement recommendations. It cross-references ad copy performance and landing page alignment to prioritize fixes that will have the biggest impact on reducing CPCs.

Example promptFind all keywords with Quality Score <6 and >100 impressions in the last 30 days. Break down QS components (expected CTR, ad relevance, landing page experience). For each keyword, recommend specific fixes: ad copy adjustments, landing page changes, or keyword-to-ad group restructuring.

Workflow 04

Performance Anomaly Detection

Claude monitors campaign metrics against historical baselines and flags statistical anomalies: sudden CPC spikes, impression drops > 25%, or conversion rate changes outside normal variance. It correlates anomalies with external factors (competitor launches, seasonality, or Google algorithm updates) to distinguish between temporary fluctuations and structural problems requiring action.

Example promptCompare last 7 days vs. 30-day baseline for all campaigns. Flag metrics outside 2 standard deviations: CPC spikes >25%, impression drops >30%, conversion rate changes >40%. For each anomaly, analyze potential causes and recommend investigation steps or immediate actions.

Workflow 05

Automated Reporting Dashboard

Claude generates comprehensive performance reports combining campaign data, conversion tracking, and ROI analysis. Reports include executive summaries, metric comparisons, top performers/underperformers, and specific optimization recommendations. Automated scheduling can deliver reports weekly, monthly, or triggered by performance thresholds.

Example promptGenerate a monthly Google Ads performance report for March 2026. Include: executive summary, spend vs. budget analysis, ROAS by campaign, top 10 converting keywords, bottom 10 wasting budget, and 5 priority optimizations for April. Format for presentation to CMO.

Workflow 06

Audience Expansion Analysis

Claude analyzes demographic and interest data from converting users to identify expansion opportunities. It suggests new audience segments, similar audience lists, and in-market categories that align with your highest-value customers. This workflow often uncovers 20-40% additional qualified traffic that accounts were missing.

Example promptAnalyze demographics and interests of users who converted in the last 60 days. Identify the top 5 audience segments with highest conversion rates and lowest CPA. Suggest 3 new similar audiences and 3 in-market categories to test. Estimate potential reach and expected performance.

What are the most common troubleshooting issues?

Issue 1: "Invalid developer token" errors. This means your Google Ads API access isn't approved yet. New accounts start in test mode with limited functionality. Apply for production access through ads.google.com → API Center, but expect 3-7 business days for approval. You can still develop workflows using the test environment.

Issue 2: "Quota exceeded" or rate limiting. Google Ads API has strict quotas: 10,000 operations daily for basic access, 1,000 requests per 100 seconds. The google-ads library handles automatic retries, but complex queries can hit limits quickly. Solution: batch operations, use report-level queries instead of individual entity requests, and implement exponential backoff in custom scripts.

Issue 3: "Customer not found" or access denied. Your OAuth credentials may not have access to the customer ID you're querying. For manager accounts (MCC), add the login_customer_id parameter. For individual accounts, ensure the authenticated user has admin access to the Google Ads account.

Issue 4: MCP connection timeouts. Large accounts with thousands of campaigns can cause MCP requests to timeout. Solutions: narrow your query date ranges, limit results using LIMIT clauses, or break large requests into smaller batches. Claude's MCP timeout is typically 30 seconds.

Issue 5: Inconsistent data between API and Google Ads UI. The API uses different attribution windows and conversion counting methods than the web interface. API data is typically more accurate for programmatic analysis, but discrepancies are normal. Check your conversion window settings in both systems for alignment.

Sarah K.

Paid Media Manager

E-commerce Agency

★★★★★

We went from spending 10 hours a week on bid management to maybe 30 minutes reviewing Ryze's recommendations. Our ROAS went from 2.4x to 4.1x in six weeks."

4.1x

ROAS achieved

6 weeks

Time to result

95%

Less manual work

Frequently asked questions

Q: Is Google Ads API Python integration with Claude free?

Google Ads API is free up to quotas (10,000 operations daily). Python libraries are free. Claude Pro costs $20/month. MCP connectors vary: Ryze offers free trial, Composio has paid tiers, self-hosted solutions are free but require technical setup.

Q: How long does the setup take?

Complete setup takes 25-30 minutes for first-time users. If you already have Python and Google Cloud configured, expect 8-10 minutes. Google Ads API approval adds 2-5 business days, but test environment works immediately for development.

Q: Can Claude make changes to my Google Ads campaigns?

With MCP and proper API permissions, Claude can execute campaign changes: bid adjustments, keyword additions, ad copy updates, budget changes. However, most setups use read-only access for safety, with Claude providing recommendations for manual implementation.

Q: What are API quotas and how do they affect usage?

Google Ads API limits operations to 10,000 per day (basic access) or 1,000,000 (approved accounts). Rate limits are 1,000 requests per 100 seconds. Exceeding quotas causes temporary suspensions. Most analysis workflows use 50-200 operations per query.

Q: Does this work with Google Ads Manager (MCC) accounts?

Yes. Use the login_customer_id parameter set to your MCC account ID, then specify individual customer_id in queries. This lets Claude access all client accounts under your manager account with a single authentication setup.

Q: How does this compare to fully automated platforms?

Claude + API integration provides analysis and recommendations but requires manual implementation. Fully automated platforms like Ryze AI execute optimizations 24/7 without human intervention. Choose Claude for learning and control, automated platforms for hands-off growth.

Related guides

Google Ads

15 Claude Skills for Google Ads Management

Advanced Claude prompts for keyword research, ad copy optimization, bidding strategies, and performance analysis.