AI-Powered Customer Feedback Analysis and HTML Report Generator - n8n Workflow

• Marketing Managers and Analysts seeking automated social listening solutions.


• Data scientists needing structured extraction and categorization of qualitative feedback.


• Businesses that frequently analyze large volumes of customer comments or survey responses.


• Users looking for advanced examples of complex n8n templates utilizing Langchain agents.

Analyzing large volumes of unstructured customer feedback can be time-consuming and inconsistent. This sophisticated n8n workflow solves this by creating a structured, multi-step AI pipeline to categorize, analyze, summarize, and report on the data automatically. Starting with raw input from a Google Sheets n8n node, the process leverages several specialized OpenAI-powered agents to propose topics, unify analysis, refine proposals, and generate a comprehensive final report.

This high-value n8n workflow showcases the power of integrating specialized LLM agents within a single automation, turning raw data into an actionable, HTML-formatted report that is instantly delivered to stakeholders using the Gmail n8n node. It is a prime example of expert-level n8n development.

The process begins with the 'When clicking ‘Test workflow’' n8n trigger, designed for manual execution or testing.

1. Data Ingestion: The Google Sheets n8n node reads the customer feedback data. Initial Filter and Code n8n nodes prepare a sample set for the first analytical pass.


2. Initial Analysis: The analysis topics proposal agent, connected to an OpenAI Chat Model, suggests the initial classification topics. This process is structured using a Structured Output Parser n8n node.


3. Unified Analysis: The Merge original table and the prompts node combines the initial topics with the full dataset. The Unified AI Agent for analysis then performs the main classification and sentiment assessment.


4. Data Transformation & Refinement: The Transform results into columns and All unique elements merge code n8n nodes restructure the AI output. The subsequent proposal refinement agent standardizes the categorized results.


5. Iterative Processing: A Second iteration of analysis agent provides further detail and refinement, followed by a Code n8n node that aggregates all results into a single JSON structure.


6. Report Generation: The Summarization of the unalysed results agent creates executive summaries. The Final report editor and Revisor and HTML formating agent then compile the final analysis and apply professional HTML formatting.


7. Delivery: The data is prepared by an Edit Fields n8n node, and the complete HTML report is automatically sent to the relevant recipient via the Gmail n8n node, completing the entire n8n workflow.

To deploy this specialized n8n workflow, follow these steps:

1. Import: Copy the provided JSON data and import it directly into your n8n instance via the 'Workflows' menu using the 'New' -> 'Import from JSON' option.


2. Google Sheets Credentials: Configure the Google Sheets n8n node with credentials authorized to read the specific spreadsheet containing your customer feedback data.


3. OpenAI/Langchain Credentials: Configure the required credentials for the Langchain n8n node agents. This typically involves setting up your OpenAI API key as an active credential.


4. Gmail Setup: Configure the final Gmail n8n node with your sending credentials and update the recipient email address in the node's parameters.


5. Activate: Ensure the n8n workflow is active. You can use the manual 'When clicking ‘Test workflow’' n8n trigger for initial testing and validation.

When clicking ‘Test workflow’ (Manual Trigger n8n trigger): The starting point of this automation. It allows for easy testing before scheduling or integration.
Google Sheets: Reads the customer feedback data from a specified spreadsheet and sheet. Essential for feeding the raw data into the n8n workflow.
analysis topics proposal (Langchain Agent n8n node): The first layer of AI analysis. It uses an OpenAI Chat Model to generate potential topics and categories found in the initial data sample.
Unified AI Agent for analysis (Langchain Agent n8n node): The primary processing unit. It applies the analysis framework (topics, sentiment) across the entire dataset, powered by an associated OpenAI Chat Model.
Code n8n nodes (Transform results into columns, All unique elements merge): Crucial custom scripting steps used to structure and normalize the complex JSON output generated by the Langchain agents into a clean, tabular format suitable for subsequent steps.
Revisor and HTML formating agent (Langchain Agent n8n node): Takes the finalized summary and analysis, ensures clarity, and wraps the entire report into valid HTML code for professional delivery.

• Gmail (Email n8n node): The final action. Sends the HTML-formatted report generated by the n8n workflow to the specified recipient, completing the feedback loop.