YouTube Comment Scraper

YouTube Comment Scraper creates a structured dataset of comments collected from public YouTube videos and playlists. Each dataset item represents one comment and can include the comment text, author, likes, reply status, publish timestamp, and the source video's title, URL, view count, and channel. Query the source by video or playlist URL, control the result limit with maxItems, and retrieve records through the Apify Dataset API or export them as JSON, CSV, or Excel.

Dataset at a glance

Property	Value
Source	youtube.com (public videos and playlists)
Record unit	One comment (or reply)
Input methods	Video or playlist URLs (videoUrls)
Main identifiers	commentId, videoId
Delivery	Apify Dataset and API
Export formats	JSON, CSV, Excel
Update model	Fresh records per Actor run
Pricing	$2.50 per 1,000 comments

Coverage and available records

The Actor collects comments from one or more YouTube videos or playlists.

• Video input: videoUrls accepts individual video URLs. Playlist URLs are expanded into their videos automatically; a video URL carrying both v= and list= parameters is treated as a single video.
• Sort order: commentSort selects new (newest first) or top (top comments).
• Result limit: maxItems caps the total number of comments collected across all videos (0 = unlimited).
• Parallelism: maxConcurrency controls how many videos are processed at once (1–10).
• Replies: Replies are collected as individual records and carry isReply and parentCommentId.

Known exclusions: comments are limited to what YouTube returns for the chosen sort order; each run captures comment state at run time (no historical snapshots); the Actor does not retrieve the full nested reply tree beyond returned replies.

Data dictionary

Field names below match dataset record JSON properties exactly.

Field	Type	Nullable	Description	Example
commentId	string	No	YouTube comment identifier; best stable deduplication key	UgzuC3zzpRZkjc5Qzsd4AaABAg
text	string	No	Comment text content	We're so honored...
author	string	Yes	Display name of the comment author	@SanDiegoZoo
authorChannelId	string	Yes	Channel ID of the comment author	UCC5NfQ6Mf0dq_eEwv4P_hWA
likeCount	integer	Yes	Number of likes on the comment	4600000
isFavorited	boolean	Yes	Whether the video creator hearted the comment	true
isReply	boolean	Yes	Whether the comment is a reply to another comment	false
parentCommentId	string	Yes	Parent comment ID when isReply is true	UgzuC3zzpRZkjc5Qzsd4AaABAg
publishedAt	string	Yes	ISO 8601 timestamp of the comment	2020-07-09T00:00:00+00:00
videoId	string	Yes	YouTube video ID of the source video	jNQXAC9IVRw
videoUrl	string	Yes	Canonical URL of the source video	https://www.youtube.com/watch?v=jNQXAC9IVRw
videoTitle	string	Yes	Title of the source video	Me at the zoo
channelName	string	Yes	Name of the channel that published the video	jawed
channelId	string	Yes	Channel ID of the video publisher	UC4QobU6STFB0P71PMvOGN5A
videoPublishedAt	string	Yes	Publication date of the source video	2005-04-24
videoViewCount	integer	Yes	View count of the source video at scrape time	398675009
videoCommentCount	integer	Yes	Total comment count reported for the source video	10

Example dataset record

A representative record produced from the video https://www.youtube.com/watch?v=jNQXAC9IVRw:

{
  "commentId": "UgzuC3zzpRZkjc5Qzsd4AaABAg",
  "text": "We're so honored that the first ever YouTube video was filmed here!",
  "author": "@SanDiegoZoo",
  "authorChannelId": "UCC5NfQ6Mf0dq_eEwv4P_hWA",
  "likeCount": 4600000,
  "isFavorited": true,
  "isReply": false,
  "parentCommentId": null,
  "publishedAt": "2020-07-09T00:00:00+00:00",
  "videoId": "jNQXAC9IVRw",
  "videoUrl": "https://www.youtube.com/watch?v=jNQXAC9IVRw",
  "videoTitle": "Me at the zoo",
  "channelName": "jawed",
  "channelId": "UC4QobU6STFB0P71PMvOGN5A",
  "videoPublishedAt": "2005-04-24",
  "videoViewCount": 398675009,
  "videoCommentCount": 10
}

Query and input reference

Input	Type	Required	Default	Accepted values	Description
videoUrls	array[string]	Yes	—	YouTube video or playlist URLs	Source videos; playlists are expanded automatically
maxItems	integer	No	100	0 or a positive integer	Maximum comments across all videos; 0 = unlimited
commentSort	string	No	new	new, top	Newest first, or top comments
maxConcurrency	integer	No	3	1–10	Number of videos processed in parallel
proxyConfiguration	object	No	Apify residential proxy	Apify proxy groups or custom proxies	Residential proxies are recommended

Minimal request:

{ "videoUrls": ["https://www.youtube.com/watch?v=jNQXAC9IVRw"] }

Advanced request:

{
  "videoUrls": ["https://www.youtube.com/playlist?list=..."],
  "maxItems": 1000,
  "commentSort": "top",
  "maxConcurrency": 3,
  "proxyConfiguration": { "useApifyProxy": true, "apifyProxyGroups": ["RESIDENTIAL"] }
}

Retrieve the data through the API

1. Start the Actor with a JSON input (console or API).
2. Wait for the run to finish, or use a synchronous endpoint for an inline response.
3. Retrieve items from the run's default dataset.
4. Paginate or export the dataset.

Python example:

from apify_client import ApifyClient

client = ApifyClient("YOUR-APIFY-TOKEN")

run = client.actor("datascrapers/youtube-comment-scraper").call(run_input={
    "videoUrls": ["https://www.youtube.com/watch?v=jNQXAC9IVRw"],
    "maxItems": 1000,
    "commentSort": "new",
})
for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item["commentId"], item["author"], item["likeCount"])

Apify generates ready-to-run Python, JavaScript, and cURL examples on the Actor's API tab. Do not put a real API token in shared code or URLs.

Data quality and record handling

• Conditional fields: Reply records carry isReply and parentCommentId; top-level comments have parentCommentId null. Fields YouTube does not expose are returned as null.
• Source changes: YouTube page structure and values can change; unreadable fields are returned as null rather than fabricated.
• Deduplication: Each run appends fresh records; the Actor does not deduplicate across runs. Use commentId as the stable key and filter repeated runs against previously stored IDs.
• Rate limits: YouTube rate-limits automated traffic. Residential proxies (enabled by default) and bounded concurrency are used to maintain coverage.
• Normalization: publishedAt is normalized to ISO 8601; view and like counts are read as integers.

Export and pipeline examples

Destination	Recommended method	Typical use
PostgreSQL / Supabase	Dataset API poll or webhook consumer	Store comments for sentiment pipelines
Google Sheets	Apify Google Sheets integration	Share comment threads with community teams
S3 / cloud storage	Scheduled export via Apify scheduler	Archival of comment snapshots
BI tools	CSV / JSON export	Audience and sentiment analysis

Pricing and cost examples

The Actor uses pay-per-event pricing, billed per comment record:

Event	Trigger	Rate
Comment	Every comment pushed to the dataset	$2.50 per 1,000 comments

A one-time Actor start event of $0.00005 applies to each run (effectively $0).

Comments	Estimated base cost
1,000	$2.50
10,000	$25.00
100,000	$250.00

Compute units consumed by the run are billed by your Apify plan. Estimates depend on the verified pricing model and the options selected for the run.

Limitations and responsible data use

• The Actor collects publicly accessible comment data from YouTube only.
• Field availability depends on what YouTube returns at run time; some values can be null or missing, and site changes can alter fields.
• The Actor does not provide historical snapshots unless you store them yourself.
• Large runs require residential proxies; without them, coverage may degrade due to rate limiting.
• You are responsible for compliance with YouTube's terms of service, applicable privacy law, and any contractual obligations before using the data.

Dataset questions

What does one dataset item represent?

One comment or reply. Replies are separate records linked to their parent via parentCommentId.

Which field should I use as a unique identifier?

commentId is the stable YouTube comment identifier and the recommended deduplication key.

Are fields nullable or conditional?

Yes. Reply records carry isReply and parentCommentId; top-level comments have parentCommentId null. Fields YouTube does not expose for a given comment are returned as null.

Can I retrieve the records as CSV or JSON?

Yes. The dataset can be exported as JSON, CSV, or Excel from the Apify Console, and queried through the Dataset API.

Does the Actor return historical data?

No. Each run captures the state of the comments at run time. To track changes, schedule repeated runs and store the outputs yourself.

What counts as a billable result?

Each comment pushed to the dataset is one billable result, charged at $2.50 per 1,000 comments. Use maxItems to control cost per run.

Related datasets from Data Scrapers

• YouTube Channel Scraper — Channel and video metadata that can be joined to comment data by channel or video ID.
• TripAdvisor Reviews Scraper — User review text for sentiment and reputation analysis.
• Google Play Store App Reviews Scraper — App review text for product feedback analysis.
• Twitter X Profile Scraper — Public post data for cross-platform audience research.

Data Scrapers support

Need an additional field, record type, or export workflow? Contact Data Scrapers at stardustspotlight@gmail.com. Include a sample source URL, required fields, expected record volume, and preferred delivery format.