YouTube Search Scraper: Videos, Shorts, Channels & Playlists
Pricing
from $2.00 / 1,000 search results
YouTube Search Scraper: Videos, Shorts, Channels & Playlists
Scrape YouTube search results for any keyword: rank, title, channel, views, upload age, length and badges for videos, Shorts, channels and playlists. Use YouTube's own filters (upload date, length, 4K, live, Creative Commons), sort by popularity, set the country. No login or API key.
Pricing
from $2.00 / 1,000 search results
Rating
0.0
(0)
Developer
Philip Kirkbride
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
3 days ago
Last modified
Features
Share
Type a keyword and get YouTube's search results as a clean spreadsheet: rank, title, channel, views, upload age, length, thumbnail and badges for every video, Short, channel or playlist YouTube shows. Use the same filters as YouTube's own search menu, and track where videos rank for the keywords you care about.
What you get
- One row per search result, in YouTube's order: rank, result type, video / channel / playlist ID, link, title, channel name and ID, view count, upload age with an estimated date, duration, thumbnail, and the Shorts, live and badge markers (4K, CC, New...).
- YouTube's own filters: result type (videos, Shorts, channels, playlists or everything), upload date (today, this week, month or year), length (under 3, 3–20 or over 20 minutes), features (live now, 4K, 360°, VR180, Creative Commons) and relevance or popularity order.
- Only recent videos:
publishedAfterkeeps results published after a date such as2026-10-01or7 days. - Any market: search as a viewer in another country and interface language, for example
GBanden, orDEandde. - Paste searches the way you have them: plain keywords, YouTube search-results links, or the
startUrlslist and the input keys you already use with other YouTube scrapers. - No login, API key or quota. Export JSON, CSV or Excel, or connect through the API.
Use cases
- Keyword and rank tracking: schedule a daily run and see which videos rank for your keywords, and where your own videos sit.
- Content research: find the most popular videos on a topic this month, with views and length, before you plan your own.
- Competitor and trend monitoring: new uploads on a keyword this week, across countries.
- Lead and creator discovery: channel results for a niche, with channel IDs ready for a channel or comments scraper.
Also available
The YouTube suite: pick the job.
| Actor | Use when |
|---|---|
| YouTube Search (this Actor) | Ranked keyword results with YouTube's search filters |
| YouTube Scraper | Videos, channels, search results and comments |
| YouTube Transcripts | Bulk transcripts as text, SRT or VTT |
| YouTube Outlier Finder | Breakout videos that beat their channel median |
Input
The Console prefill collects the top 10 results for one keyword:
{"searchTerms": ["python tutorial"],"maxResultsPerQuery": 10}
The most popular long videos on two keywords from the last 30 days, as shown in the UK:
{"searchTerms": ["home espresso", "budget headphones"],"maxResultsPerQuery": 50,"sort": "popularity","duration": "long","publishedAfter": "30 days","country": "GB"}
| Field | What it does |
|---|---|
searchTerms | Keywords, one per line. A pasted youtube.com/results?search_query=… link runs as that search. |
startUrls | Search-results links as plain strings or {"url": "…"} objects. |
maxResultsPerQuery | Results per search (1–5,000). If left unset, uses 20. |
type | video (default), shorts, channel, playlist or all. |
sort | relevance (default) or popularity. |
uploadDate | today, week, month or year. For videos, Shorts or all types. |
duration | short (under 3 min), medium (3–20 min) or long (over 20 min). For videos or all types. |
features | Any of live, 4k, 360, vr180, creative_commons; videos must have all of them. |
publishedAfter | A date (2026-10-01) or a period (7 days, 6 months). |
country, language | Search market, for example GB and en. If left unset, uses US and en. |
maxItems | Hard total across all searches (default 1,000). |
maxPages | Result pages per search (default 50). |
Switching from another YouTube scraper? Your usual keys work: searchQueries, keywords, query, maxResults, sortingOrder, dateFilter, lengthFilter, isLive / is4K / is360 / isVR180 / hasCC, oldestPostDate, searchSort, searchType, searchFeatures, regionCode, geo and lang. Each translated key is listed in OUTPUT.notices; when the same setting is given twice, this Actor's own field wins.
Output example
Captured from a public search on November 13, 2026:
{"query": "python tutorial","rank": 1,"type": "video","videoId": "nKPbfIU442g","channelId": "UCtoo4_P6ilCj7jwa4FmA5lQ","playlistId": null,"url": "https://www.youtube.com/watch?v=nKPbfIU442g","title": "PYTHON course from ZERO (Complete)","channelName": "Soy Dalto","viewCount": 6270466,"publishedText": "3 years ago","publishedAtEstimate": "2023-11-14T00:34:07Z","publishedAtIsEstimate": true,"duration": 29190,"durationText": "8:06:30","thumbnailUrl": "https://i.ytimg.com/vi/nKPbfIU442g/hq720.jpg","isShort": false,"isLive": false,"badges": ["4K","CC"],"scrapedAt": "2026-11-13T18:34:07Z"}
Run summary
OUTPUT reports each search with its delivered rows and pages, any errors, the stopping reason and, with publishedAfter, how many older results were skipped. A search that YouTube answers with "No results found" ends the run successfully with OUTPUT.status: "no_results", and so does a run where every result was older than publishedAfter. Earlier rows stay available when a later page fails. A charge cap reports "allowed N of M requested rows". Results are saved throughout the run.
Pricing
$0.002 per result ($2.00 per 1,000) plus $0.005 per Actor start event. The start event is charged per GB of memory, rounded up with a minimum of one event. The default 512 MB run is one start event. Only delivered dataset rows generate row events.
These examples use the default 512 MB memory:
| Example | Results | Cost |
|---|---|---|
| Console prefill | 10 | $0.025 |
| 100 results | 100 | $0.205 |
| 1,000 results | 1,000 | $2.005 |
Good to know
- Public, logged-out YouTube search only. Private, deleted, members-only and age-restricted results are not available.
publishedAfterreads the age YouTube shows ("3 weeks ago"), so it is approximate: a video uploaded just before your date can be kept, and a newer one is never dropped. WhenuploadDateis empty, the smallest YouTube upload-date filter that covers your date is applied first. Shorts results show no upload age, so for Shorts the date comes from that YouTube filter alone;OUTPUT.publishedAfter.undatedcounts them.publishedAfterdoes not apply to channel or playlist results.- Popularity is YouTube's own ranking, not a strict view-count sort. YouTube's search menu no longer offers date or rating order; those values from other scrapers are ignored with a notice. The HD, subtitles, 3D, HDR, location and purchased filters are not applied, because YouTube does not apply them reliably.
- View counts can be rounded as YouTube shows them.
publishedAtEstimateis an estimate from the upload age (flagged bypublishedAtIsEstimate); labels in languages other than English stay null. Fields missing from a result card are null. - A video, channel or playlist link pasted as a search term is skipped with a note in
OUTPUT.errorsand is not charged. Filters inside a pasted search link are not used; settype,uploadDate,duration,featuresandsortinstead. - Duplicate results within one search are collapsed, including the watch and Shorts forms of the same video; the same video can appear under different searches. Pagination was verified through 19 pages and 300 results in one search; YouTube can end a search earlier.
- Country and language set the search market only; YouTube decides which results it shows there. Empty input (
{}) runs a five-result example for "python tutorial" and says so in the run status.