Brand Content Extraction, Summarization, and Structured Sentiment Analysis - n8n Workflow

This n8n template is ideal for:
Marketing Analysts: Who need quick summaries and sentiment scores for competitor product pages or brand mentions.
Data Engineers: Looking for robust web scraping solutions combined with LLM processing within a single n8n workflow.
AI Developers: Interested in seeing how the n8n LangChain nodes can enforce structured output for sentiment analysis.
Automation Specialists: Seeking advanced, multi-step n8n solutions utilizing premium scraping tools like Bright Data.

Analyzing large volumes of brand content scattered across the web is challenging. This powerful n8n workflow solves this by seamlessly integrating Bright Data's Web Unlocker for efficient, anti-blocking web scraping, followed by sophisticated analysis using Google Gemini via the n8n AI nodes. This comprehensive n8n solution ensures that raw, messy web data is transformed into three actionable outputs: a concise summary, clean textual data, and a structured JSON array detailing sentiment scores. This n8n workflow provides immediate, high-value insights, saving hours of manual data processing and cleaning.

This automation begins with a manual n8n trigger, but it can easily be adapted into an API endpoint or scheduled n8n trigger for production use.

1. Trigger and Setup: The n8n workflow starts with the 'When clicking ‘Test workflow’' n8n trigger. A subsequent 'Set' n8n node defines the target URL for extraction (e.g., an Amazon product page) and the Bright Data zone ('webunlocker1').


2. Web Extraction: The 'Perform Bright Data Web Request' n8n node scrapes the defined URL using Bright Data, requesting the output content in clean Markdown format.


3. Parallel AI Processing: The scraped content is split into three simultaneous processing paths, all utilizing the Google Gemini Flash Exp model via various n8n LangChain nodes:

1. Output and Archival: Each processed output (Summary, Textual Data, Sentiment Analysis) is archived to a local disk using n8n file nodes and simultaneously sent to a third-party webhook for immediate notification, completing the data flow of the n8n workflow.

To deploy this comprehensive n8n workflow template, follow these steps:

1. Import: Copy the provided JSON and paste it directly into your n8n instance using the 'New' menu and selecting 'Import from JSON'. This creates the full n8n workflow.


2. Credentials - Bright Data: Locate the 'Perform Bright Data Web Request' n8n node. You must set up an 'HTTP Header Auth' credential using your Bright Data account details (API key).


3. Credentials - Google Gemini: Ensure you have configured a Google Gemini (PaLM) API credential. This credential must be linked to the three Google Gemini Chat Model n8n nodes used in the workflow.


4. Configuration: In the 'Set URL and Bright Data Zone' n8n node, update the url value to the specific webpage you wish to scrape.


5. Webhook Update: In all 'Initiate a Webhook Notification' n8n nodes, replace the placeholder URL with your actual desired endpoint (e.g., a Slack incoming webhook or an internal logging service).


6. File Paths: If running this n8n workflow on a self-hosted or Docker instance, adjust the file paths (d:\...) in the 'Write to disk' n8n nodes to a location accessible by your n8n container or server.

This n8n workflow utilizes multiple specialized n8n node types to achieve sophisticated scraping and AI processing:

Manual Trigger n8n Trigger: Starts the n8n workflow upon manual execution, useful for testing.
Set n8n Node: Used to define dynamic data, specifically the target url and the zone for Bright Data, making the n8n workflow easily configurable.
HTTP Request n8n Node (Perform Bright Data Web Request): This is the core scraping n8n node. It POSTs the request to the Bright Data API, utilizing the 'web_unlocker1' zone and requesting the output in markdown format for structured AI input.
@n8n/n8n-nodes-langchain.lmChatGoogleGemini n8n Node: This n8n node is configured multiple times using the models/gemini-2.0-flash-exp model, providing the backbone for all AI operations.
@n8n/n8n-nodes-langchain.chainSummarization n8n Node (Summarize Content): Uses the Gemini LLM to generate a concise summary of the scraped brand content.
@n8n/n8n-nodes-langchain.chainLlm n8n Node (Markdown to Textual Data Extractor): A specialized AI n8n node that cleans the Markdown input, ensuring only pure textual data remains for further analysis.
@n8n/n8n-nodes-langchain.informationExtractor n8n Node (AI Sentiment Analyzer): This powerful n8n node uses a defined JSON schema to force the LLM to provide a structured response for sentiment analysis (Positive, Neutral, Negative, Confidence Score).
Function n8n Node: Used to convert the final JSON output into binary data, preparing the data for storage via the Read/Write File n8n nodes.

• Read/Write File n8n Node: Archives the three key outputs (Summary, Textual Data, Sentiment Analysis JSON) to persistent storage.