TikTok Comment Scraper — Comments, Replies & Analytics

Not just comments — comment intelligence.

Most TikTok comment scrapers hand you raw API fields to clean up yourself. This one gives you comments, replies, author profiles, video context, and built-in text analytics (language, emoji, hashtags, mentions) in clean, flat, spreadsheet-ready rows — with separate Unified, Overview, Authors, and Replies views. ~41 structured fields per record. Export in JSON, CSV, or Excel.

---

Key Features

• ~41 Structured Fields per record — comments, replies, authors, analytics
• Reply Threads — full reply chains linked to parent comments via parentCommentId
• Replies After Parent — each comment's replies appear right after it, numbered via threadPosition (comment, reply #1, reply #2, …)
• Author Metadata — username, nickname, verified status, private account flag, avatar (always included)
• Comment Analytics — language detection, word/emoji counts, hashtag/mention extraction
• Video Context — video caption, views, likes, shares, duration, music info on every record (always included)
• Sorting — top comments or newest first
• Filters — minimum likes, date range, exclude pinned, scrape-all
• Profile Input — scrape comments across a creator's recent videos from just their @handle
• Unified Fields — shared id / parentId / text / likes / createdAt / author* columns on both comments and replies, so a single table has no empty cells
• 4 Dataset Views — Unified, Overview, Authors, Replies on Apify
• Bare IDs, URLs, or Profiles — accepts full video URLs, bare numeric video IDs, or creator handles
• Error Rows — failed/blocked videos emit a visible error row instead of vanishing silently
• No Proxy Needed — data is fetched through TikHub's hosted API (signing + routing handled server-side)
• Pay-Per-Event — only pay for results you get

Best for: Sentiment analysis, brand monitoring, content research, AI training data, moderation analysis

---

At a Glance

Feature	Details
Output Fields	~41 structured fields per record
Dataset Views	4 tabs (Unified, Overview, Authors, Replies)
Input	TikTok video URLs, bare video IDs, or creator profiles
Max Comments	Up to 10,000 per video
Reply Threads	Enabled by default, linked via parentCommentId / replyId
Data source	TikHub hosted API (no proxy setup required)

---

How It Works

1. Provide inputs — one or more TikTok video URLs, bare numeric video IDs, or creator profiles (@handle)
2. Fetch browserlessly via TikHub — comments, reply threads, and video metadata are pulled through TikHub's hosted API with cursor-based pagination (no browser, no proxy). Creator profiles are expanded to their recent videos first.
3. Normalize — each comment and reply becomes a flat row with unified + type-specific fields, plus built-in text analytics
4. Filter & cap — apply minLikes, date range, and excludePinned; stop at maxComments per video / maxItems overall
5. Export — push to the dataset (4 views) — each comment followed by its replies

If TikHub can't serve a particular video, the actor automatically falls back to a headful browser path for that video, so a single hard-to-reach video doesn't fail the run.

---

Output

Comment Record

A single dataset row for a comment \u2014 every field an actual record contains (unified + comment + analytics + author + video):

{
  "recordType": "comment",
  "id": "7482114472766325535",
  "parentId": null,
  "threadPosition": "comment",
  "text": "\u201cShe DIED\u201d \ud83e\udd19\ud83c\udffc\ud83e\udd19\ud83c\udffc\ud83e\udd19\ud83c\udffc\ud83d\ude02\ud83d\ude2d",
  "likes": 1507,
  "createdAt": "2025-03-15T19:05:59.000Z",
  "commentId": "7482114472766325535",
  "videoId": "7481695035033980190",
  "videoUrl": "https://www.tiktok.com/@aaron.vankampen/video/7481695035033980190",
  "inputUrl": "https://www.tiktok.com/@aaron.vankampen/video/7481695035033980190",
  "commentText": "\u201cShe DIED\u201d \ud83e\udd19\ud83c\udffc\ud83e\udd19\ud83c\udffc\ud83e\udd19\ud83c\udffc\ud83d\ude02\ud83d\ude2d",
  "createTime": "2025-03-15T19:05:59.000Z",
  "likeCount": 1507,
  "replyCount": 4,
  "isPinned": false,
  "commentPermalink": "https://www.tiktok.com/@aaron.vankampen/video/7481695035033980190?comment=7482114472766325535",
  "scrapedAt": "2026-03-14T13:29:33.073Z",
  "commentLanguage": "en",
  "wordCount": 3,
  "charCount": 27,
  "emojiCount": 8,
  "hashtags": [],
  "mentions": [],
  "authorId": "2904724",
  "authorUsername": "nellyyybaby",
  "authorNickname": "baby J\ud83d\udc8b",
  "authorVerified": false,
  "authorPrivateAccount": false,
  "authorProfileUrl": "https://www.tiktok.com/@nellyyybaby",
  "authorAvatar": "https://p16-common-sign.tiktokcdn-us.com/...",
  "videoDescription": "when the cat picked the crib \ud83d\ude02\ud83d\udc31",
  "videoCreateTime": "2025-03-14T21:40:12.000Z",
  "videoLikeCount": 2100000,
  "videoCommentCount": 8452,
  "videoShareCount": 15300,
  "videoViewCount": 18500000,
  "videoDuration": 32,
  "videoRegion": "US",
  "musicTitle": "original sound",
  "musicAuthor": "aaron.vankampen"
}

Reply Record (appears right after its parent comment)

Replies carry the same unified, analytics, and always-on video* fields as comments (omitted below for brevity), plus reply-specific fields:

{
  "recordType": "reply",
  "id": "7482124949542011655",
  "parentId": "7482114472766325535",
  "threadPosition": "reply #1",
  "text": "\"So now aiiii got da crib!!\" \ud83d\ude03",
  "likes": 32,
  "createdAt": "2025-03-15T19:46:40.000Z",
  "parentCommentId": "7482114472766325535",
  "replyId": "7482124949542011655",
  "videoId": "7481695035033980190",
  "videoUrl": "https://www.tiktok.com/@aaron.vankampen/video/7481695035033980190",
  "inputUrl": "https://www.tiktok.com/@aaron.vankampen/video/7481695035033980190",
  "replyText": "\"So now aiiii got da crib!!\" \ud83d\ude03",
  "replyLikeCount": 32,
  "replyCreateTime": "2025-03-15T19:46:40.000Z",
  "replyAuthorId": "6812290091522016262",
  "replyAuthorUsername": "tony6ixx",
  "replyAuthorVerified": false,
  "replyAuthorNickname": "Tony \ud83c\udde6\ud83c\uddf1",
  "replyAuthorAvatar": "https://p19-common-sign.tiktokcdn-us.com/...",
  "authorProfileUrl": "https://www.tiktok.com/@tony6ixx",
  "commentLanguage": "en",
  "wordCount": 7,
  "charCount": 31,
  "emojiCount": 1,
  "hashtags": [],
  "mentions": [],
  "scrapedAt": "2026-03-14T13:29:33.329Z"
}

---

Output Schema

Unified Fields (8) — present on every comment and reply row

Comments and replies use different field names (commentId/replyId, commentText/replyText, …), so in a single flat table the "other" type's columns show as blank. These unified fields solve that — they're populated on both row types (in addition to the type-specific fields), so you can build a clean, empty-free table from just these columns. The Unified dataset view uses exactly these.

Field	Type	Description
recordType	String	comment, reply, or error
id	String	The row's own ID — commentId for comments, replyId for replies
parentId	String	Parent comment's ID for replies; null for top-level comments
threadPosition	String	comment for a top-level comment; reply #1, reply #2, … for its replies in order
text	String	Comment or reply text
likes	Integer	Like count (comment or reply)
createdAt	String	ISO 8601 creation timestamp
author*	—	Shared authorId, authorUsername, authorNickname, authorVerified, authorProfileUrl on both types

Rows are pushed in reading order — each comment, then its replies (reply #1, reply #2, …), then the next comment — so threadPosition lets you see the thread structure at a glance.

To join replies to comments: reply.parentId (= reply.parentCommentId) equals the parent's id (= commentId).

Core Comment (12 fields)

Field	Type	Description
recordType	String	Row type: comment, reply, or error
commentId	String	Unique comment ID
videoId	String	Video identifier
videoUrl	String	Canonical TikTok video URL
inputUrl	String	The original URL or ID you supplied for this row
commentText	String	Full comment text
createTime	String	ISO 8601 timestamp
likeCount	Integer	Number of likes
replyCount	Integer	Number of replies
isPinned	Boolean	Pinned by creator
commentPermalink	String	Direct link to comment
scrapedAt	String	Scraping timestamp

Comment Analytics (6 fields)

Field	Type	Description
commentLanguage	String	Detected language code (e.g. en, zh, ja)
wordCount	Integer	Word count
charCount	Integer	Character count
emojiCount	Integer	Number of emojis
hashtags	Array	Extracted #hashtags
mentions	Array	Extracted @mentions

Author Metadata (7 fields)

Always included on every comment/reply record.

Field	Type	Description
authorId	String	Author's TikTok user ID
authorUsername	String	Author's @username
authorNickname	String	Display name
authorVerified	Boolean	Verified badge
authorPrivateAccount	Boolean	Private account
authorProfileUrl	String	Link to the author's TikTok profile
authorAvatar	String	Profile image URL

Reply Fields (10 fields)

Included when Include Replies is enabled (default: on). Each reply is a separate dataset row linked to its parent via parentCommentId.

Field	Type	Description
parentCommentId	String	Parent comment's ID
replyId	String	Reply's unique ID
replyText	String	Reply text
replyLikeCount	Integer	Reply likes
replyCreateTime	String	ISO 8601 timestamp
replyAuthorId	String	Reply author's user ID
replyAuthorUsername	String	Reply author's username
replyAuthorVerified	Boolean	Reply author verified
replyAuthorNickname	String	Reply author's display name
replyAuthorAvatar	String	Reply author's avatar URL

Reply records also include the unified fields (recordType: "reply", id, parentId, text, likes, createdAt, plus author* mirrors), inputUrl, authorProfileUrl (reply author's profile), the analytics fields (commentLanguage, wordCount, charCount, emojiCount, hashtags, mentions) computed on the reply text, videoId, videoUrl, scrapedAt, and the same always-on video metadata (10 fields) as comments.

Video Metadata (10 fields)

Always included on every comment/reply record. Fetched browserlessly via TikHub — same fast path as comments, so it adds no browser launch and negligible time.

Field	Type	Description
videoDescription	String	Video caption
videoCreateTime	String	Publish timestamp
videoLikeCount	Integer	Video likes
videoCommentCount	Integer	Video comment count
videoShareCount	Integer	Video shares
videoViewCount	Integer	Video views
videoDuration	Integer	Duration in seconds
videoRegion	String	Region code
musicTitle	String	Background music title
musicAuthor	String	Music creator

Error Rows

When a video can't be scraped (comments disabled, private/deleted, soft-blocked IP, or an invalid input), the actor emits a visible error row instead of silently returning nothing — so failures are easy to spot and retry. Error rows are not charged.

Field	Type	Description
recordType	String	Always error
error	String	Human-readable failure reason
errorCode	String	INVALID_URL, NO_COMMENTS, EXTRACTION_FAILED, REQUEST_FAILED, or PROFILE_NO_VIDEOS
videoUrl	String	The video URL that failed
inputUrl	String	The original URL or ID you supplied
scrapedAt	String	Timestamp of the failure

---

Input Parameters

Provide at least one of videoUrls or profileUrls.

Parameter	Type	Default	Required	Description
videoUrls	Array	—	*	TikTok video URLs or bare video IDs to scrape
profileUrls	Array	—	*	Creator profiles (@handle, profile URL, or bare handle) — scrapes their recent videos
maxVideosPerProfile	Integer	10	No	Recent videos to scrape per profile (1–500)
maxComments	Integer	100	No	Max comments per video (1–10,000)
maxItems	Integer	—	No	Global cap on total results (comments + replies) across all videos. Empty = no cap
scrapeAllComments	Boolean	false	No	Ignore maxComments and fetch every comment (global maxItems still applies)
minLikes	Integer	—	No	Keep only comments/replies with at least this many likes
commentDateFrom	String	—	No	Keep only comments/replies created on/after this date (ISO, e.g. 2025-01-01)
commentDateTo	String	—	No	Keep only comments/replies created on/before this date
excludePinned	Boolean	false	No	Skip creator-pinned comments (and their replies)
includeReplies	Boolean	true	No	Include reply threads
maxRepliesPerComment	Integer	50	No	Max replies per comment (1–500)
sort	String	"top"	No	Sort order: top or new

---

Quick Start

Example 1: Single video with replies

{
  "videoUrls": [
    "https://www.tiktok.com/@aaron.vankampen/video/7481695035033980190"
  ],
  "maxComments": 100,
  "includeReplies": true,
  "maxRepliesPerComment": 50,
  "sort": "top"
}

Example 2: Creator profiles + filters

Scrape recent videos from two creators, keep only comments with ≥10 likes from 2025.

{
  "profileUrls": ["@aaron.vankampen", "https://www.tiktok.com/@tiktok"],
  "maxVideosPerProfile": 5,
  "minLikes": 10,
  "commentDateFrom": "2025-01-01"
}

Example 3: Bare video IDs with a global cap, no pinned comments

{
  "videoUrls": [
    "7481695035033980190",
    "7546661132002413854"
  ],
  "maxItems": 500,
  "excludePinned": true,
  "includeReplies": true,
  "sort": "new"
}

Example 4: Multiple videos, comments only

{
  "videoUrls": [
    "https://www.tiktok.com/@aaron.vankampen/video/7481695035033980190",
    "https://www.tiktok.com/@aaron.vankampen/video/7546661132002413854"
  ],
  "maxComments": 100,
  "includeReplies": false,
  "sort": "top"
}

Example 5: Every comment on a video, newest first

{
  "videoUrls": [
    "https://www.tiktok.com/@creator/video/7234567890123456789"
  ],
  "scrapeAllComments": true,
  "includeReplies": true,
  "sort": "new"
}

---

Dataset Views

The output is organized into 4 focused tabs on the Apify platform:

Unified

Comments and replies in one table with no empty columns — uses the shared unified fields.

{
  "recordType": "reply",
  "id": "7482124949542011655",
  "parentId": "7481790853061559070",
  "threadPosition": "reply #1",
  "text": "So now aiiii got da crib!!",
  "likes": 32,
  "createdAt": "2025-03-15T19:46:40.000Z",
  "authorUsername": "tony6ixx",
  "authorNickname": "Tony",
  "authorProfileUrl": "https://www.tiktok.com/@tony6ixx"
}

Overview

Shows comments and replies together with key fields.

{
  "recordType": "comment",
  "commentId": "7481802613793686318",
  "videoId": "7481695035033980190",
  "commentText": "YOU FOUND CODY!!!",
  "authorUsername": "duncanstanko",
  "likeCount": 489,
  "replyCount": 1,
  "isPinned": false,
  "createTime": "2025-03-14T22:55:34.000Z",
  "scrapedAt": "2026-03-14T13:29:31.670Z"
}

Authors

Focused view on commenter profiles.

{
  "commentId": "7481802613793686318",
  "authorId": "7158470338833777710",
  "authorUsername": "duncanstanko",
  "authorNickname": "Duncan Stanko",
  "authorVerified": false,
  "authorPrivateAccount": false,
  "authorProfileUrl": "https://www.tiktok.com/@duncanstanko",
  "authorAvatar": "https://p16-common-sign.tiktokcdn-us.com/..."
}

Replies

Focused view on reply threads with parent-child linking.

{
  "parentCommentId": "7482422764545753902",
  "replyId": "7482493179494253342",
  "replyText": "He is a comedian. Calm down lol",
  "replyAuthorUsername": "wizrd.618",
  "replyLikeCount": 107,
  "replyCreateTime": "2025-03-16T19:35:28.000Z"
}

---

Use Cases

Sentiment Analysis

Export comments for sentiment scoring and reaction analysis. Use commentLanguage, wordCount, and emojiCount for NLP preprocessing.

Brand Monitoring

Track public feedback on brand campaigns. Filter by video URL and analyze comment engagement patterns.

Content Research

Study audience reactions to viral content. Compare top comments vs. newest to understand engagement dynamics.

AI Training Data

Collect structured comment datasets for training text classification, sentiment, or moderation models.

Moderation Analysis

Identify patterns in toxic comments, spam, or abuse. Use author metadata to detect bot-like accounts.

---

Performance

Metric	Value
Concurrency	1 request at a time
Request timeout	120 seconds
Navigation timeout	60 seconds
Retries	Up to 4 per request (exponential backoff)
Session pooling	5 sessions, 20 uses each

---

Data Export

Supported formats via Apify platform:

• JSON
• CSV
• Excel
• HTML

---

Tips

• No proxy setup needed — data is fetched through TikHub's hosted API.
• Replies appear after their parent in the dataset output — look for parentCommentId to match them.
• Each reply is its own row with a flat structure, making it easy to filter comments vs. replies.
• commentPermalink links directly to the comment on TikTok.
• Language detection uses Unicode script analysis — returns null for ambiguous text.

---

Pricing

This actor uses Apify's Pay-Per-Event model — you only pay for results you get.

Event	Trigger	Description
result-scraped	Each comment or reply extracted	Charged once per record pushed to dataset

Only successful results are charged — error rows and skipped (pinned) comments are free.

Example cost

At a list price of $0.49 per 1,000 results:

Results scraped	Approx. cost
100	$0.05
1,000	$0.49
10,000	$4.90

The exact event price is set in the Apify Console. Platform compute is billed separately to your Apify account. No proxy is required — data is fetched through TikHub's hosted API.

---

How data is fetched

This actor uses TikHub's hosted API for fast, reliable, browserless access — no proxy configuration required. TikHub handles request signing and routing server-side, which is what makes runs fast and consistent. It powers:

• Comment & reply extraction — the primary, browserless data path
• Video metadata — the 10 video-context fields on every record, fetched browserlessly (no page load)
• Profile input (profileUrls) — resolving a creator's recent videos

The TikHub key is configured once by the actor operator (as the Apify secret tikhub_api_key, wired in actor.json as @tikhub_api_key) — run users never need to provide anything. TikHub usage is billed to the operator's TikHub account, separate from Apify's result-scraped charge.

---

Compliance

• Collects only publicly available TikTok comment data
• Does not bypass CAPTCHAs or access gated content
• Does not collect private account data
• Intended for legitimate research, analytics, and business intelligence
• Users are responsible for compliance with applicable laws in their jurisdiction

---

Support

• Issues: Report via GitHub or Apify support
• Feature requests: Open an issue with your use case

---

Built by Coregent Digital Studio — Production-ready TikTok comment extraction