Keyword Research with AI Agents and Google Search Console

Description and use

An automated workflow that pulls Google Search Console data, clusters keywords by page and intent, detects cannibalization, and outputs JSON for content and SEO decisions.

Problem

Manual keyword research is slow and disconnected from real user behavior, relying on guesswork instead of actual search query data.

Outcome

The workflow automates keyword clustering and intent analysis, turning raw GSC data into actionable insights that drive content and SEO decisions.

Workflow

• Set up Google Cloud project with Search Console API enabled and configure Application Default Credentials.
• Run Node.js script to authenticate, fetch up to 25,000 rows of GSC query data for the last 12 months.
• Map query results to canonical routes, aggregate metrics per page, and group queries by intent (informational, commercial, etc.).
• Detect cannibalization by flagging pages competing for similar queries.
• Generate JSON mapping of pages to keyword clusters, including metrics and intent labels.
• Use the JSON to prioritize optimization targets (high impressions, low clicks) and inform internal linking strategy.

Evidence

Manual keyword research is slow and disconnected from real user behavior.

Source

The agent runs a Node.js script that authenticates with Google Cloud, fetches 12 months of search query data, and processes it into keyword clusters organized by page and intent.

Source