YouTube Audience & Competitor Intelligence

Turn public YouTube comments into structured audience intelligence using the official YouTube Data API v3. No browser automation, private YouTube endpoints, login automation, CAPTCHA solving, or proxy rotation is used.

What it does

For each public video, the Actor returns comment-level intelligence including sentiment, complaint categories, complaint score, audience intent, feature-request signals, questions, purchase intent, competitor mentions, and switching signals. It also emits a free per-video summary and supports monitoring newly observed comments across scheduled runs.

Why this Actor exists

Raw YouTube comment scrapers already exist. This Actor focuses on the commercial layer above raw extraction: product feedback, customer pain points, competitive switching, feature demand, and audience questions.

Source and stability

The Actor uses Google's official YouTube Data API v3 only. commentThreads.list, comments.list, and videos.list are documented public endpoints. A user-supplied API key is required and is marked isSecret: true in the Apify input schema.

Typical public-read quota cost is 1 unit per list request. Your Google Cloud project's YouTube quota is separate from Apify billing.

Inputs

• youtubeApiKey: encrypted YouTube Data API v3 key.
• videoUrls: YouTube URLs or 11-character video IDs.
• maxCommentsPerVideo: result cap.
• order: newest (time) or relevance.
• searchTerms: optional API-side comment text filter.
• competitorNames: optional names to detect.
• includeReplies: fetch replies; may consume more quota.
• monitorChanges / onlyNewComments: incremental monitoring.
• maxConcurrency: bounded video concurrency (default 2, maximum 5).
• request, timeout, and retry caps prevent runaway runs.

Output

Dataset records are one of:

• comment — useful comment intelligence row.
• video_summary — free aggregate summary for that video.
• diagnostic — free actionable error/limit row.

Public author display names are excluded by default. Enable includeAuthorName only when needed.

Comment rows keep video-level likeCount and expose the comment's own likes separately as commentLikeCount, so the two metrics cannot overwrite each other.

Example input

{
  "youtubeApiKey": "YOUR_KEY",
  "videoUrls": ["https://www.youtube.com/watch?v=dQw4w9WgXcQ"],
  "maxCommentsPerVideo": 100,
  "competitorNames": ["Spotify", "TikTok"],
  "monitorChanges": true
}

Example intelligence row

{
  "recordType": "comment",
  "videoId": "dQw4w9WgXcQ",
  "commentText": "Too many ads. I'm switching to Spotify.",
  "sentiment": "negative",
  "complaintCategories": ["ads"],
  "switchingSignal": true,
  "competitorsMentioned": ["Spotify"],
  "audienceIntent": "switching"
}

Monitoring behavior

Monitoring state is Actor-scoped and keyed by the video plus order/search/reply/competitor settings that affect the result set. Changing those settings creates a separate baseline instead of producing false change alerts. On the first monitoring run, isNewSinceLastRun and newCommentsSinceLastRun are null because there is no earlier run to compare against. When monitoring is disabled, both change fields are also null.

Error behavior

• commentsDisabled: diagnostic row, no retry loop.
• invalid video: diagnostic row.
• 408/429/5xx/network failures: bounded exponential retries.
• quotaExceeded, dailyLimitExceeded, invalid key, or API-not-enabled: circuit breaker stops additional YouTube requests.
• request budget reached: graceful partial success.

Pricing

Suggested PPE event: comment-returned = $0.0015 per successfully delivered new comment intelligence row ($1.50 / 1,000). Summaries, diagnostics, retries, API failures, cached duplicates, and previously seen monitoring rows are not custom-billed.

The event definition in this package does not activate monetization by itself. Configure Pay Per Event in Apify Console before publishing.

Limits

This version targets public video comments. Comments can be disabled by a video owner. The YouTube Data API has its own quota limits. The Actor does not bypass those limits.

Independence

DataHunter_Labs is independent and is not affiliated with or endorsed by YouTube or Google.