n8n is a workflow automation platform that connects APIs, data processing steps, and delivery destinations without requiring custom server code. For YouTube competitor research, it connects the YouTube Data API, a transcript extraction step, an AI analysis node, and an email or Google Sheets delivery step into a single workflow that runs on a Monday morning schedule. Here is how each component works.

The Schedule Trigger

The workflow starts with a Schedule Trigger node configured for Monday at 7 AM. This fires the rest of the workflow automatically every week. No manual action is required after the initial setup. The cron expression for "every Monday at 7 AM" is 0 7 * * 1. The n8n Schedule Trigger node accepts this format directly.

The Channel Loop

After the trigger, a Code node or a Split in Batches node loops through a JSON array of competitor channel IDs. Each channel ID feeds into the next step sequentially. Adding or removing a channel from monitoring requires editing one node in the workflow — no code changes, no redeployment.

The YouTube API Calls

Two HTTP Request nodes handle the YouTube Data API calls. The first calls the search.list endpoint with the channel ID and a publishedAfter parameter set to 7 days ago, returning all videos published in the past week. The second calls the captions.list endpoint for each video ID returned, checking for available transcript tracks and downloading the caption text.

The full workflow is demonstrated in the YouTube Competitor Intelligence Monitor — showing exactly how the channel loop, API calls, filtering, AI analysis, and CSV output connect in sequence.

The AI Analysis Node

n8n's OpenAI node or HTTP Request node sends each transcript to the AI with a structured prompt. The prompt asks for a JSON response with specific fields: primary_topic, messaging_angle, audience_target, key_claims, and gap_flag. Requesting JSON output directly from the AI simplifies the downstream data processing — the output can be mapped directly to CSV columns without additional parsing.

The Output Node

The final node compiles all analyzed video records and sends them as a CSV attachment via email using n8n's Send Email node. Alternatively, a Google Sheets node appends each record as a new row for historical accumulation. Both options take under 10 minutes to configure once the upstream analysis is working.