Building an Automated Customer Sentiment Analysis System with n8n and Google Sheets: Real-time Data Updates and Visualization

Tired of manually analyzing countless customer feedback every day? By combining n8n and Google Sheets to build an automated system that analyzes and visualizes customer sentiment in real-time, you can save time and money, and make data-driven decisions. This article provides a detailed guide on how to build it and its practical applications.

1. The Challenge / Context

Today, businesses must process an enormous amount of data generated from customer interactions. This data, collected through various channels such as social media comments, customer support tickets, and product reviews, contains crucial information for understanding customer needs, preferences, and dissatisfactions. However, manually analyzing such data is time-consuming, prone to errors, and makes it difficult to quickly grasp real-time changes in customer sentiment. By building an automated sentiment analysis system, these problems can be solved, and it can be utilized to improve customer experience and promote business growth.

2. Deep Dive: n8n

n8n is a node-based workflow automation platform. Even users with limited coding experience can connect various applications and services and build automated workflows through its drag-and-drop interface. n8n has the advantage of being able to operate in a self-hosting manner, which enhances control over data security and privacy. Furthermore, it supports integration with various APIs through its extensive library, and custom nodes can be developed to build workflows tailored to specific requirements.

3. Step-by-Step Guide / Implementation

Now, let's look at how to build an automated customer sentiment analysis system using n8n and Google Sheets, step by step. This system operates with the following flow:

  1. Collect customer feedback from data sources. (e.g., Twitter, Google Sheets, Customer Support System API)
  2. Analyze the sentiment of the feedback using a sentiment analysis API. (e.g., Google Cloud Natural Language API, Hugging Face API)
  3. Save the analysis results to Google Sheets.
  4. Create a dashboard based on Google Sheets data to visualize sentiment changes in real-time.

Step 1: n8n Installation and Setup

First, you need to install n8n. n8n can be installed in various ways, including Docker, npm, and Cloudron. Here, we will explain how to install it using Docker.

# Run n8n using Docker
docker run -it --rm -p 5678:5678 -v ~/.n8n:/home/node/.n8n n8nio/n8n

Executing the command above will run n8n on port 5678 of your local environment. You can access the n8n interface by connecting to http://localhost:5678 via a web browser. To permanently store n8n data, you need to set up a Docker Volume. The -v ~/.n8n:/home/node/.n8n part in the command above is for setting up the Docker Volume. With this setting, n8n data will be stored in the .n8n folder of the user's home directory.

Step 2: Google Sheets API Setup

To access Google Sheets from n8n, you need to enable the Google Sheets API and create a service account key.

  1. Access Google Cloud Console and create a new project or select an existing one.
  2. In the "APIs & Services" menu, click "Enable APIs and Services".
  3. Search for "Google Sheets API" and enable it.
  4. In the "Credentials" menu, click "Create Credentials" and select "Service Account".
  5. Enter a service account name and description, then click "Create".
  6. Set the service account permissions (role). Granting the "Editor" role is common.
  7. Click "Create Key" and download the key file in JSON format. This file will be used by n8n to access Google Sheets.
  8. You need to share the Google Sheet and grant edit permissions. Share the Google Sheet with the service account's email address (which can be found in the key file) and grant edit permissions.

Step 3: Building the n8n Workflow

Now, build the workflow in the n8n interface. Below is an example of a workflow that searches for specific keywords on Twitter, analyzes the sentiment of the search results, and then saves them to Google Sheets.

  1. Twitter Trigger: Add a node to search for specific keywords using the Twitter API. To access the Twitter API, you need to create a Twitter Developer account and obtain an API key.
  2. Function: Add a JavaScript Function node to extract only the necessary text information from the data collected from Twitter.
    3. // Text extraction example
items.forEach(item => {
  item.json.text = item.json.full_text || item.json.text; // Handle Retweet
  item.text = item.json.text;
});
return items;
  3. HTTP Request: Add an HTTP Request node to send text data to a sentiment analysis API (e.g., Hugging Face API) and receive the sentiment analysis results. To use the Hugging Face API, you need to create a Hugging Face account and obtain an API key.
    4. // Hugging Face API request settings example (POST method)
{
  "url": "https://api-inference.huggingface.co/models/nlptown/bert-base-multilingual-uncased-sentiment",
  "method": "POST",
  "headers": {
    "Authorization": "Bearer YOUR_HUGGING_FACE_API_KEY"
  },
  "body": {
    "inputs": "{{$json.text}}"
  },
  "json": true
}
  4. Function: Add a JavaScript Function node to convert the JSON data returned from the sentiment analysis API into the required format.
    5. // Sentiment analysis result conversion example
items.forEach(item => {
  let result = item.json[0];
  let sentiment = "Neutral";
  let score = 0;

  result.forEach(s => {
    if (s.score > score) {
      score = s.score;
      sentiment = s.label;
    }
  });

  item.sentiment = sentiment;
  item.score = score;
});

return items;
  5. Google Sheets: Add a Google Sheets node to save the original text data along with the sentiment analysis results to Google Sheets. In this node, use the service account key file created earlier to access Google Sheets.
    6. // Google Sheets node settings example
{
  "authentication": "serviceAccount",
  "serviceAccountAuthType": "Key File",
  "keyFileData": "JSON formatted service account key file content",
  "spreadsheetId": "Google Sheets Spreadsheet ID",
  "range": "Sheet1!A1",
  "operation": "append",
  "valueInputOption": "USER_ENTERED",
  "includeValuesInResponse": false,
  "rawData": [
    [
      "{{$json.text}}",
      "{{$json.sentiment}}",
      "{{$json.score}}"
    ]
  ]
}

After completing the settings for each node, activating the workflow will automatically search for specific keywords on Twitter, analyze the sentiment of the search results, and save them to Google Sheets.

Step 4: Google Sheets Data Visualization

Based on the sentiment analysis data stored in Google Sheets, you can create charts and graphs to visualize sentiment changes in real-time. You can use Google Sheets' built-in charting features or build more complex and interactive dashboards using Google Data Studio.

4. Real-world Use Case / Example

Customer Service Improvement: A startup analyzed the sentiment of product mentions on social media in real-time after a new product launch to quickly identify and respond to customer complaints. Through an automated system built by integrating n8n, Hugging Face API, and Google Sheets, the customer service team was able to respond immediately when issues arose, thereby improving customer satisfaction. Previously, a lot of time was spent manually monitoring social media, but after building the automated system, time was saved and efficiency was increased.

5. Pros & Cons / Critical Analysis

  • Pros:
    • Automation: Saves time and effort required for manual sentiment analysis.
    • Real-time Analysis: Allows for understanding real-time changes in customer sentiment and responding quickly.
    • Data-driven Decision Making: Enables making data-driven decisions based on sentiment analysis results.
    • Flexibility and Scalability: n8n can integrate with various APIs and services, and custom nodes can be developed to build workflows tailored to specific requirements.
  • Cons:
    • Initial Setup Complexity: Initial setup, including n8n installation and configuration, API key issuance, and workflow building, can be time-consuming.
    • API Usage Costs: Using sentiment analysis APIs may incur API usage costs. While free APIs exist, they may have usage limitations.
    • Accuracy Issues: The accuracy of sentiment analysis APIs can vary depending on the quality of text data, language, and context. Accuracy may be lower, especially for sarcastic or metaphorical expressions.

6. FAQ

  • Q: Can I use other automation tools instead of n8n?
    A: Yes, various automation tools such as Zapier, IFTTT, and Make (formerly Integromat) can be used. However, n8n offers the advantage of self-hosting and provides more powerful features and flexibility.
  • Q: What should I consider when choosing