Automate SEO Keyword Research With AI Agents: Topical Map Expansion

AI-Powered Topic and Keyword Research Automation
📌 The presented system uses AI agents built with NADN to automate topic and keyword research, running processes like finding categories and feeding data into an agent.
🤖 The workflow includes a Topic Research Agent that generates topics based on a Content Configuration Document defining the website's focus, intent, and context.
🔄 A Topic Revision Agent handles updates, avoids keyword cannibalization by checking existing records and performing Google site searches, and corrects records flagged for review.

Topical Map Structure and Execution
📌 Content strategy is defined using a Topical Map categorized as "Off," "Core Section" (monetizable pages), or "Outer Section" (informational content/blog).
⏱️ The main agent is set to run on a schedule (e.g., every hour) and processes one topic per active category sequentially within its loop.
📊 The process incorporates existing records to exclude already covered topics and uses a list of defined Content Types (crystallized search intent templates) to ensure proper page structure.

Topic Validation and Modern SEO Approach
🔎 Validation requires a human in the loop to check live Google results (using tools like Ahrefs) for search volume and intent alignment before promotion to "Production."
↔️ The new topic-focused approach prioritizes identifying the correct root topic/page first, reversing the old keyword-focused method of clustering keywords into pages.
✅ Successful validation hinges on matching the search intent and content type (e.g., an "Ultimate Guide" vs. a nuanced subtopic) to the validated search demand.

Automation Workflow and Future Scope
⚙️ Once validated, topics moved to "Production" trigger downstream automations to write content, upload to Google Docs for review, publish to WordPress, and even create featured images.
⚠️ The revision agent manages cannibalization conflicts by flagging records for "cannibal review" or marking them for "do not make," ensuring records of reviewed keywords are kept.
🚀 The speaker plans to expand automation to cover content updates and internal link building to streamline the entire SEO production cycle.

Key Points & Insights
➡️ Utilize an AI agent system integrated with tools like Airtable and Google Docs to run topic research 24/7, moving away from manual keyword-first processes.
➡️ Define content types clearly as search intent templates (e.g., guide, how-to) to improve indexing efficiency and lower the cost of retrieval for search engines.
➡️ Validate every AI-generated idea against live search results to confirm search demand and match the required content format/angle, even when using AI ideation.
➡️ The revision agent's cannibalization checks are crucial; if a topic conflicts, it annotates the main topic instead of deleting the derivative idea, keeping it logged under "cannibal review."

📸 Video summarized with SummaryTube.com on Dec 22, 2025, 18:49 UTC