Automated Customer Feedback Sentiment Analysis

Client and Business Context

The project was implemented for a medium-sized company in the meal kit delivery service industry. The company was growing rapidly but lacked an automated way to track and respond to the increasing volume of online customer reviews.

Challenge: Efficient Monitoring and Response to Feedback

The main challenge was to create a system for continuous monitoring of online reviews appearing across various channels (social media, review portals, website forms). Key objectives included quickly identifying issues (e.g., regarding meal quality, delivery timeliness, customer service), understanding overall customer satisfaction, and leveraging positive feedback. Manual tracking of reviews was time-consuming and did not allow for rapid response.

• Need to aggregate reviews from multiple online sources.
• Requirement for automatic sentiment classification (positive, negative, neutral).
• Need for prompt notification of relevant departments about issues or praise.
• Desire to use data for service quality improvement and marketing efforts.
• Limited human resources for manual monitoring.

Solution: Automation with Airtable, Make.com, and AI

An automated system based on the Make.com platform and the Airtable database was implemented. The system aggregates reviews from selected sources, performs sentiment analysis using OpenAI models from the GPT-4 family (called via API), and then classifies and saves the results in Airtable. Depending on the detected sentiment, the system automatically initiates appropriate actions (e.g., alerts, task creation). The solution was also successfully tested with other language models like Gemini and Claude.

Applied Technologies:

• Make.com (formerly Integromat): The automation platform acting as the operation's "brain." Responsible for:
  • Integration with data sources (social media APIs, Google Reviews, forms).
  • Process orchestration: fetching reviews, sending for analysis, saving results, triggering actions.
  • Invoking sentiment analysis: Passing review text to the OpenAI API (GPT-4).
  • Conditional logic defining actions based on the analysis result.
• Airtable: A flexible cloud-based database platform serving as:
  • Central review database (content, source, date, author, sentiment, keywords, status).
  • Interface for teams to manage and analyze data.
  • Dashboard for visualizing sentiment trends.
• OpenAI Models (GPT-4) and a Well-Crafted Prompt: Sentiment analysis is performed by calling the GPT-4 API with a precisely constructed prompt that instructs the model on how to classify the emotional tone of the text (positive, negative, neutral) and identify key topics or words within the review.

Data Flow Process:

1. Source Monitoring: Make.com scenarios periodically check configured sources (social media, reviews, forms).
2. Data Retrieval: Make.com fetches the content of new reviews and metadata.
3. Sentiment Analysis: Make.com sends the review text along with the appropriate prompt to the OpenAI API (GPT-4).
4. Result Reception: The OpenAI API returns the result (e.g., 'Negative', 'Positive', 'Neutral') and possibly identified keywords/topics.
5. Saving to Airtable: Make.com creates a new record in the "Customer Reviews" base with all the data.
6. Triggering Conditional Actions: Based on the sentiment, Make.com initiates further actions.

Response Strategy Based on Detected Sentiment:

• Negative Sentiment:
  • Immediate Alert: Make.com sends a real-time notification (e.g., to a dedicated Slack channel #alert_negative_reviews or email) to the customer service team and/or quality manager. The notification includes the review content, source, and link.
  • Automatic Task Creation: Make.com can automatically create a task in a project management system (e.g., Trello, Asana, Jira - if integrated) or in a dedicated "Actions Needed" table in Airtable. The task is assigned to the appropriate person/team with high priority.
  • Marking in Airtable: The record in Airtable is marked as requiring urgent intervention.
• Positive Sentiment:
  • Marketing Notification: Make.com sends a notification (e.g., to Slack channel #customer_praise or email) to the marketing team. This can serve as inspiration for content creation, social proof, or identifying brand ambassadors.
  • Marking in Airtable: The record is marked as 'Positive,' facilitating filtering and searching for potential testimonials.
  • Optionally: Flagging for the customer service team to send a thank-you note.
• Neutral Sentiment:
  • Archiving and Monitoring: The review is saved in Airtable without triggering immediate alerts. It primarily serves for trend analysis and understanding context (e.g., frequently asked questions, suggestions).
  • Identifying Questions/Suggestions: If the analysis (or additional logic in Make.com) detects a question or suggestion within a neutral review, it can be flagged for review by the appropriate team (e.g., customer service, product development).

Key Benefits for the Client

• Speed of Response: Instantly identifying and reacting to negative reviews, minimizing potential brand crises.
• Operational Efficiency: Automation of the time-consuming process of monitoring and classifying reviews, freeing up employee time.
• Better Customer Understanding: Centralization of data in Airtable allows for easy analysis of trends, identification of recurring issues (e.g., with a specific dish, courier), and strengths of the offering.
• Marketing Support: Easy access to positive reviews for use in marketing communications.
• Quality Improvement: Systematic feedback collection supports the continuous improvement of products and services.
• Scalability: The solution can be easily expanded with new data sources or more advanced analytical functions.