YouTube Scraper: Videos, Channels, Shorts & Search

Scrape YouTube videos from searches, channels, Shorts, live streams and playlists, with no API key and no login, and get a report for every channel and search that tells you what works:

• Search with YouTube's own filters: sort by views or date, upload date (hour to year), length, videos or Shorts
• Channels by link or @handle: the Videos, Shorts and Live tabs, sorted by latest, most popular or oldest, up to 5,000 videos per tab
• Playlists and single videos
• Every video: title, channel, views, date, length, Short or live, thumbnail and link
• Details (optional): exact views, likes, comment count, full description, hashtags and exact publish date
• Reports: top videos, fastest-growing videos, outliers (videos that beat the channel's usual views), upload pace, title words, top channels in a search, and channel facts (subscribers, total views, join date, country, links)

What does this YouTube scraper do?

Marques Brownlee · @mkbhd · 21.3M subscribers · 5.73B total views · joined Mar 21, 2008
 Last 40 videos: median 4.8M views, 1.5 uploads a week, 9 in the last 30 days
 Outlier: "iPhone 18 Pro/Duo Impressions: Mogged" 27.5M views, 5.7× the usual

Search "best budget laptop" (this month, by views)
 Median 183K views · top channels: Just Josh, Techum, REVIEW SHEVIEW
 Outlier: "My Son Needs a New Laptop" 4.3M views, 23.6× the median

1. Add search terms, channels, playlists or video links, in any mix.
2. Pick search filters, channel tabs and sort order, and how many videos to read from each.
3. You get one row per video in the dataset, one report per search, channel tab and playlist, and a file of channel facts.

flowchart LR
    A[Search terms<br/>+ filters] --> D[Videos<br/>one row each]
    B[Channels<br/>videos, Shorts, live] --> D
    C[Playlists,<br/>video links] --> D
    D --> E[Details<br/>likes, comments,<br/>description, date]
    D --> F[Reports<br/>top, outliers,<br/>upload pace,<br/>channel facts]

Who is it for?

• YouTubers: find the outliers in your niche, the titles and topics that beat a channel's usual views, and how often competitors upload
• Marketers and agencies: shortlist creators for a campaign by views, growth and upload pace; track a brand's channel
• Researchers and analysts: what YouTube ranks for a search, who dominates it, and how fresh the results are
• Product and e-commerce teams: every review of your product (or a competitor's) with views, likes and comments
• AI and data teams: titles, descriptions and hashtags at scale

What data do you get?

Videos (dataset, one row per video)

Field	What it tells you
sourceType, source, rank	Which search, channel tab or playlist it came from, and its position
videoId, title, url, thumbnailUrl	The video
channelName, channelId, channelUrl, channelSubscribers	Who posted it
views, viewsRounded	Views (exact in search results and with details; rounded on channel pages and playlists)
publishedTimeText, approxPublishedAt, ageDays	When it was posted
duration, durationSeconds	Length
isShort, isLive, isUpcoming, wasLive	Short, live now, scheduled, or a past live stream
descriptionSnippet	Description snippet (search results)
likes, comments, publishDate, description, hashtags	With details on

Report (REPORTS and REPORT_<type>-<name>, one per search, channel tab and playlist)

Field	What it tells you
videosRead, totalViews, medianViews, meanViews, medianViewsPerDay	How the videos perform
topVideos, fastestGrowing	The 10 most viewed, and the 10 with the most views per day since posting
outliers	Videos with 3× or more the median views, and by how much (needs 10+ videos)
uploads, uploadsPerWeek, newestVideo	Upload pace (channels read by latest)
medianLengthSeconds, shortsSharePercent, liveSharePercent	What kind of videos
topTitleWords	Words that recur in titles
topChannels	Who dominates a search or playlist
channel	Subscribers, total views, video count, join date, country, description, links

CHANNELS lists every channel read, REPORTS_CSV has one flat row per report, and RUN_SUMMARY lists problems and the limits of the data.

Example: 2 October 2026

One run with 2 searches (this month, sorted by views), 4 channels, a playlist and a video, 40 videos each with details: 243 videos and 7 reports.

Source	Median views	Upload pace	Biggest outlier
Search "iphone 18 review"	4.0M	-	Apple Event, 50.4M views (12.4×)
Search "best budget laptop"	183K	-	"My Son Needs a New Laptop", 4.3M (23.6×)
Marques Brownlee, videos	4.8M	1.5 a week	"iPhone 18 Pro/Duo Impressions", 27.5M (5.7×)
Veritasium, videos	10.3M	0.75 a week	"The Closest Thing We Have to Alien...", 59.7M (5.8×)
MrBeast, Shorts	128M	1.4 a week	"Who's The Strongest Streamer?", 1.01B (7.9×)
Lex Fridman, videos	1.0M	0.8 a week	"Pavel Durov: Telegram...", 3.1M (3.1×)

Apple posted 8 of the top 40 "iphone 18 review" videos this month; Marques Brownlee and Mrwhosetheboss 4 each.

A second run read 1,500 videos and 123 Shorts from one channel plus 293 search results in 44 seconds.

Pricing

$0.002 per video ($2 per 1,000), $0.002 per video for details (when turned on), and $0.02 per report.

• A channel's last 50 videos with a report: $0.12
• The same with details: $0.22
• 1,000 videos from one channel, no details, one report: $2.02
• Single video links include details at $0.002 each

Turn off reports to pay for videos only.

Good to know

• Views: search results and details give exact views. Channel tabs and playlists show YouTube's rounded figure ("4.6M views", viewsRounded: true); turn on details for exact numbers.
• Dates: listings show "2 days ago", so approxPublishedAt is approximate. Details add the exact publish date. Shorts listings carry no date or length; details add the date.
• Likes are exact when YouTube sends them, otherwise rounded (likesRounded). Comment counts are rounded by YouTube; empty when comments are turned off.
• YouTube's search ends after a few hundred results (293 for "python tutorial"). Use filters or more specific terms to go further.
• Upload pace is only worked out for channels read by "Latest"; "Most popular" gives a channel's all-time best videos instead.
• No personal data: public channel and video information only. Email addresses in channel and video descriptions are hidden.
• Large runs keep going: YouTube blocks an address after a few hundred pages. The Actor then switches to Apify Proxy on its own (datacenter first, residential only if needed), at no extra cost to you. If every address is blocked, the run stops cleanly and keeps what it read.
• Data comes from YouTube's own web API, the one youtube.com uses for search, channel pages and playlists.

Input

Field	Meaning
Search terms	Words, or YouTube search links
Channels	Channel links, @handles or channel ids (a link to a Shorts or Live tab reads that tab)
Playlists	Playlist links or ids
Videos	Video or Shorts links or ids
Videos per search, channel tab or playlist	1 to 5,000 (default 50)
Search: sort, upload date, length, videos or Shorts	YouTube's search filters
Channel: tabs	Videos, Shorts, Live streams
Channel: sort	Latest, most popular, oldest
Add video details	Off by default
Reports	On by default

Integrations

Export to JSON, CSV, Excel or Google Sheets, or connect through the Apify API, webhooks, Make, Zapier and n8n. Schedule it to follow competitors' channels or a search every day.