YouTube Comment Scraper: Analyze Viewer Sentiment & Feedback

If you have a viral video, you might get 5,000+ comments.

Reading them all is impossible. But ignoring them means missing out on:

1. Video Ideas: "Can you do a part 2 on X?"
2. Bug Reports: "The audio is out of sync at 2:30."
3. Sentiment: "I usually love your stuff, but this one felt rushed."

In this guide, we'll build a YouTube Comment Analyzer using SociaVault. We'll scrape the comments from a video and automatically categorize them.

The Strategy

We will write a script that:

1. Takes a YouTube video URL.
2. Scrapes the top 100 comments.
3. Filters for Questions (potential video topics).
4. Filters for Sentiment (positive vs negative).

Prerequisites

You'll need a SociaVault API key. You can get one here.

Step 1: Scrape the Comments

We'll use the /scrape/youtube/video/comments endpoint.

const axios = require('axios');

const API_KEY = 'YOUR_SOCIAVAULT_API_KEY';
const BASE_URL = 'https://api.sociavault.com/v1';

async function getVideoComments(videoUrl) {
  console.log(`📥 Fetching comments for: ${videoUrl}`);

  try {
    const response = await axios.get(`${BASE_URL}/scrape/youtube/video/comments`, {
      params: { 
        url: videoUrl,
        order: 'top' // Get top comments first
      },
      headers: { 'x-api-key': API_KEY }
    });

    if (response.data.success) {
      return response.data.data;
    }
  } catch (error) {
    console.error("Error fetching comments:", error.message);
    return [];
  }
}

Step 2: Find Video Ideas (Questions)

The most valuable comments are questions. They literally tell you what to make next.

function findQuestions(comments) {
  const questions = comments.filter(c => c.text.includes('?'));
  
  console.log(`\n❓ Found ${questions.length} Questions:`);
  questions.slice(0, 5).forEach(q => {
    console.log(`   - "${q.text}" (by ${q.author})`);
  });
}

Step 3: Analyze Sentiment

Let's do a quick "Vibe Check" on the comments.

function analyzeSentiment(comments) {
  let positive = 0;
  let negative = 0;

  comments.forEach(c => {
    const text = c.text.toLowerCase();
    if (text.includes('love') || text.includes('great') || text.includes('awesome')) positive++;
    if (text.includes('hate') || text.includes('bad') || text.includes('boring')) negative++;
  });

  console.log(`\n📊 Sentiment Analysis:`);
  console.log(`   👍 Positive: ${positive}`);
  console.log(`   👎 Negative: ${negative}`);
  
  const ratio = positive / (positive + negative || 1);
  console.log(`   📈 Positivity Ratio: ${(ratio * 100).toFixed(1)}%`);
}

Putting It All Together

async function analyzeVideo(url) {
  const comments = await getVideoComments(url);
  
  if (comments.length > 0) {
    console.log(`✅ Successfully scraped ${comments.length} comments.`);
    findQuestions(comments);
    analyzeSentiment(comments);
  }
}

// Example: Analyze a MrBeast video
analyzeVideo('https://www.youtube.com/watch?v=0e3GPea1Tyg');

Sample Output

📥 Fetching comments for: https://www.youtube.com/watch?v=0e3GPea1Tyg
✅ Successfully scraped 85 comments.

❓ Found 12 Questions:
   - "When is the next challenge coming out?" (by User123)
   - "How much did this cost to make?" (by FanBoy)
   - "Can you do a video with PewDiePie?" (by GamerX)

📊 Sentiment Analysis:
   👍 Positive: 45
   👎 Negative: 3
   📈 Positivity Ratio: 93.8%

Why This Matters

If you are a brand or creator, your comments section is a focus group that runs 24/7.

By automating this analysis, you can:

1. Reply Faster: Filter for questions and answer them immediately to boost engagement.
2. Spot Issues: If the "Negative" count spikes, investigate immediately.
3. Plan Content: Use the questions to build your content calendar for next month.

Get Started

Stop guessing what your viewers want. Get your API key here.