AI Trustpilot Review Analyzer & Sentiment Extractor with DeepSeek & OpenAI

Unlock Customer Insights with this AI-Driven Review Analysis Agent

This AI Agent acts as your dedicated digital assistant for understanding customer feedback. It systematically scrapes Trustpilot reviews for a company you specify, then employs two distinct AI models for comprehensive analysis. First, it uses DeepSeek's advanced language understanding (via a Langchain compatible node) to intelligently extract key information from the raw HTML of each review, such as the author's name, rating, date, review title, full text, the reviewer's country, and their total review count.

Once the structured data is extracted, the Agent passes the review text to OpenAI for sophisticated sentiment analysis. OpenAI categorizes each review as 'Positive', 'Neutral', or 'Negative', providing a clear emotional signal.

All this valuable, enriched data – the extracted details and the AI-generated sentiment – is then automatically and neatly organized into a Google Sheet. The Agent also includes a smart check to prevent processing and storing duplicate reviews, ensuring your dataset remains clean and efficient. This AI-driven automation empowers you to monitor customer sentiment, track feedback trends, and gather actionable insights with minimal manual effort.

Key Features & Benefits

• Automated Trustpilot Scraping: Fetches recent reviews from any specified company page on Trustpilot, saving significant manual data collection time.
• AI-Powered Data Extraction (DeepSeek): Leverages DeepSeek for precise extraction of structured data (author, rating, date, title, text, country, review count) directly from review HTML.
• Advanced Sentiment Analysis (OpenAI): Utilizes OpenAI's powerful models to accurately determine the sentiment (Positive, Neutral, Negative) of each customer review.
• Centralized Data Repository: Consolidates all scraped reviews and their sentiment analysis into a Google Sheet, making it easy to access, filter, and analyze customer feedback.
• Duplicate Prevention: Intelligently checks against existing entries in Google Sheets to avoid redundant data, ensuring data integrity.
• Configurable & Scalable: Easily customize the target company and the number of review pages to scrape based on your needs.
• AI Agent Abilities: This agent demonstrates AI abilities in 'customer review analysis', 'natural language processing', 'AI-driven data extraction', and 'automated sentiment detection'.

• An n8n instance (Cloud or self-hosted).
• DeepSeek API Key. You'll need to configure this in the 'DeepSeek Chat Model' node, ensuring the 'Base URL' under options is set to https://api.deepseek.com/v1.
• OpenAI API Key with access to a suitable model for sentiment analysis (e.g., gpt-3.5-turbo).
• Google Sheets API credentials configured in n8n.
• A Google Sheet prepared with the following specific column headers in the first row: Id, Data, Nome, Titolo, Testo, Località, N. Recensioni, URL, Valutazione, Sentiment.

1. Download the n8n workflow JSON file.
2. Import the workflow into your n8n instance.
3. Configure the 'Set Parameters' node: update the company_id value with the target company's identifier from Trustpilot (e.g., 'mycompany.com') and set the max_page value for the number of review pages to scrape.
4. Configure the 'DeepSeek Chat Model' node: select or create your DeepSeek API credential. Under 'Options', ensure the 'Base URL' is set to https://api.deepseek.com/v1.
5. Configure the 'OpenAI Chat Model' node: select or create your OpenAI API credential.
6. Configure the 'Get rows', 'Get Google Sheets', and 'Update sheet' nodes: select your Google Sheets API credential. Update the 'Document ID' to point to your Google Sheet. Ensure the 'Sheet Name' (or 'Sheet ID' like 'gid=0') is correct for the sheet containing your review data.
7. Verify your Google Sheet has the exact column headers as specified in the Prerequisites section. The 'Id' column is used to store a unique identifier for each review (derived from the review URL) to prevent duplicates.
8. (Optional) Review and customize the attributes in the 'Information Extractor' node if you need to extract different or additional data points. Ensure these match your Google Sheet columns if changed.
9. Activate the workflow. You can test it by clicking 'Test workflow' on the 'When clicking ‘Test workflow’' node.