Google Ads Search Term Analyzer with AI & Slack

What This Agent Does

This AI-powered Google Ads Search Terms Analyzer automatically monitors your Google Ads campaigns, extracts search term performance data, analyzes it with artificial intelligence, and delivers actionable insights directly to your Slack channel. The workflow identifies high-performing and underperforming search terms, provides strategic recommendations, and formats professional reports—all without manual intervention.

Key benefits and time savings:

• Eliminate manual reporting: Automatically generate daily/weekly SEM performance reports without spreadsheet work
• AI-powered insights: Leverage Google's Gemini AI to identify optimization opportunities in seconds
• Real-time alerts: Get instant Slack notifications with actionable recommendations
• Data-driven decisions: Make faster, more informed bidding and keyword strategy decisions
• Save 3-5 hours weekly: Reduce time spent analyzing search term reports and creating presentations

Target use cases:

• SEM managers monitoring multiple Google Ads accounts
• Marketing teams requiring regular performance dashboards
• Agencies managing client campaigns with reporting requirements
• E-commerce businesses optimizing search term performance
• B2B companies analyzing search intent and keyword relevance

Who Is It For

This workflow is ideal for marketing professionals, SEM specialists, and marketing operations teams who want to:

• Automate repetitive analysis tasks
• Access AI-generated insights without technical expertise
• Keep teams informed with consistent, professional reporting
• Improve campaign performance through data-driven optimization
• Scale reporting across multiple campaigns and accounts

You'll benefit most if you're already using Google Ads, Slack, and want to reduce manual analysis time while improving decision-making speed.

Required Integrations

Google Ads

Why it's needed: This integration connects TaskAGI directly to your Google Ads account, enabling the workflow to retrieve campaign data and search term performance metrics automatically.

Setup steps:

1. Access Google Cloud Console

   • Navigate to Google Cloud Console
   • Create a new project or select an existing one
   • Enable the Google Ads API in the APIs & Services section

2. Create OAuth 2.0 Credentials

   • Go to "Credentials" → "Create Credentials" → "OAuth 2.0 Client ID"
   • Select "Desktop application" as the application type
   • Download the credentials JSON file

3. Generate Refresh Token

   • Use Google's OAuth 2.0 playground or authentication flow to obtain a refresh token
   • You'll need your Google Ads customer ID (found in your Google Ads account settings)

4. Configure in TaskAGI

   • Navigate to Integrations → Google Ads
   • Enter your Client ID and Client Secret from the credentials file
   • Paste your Refresh Token
   • Input your Google Ads Customer ID (format: 1234567890)
   • Click Test Connection to verify setup
   • Save configuration

How to obtain API keys/credentials:

• Customer ID: Found in Google Ads account under "Account settings" (top-right menu)
• OAuth credentials: Generated through Google Cloud Console (see steps above)
• Refresh Token: Obtained through OAuth 2.0 authentication flow with Google Ads API scopes

Google Gemini

Why it's needed: This AI model analyzes search term data and generates strategic recommendations, transforming raw metrics into actionable insights for your team.

Setup steps:

1. Access Google AI Studio

   • Visit Google AI Studio
   • Sign in with your Google account

2. Create API Key

   • Click "Get API Key" → "Create API Key in new project"
   • Copy the generated API key

3. Configure in TaskAGI

   • Navigate to Integrations → Google Gemini
   • Paste your API Key
   • Select Model: gemini-2.5-flash (pre-configured in this workflow)
   • Set Temperature: 0.7 (balances creativity and consistency)
   • Click Test Connection
   • Save configuration

How to obtain API keys/credentials:

• Visit Google AI Studio API Keys page
• Create a new API key (no billing required for free tier)
• Keep the key secure and never share it publicly

Slack

Why it's needed: Slack integration delivers your AI-generated reports directly to your team's communication hub, ensuring stakeholders see insights immediately.

Setup steps:

1. Create Slack App

   • Go to Slack API Dashboard
   • Click "Create New App" → "From scratch"
   • Name it (e.g., "TaskAGI SEM Reports")
   • Select your workspace

2. Configure Bot Token Scopes

   • Navigate to OAuth & Permissions
   • Under "Scopes," add these Bot Token Scopes:
     • chat:write
     • chat:write.public
     • files:write

3. Install App to Workspace

   • Click "Install to Workspace"
   • Authorize the app
   • Copy the Bot User OAuth Token (starts with xoxb-)

4. Configure in TaskAGI

   • Navigate to Integrations → Slack
   • Paste your Bot Token
   • Enter your Workspace Name (from Slack URL)
   • Click Test Connection
   • Save configuration

5. Invite Bot to Channel

   • In Slack, go to your target channel
   • Type /invite @TaskAGI (or your app name)
   • The bot can now post messages there

How to obtain API keys/credentials:

• Bot Token: Generated in Slack API Dashboard under "OAuth & Permissions"
• Workspace Name: Found in your Slack workspace URL (e.g., mycompany.slack.com)

Configuration Steps

Step 1: Trigger Setup (Manual Trigger)

The workflow starts with a manual trigger, allowing you to run analysis on-demand or schedule it.

• Configuration: No parameters needed
• Usage: Click "Run" in TaskAGI to execute the workflow, or set up a schedule in your automation rules
• Benefit: Full control over when analysis runs

Step 2: Get Google Campaigns

This node retrieves all active campaigns from your Google Ads account.

• Configuration:
  • Account ID: Your Google Ads customer ID (e.g., 1234567890)
  • Status Filter: Set to ENABLED to retrieve only active campaigns
• Output: List of all campaigns with IDs, names, and status
• Data passed to next node: Campaign list for filtering

Step 3: Filter Brand Campaigns

This processing node filters campaigns to focus on specific ones (e.g., "Brand" campaigns only).

• Configuration:
  • Filter Logic: Add a condition like campaign_name contains "Brand"
  • Example: If your campaign is named "Brand - Desktop," it will be included
• Output: Filtered list of relevant campaigns
• Tip: Adjust the filter to match your campaign naming convention

Step 4: Check Campaign Found (Conditional)

This node checks if any campaigns match your filter criteria.

• Configuration: Automatic (no manual setup required)
• True Path: Proceeds to extract search term data
• False Path: Executes "No Campaign Found" (sends notification)
• Success indicator: Workflow continues if campaigns are found

Step 5: Extract Search Terms (HTTP Request)

This node queries Google Ads API for detailed search term performance data.

• Configuration:
  • Endpoint: Google Ads API search terms endpoint
  • Query: Pre-configured SQL query selecting search terms, impressions, clicks, and conversions
  • Date Range: Modify the BETWEEN clause to analyze specific periods (e.g., last 7 days)
• Example query modification:

  WHERE segments.date BETWEEN '2024-01-01' AND '2024-01-31'
  

• Output: Raw search term data with performance metrics

Step 6: Process Search Term Data

This function node transforms raw data into a structured format for AI analysis.

• Configuration:
  • Aggregation: Groups search terms by performance tier (high, medium, low)
  • Metrics included: Impressions, clicks, conversions, cost-per-conversion
• Output: Structured JSON object ready for AI analysis
• Example structure:

  {
    "high_performers": [...],
    "low_performers": [...],
    "analysis_period": "2024-01-01 to 2024-01-31"
  }
  

Step 7: Prepare AI Analysis Data (Edit Data)

This node formats data specifically for the Gemini AI prompt.

• Configuration:
  • Add context: Include campaign name, date range, and business goals
  • Format: Convert to readable text format for AI processing
• Output: Formatted data string passed to AI model

Step 8: AI Search Terms Analyzer (Google Gemini)

This is where AI magic happens—Gemini analyzes your search term data and generates insights.

• Configuration:
  • Model: gemini-2.5-flash (fast, efficient)
  • Prompt: Pre-configured to act as an SEM expert
  • Temperature: 0.7 (balanced analysis)
  • Max tokens: 2000 (sufficient for detailed recommendations)
• AI prompt includes:
  • Identification of high-performing keywords to bid up
  • Low-performing terms to pause or refine
  • Negative keyword recommendations
  • Budget optimization suggestions
• Output: Detailed AI-generated analysis and recommendations

Step 9: Parse AI Response

This node extracts structured insights from the AI's text response.

• Configuration: Automatic parsing of AI output
• Output: Organized recommendations ready for Slack formatting
• Benefit: Converts AI text into actionable data points

Step 10: Format Slack Blocks

This function creates visually appealing Slack message blocks with your insights.

• Configuration:
  • Include: Campaign name, analysis date, key metrics
  • Formatting: Uses Slack block kit for professional appearance
  • Sections: Summary, recommendations, and action items
• Output: Slack-formatted message ready to send
• Visual result: Professional report with headers, sections, and emphasis

Step 11: Send Slack Report

Final node delivers your analysis to your team.

• Configuration:
  • Channel: Select your target Slack channel (e.g., #marketing-reports)
  • Message type: Rich formatted message with blocks
  • Notification: Set to true for channel notification
• Output: Message posted to Slack
• Verification: Check Slack channel for formatted report

Step 12: No Campaign Found (Fallback)

If no campaigns match your filter, this node sends a notification.

• Configuration: Automatic
• Message: "No matching campaigns found. Please review filter settings."
• Purpose: Prevents silent failures and alerts you to configuration issues

Testing Your Agent

Test Execution Steps

1. Initial Setup Verification

• Ensure all three integrations (Google Ads, Gemini, Slack) show "Connected" status
• Verify your Google Ads account has at least one active campaign
• Confirm your Slack bot is invited to the target channel

2. Run Test Execution

• Navigate to your workflow in TaskAGI
• Click the "Run" button (or "Test" for test mode)
• Monitor the execution progress in real-time

3. Verify Each Node

• Get Google Campaigns: Check that campaigns are retrieved (should show count > 0)
• Filter Brand Campaigns: Verify filtered list matches your criteria
• Check Campaign Found: Confirm "true" path is taken
• Extract Search Terms: Validate that search term data is retrieved
• Process Search Term Data: Check structured data format in node output
• AI Search Terms Analyzer: Review AI-generated recommendations for quality and relevance
• Send Slack Report: Confirm message appears in your Slack channel

Expected Results and Success Indicators

Successful execution shows:

• ✅ All nodes complete without errors (green checkmarks)
• ✅ Campaign data retrieved and filtered correctly
• ✅ Search term data populated with metrics (impressions, clicks, conversions)
• ✅ AI analysis includes specific recommendations (e.g., "Pause 'misspelled variant' - 0 conversions")
• ✅ Slack message displays with professional formatting, headers, and bullet points
• ✅ Execution time: typically 30-60 seconds for complete workflow

Common issues and solutions:

• "No campaigns found": Verify filter matches your campaign naming convention
• Empty search term data: Ensure campaigns have sufficient traffic and historical data
• Slack message fails: Confirm bot is invited to channel and has chat:write permission
• AI response is generic: Add more specific context in the "Prepare AI Analysis Data" node

Next steps after successful test:

1. Schedule the workflow to run daily or weekly
2. Adjust filters and AI prompts based on your specific needs
3. Share the Slack channel with stakeholders
4. Monitor recommendations and implement top suggestions
5. Refine the workflow based on team feedback

Your AI-powered SEM analysis is now live and delivering insights to your team! 🚀