A YouTube scraper for channel and video metadata
The Kavex YouTube scraper pulls public data from videos and channels into a structured file. Give it video URLs, channel URLs or @handles and get back titles, view and like counts, publish dates, descriptions and channel stats. It is built for creators, marketers and analysts who need YouTube data in a spreadsheet rather than a browser tab. You pay per video or channel returned.
Sign up to useWhat it does
The YouTube scraper accepts a mixed list of inputs — direct video URLs, channel URLs and @handles — and returns the right data for each. It sorts out which is which, so you do not have to split your list by type before running it.
For a video it returns the title, the channel, view and like counts, the publish date and the description. For a channel it returns the subscriber count, total videos and a list of recent uploads, so you can see both the size of an account and what it has been publishing lately.
Everything is returned as structured rows, which turns YouTube research into something you can sort, filter and chart. A list of competitor channels or topic videos becomes a single dataset instead of a folder of open tabs.
Numbers are most useful side by side. A single channel subscriber count means little until it sits next to ten others in the same niche, ranked, with recent view performance attached. That is what turns the YouTube scraper into a shortlisting tool: a topic search or a list of competitor channels becomes a table you can sort by reach, by upload frequency, or by how well recent videos actually performed rather than how big the channel looks. For influencer outreach that means contacting creators whose recent numbers justify the spend; for competitive work it means seeing which rivals publish consistently and which have gone quiet.
Use cases
- Influencer marketers sizing creators by subscriber count and recent view performance before outreach.
- Competitive analysts tracking rival channels for new product videos and upload cadence.
- Content teams pulling video metadata across a topic to spot trends and gaps.
- Agencies building a creator shortlist for a client with hard numbers attached to every name.
Sample output
Videos and channels return structured rows:
| Type | Title / Channel | Views / Subs | Likes | Published | Channel |
|---|---|---|---|---|---|
| video | Building a SaaS in public | 84,200 | 3,910 | 2026-04-18 | Indie Founders |
| video | Cold email teardown 2026 | 41,700 | 2,140 | 2026-04-09 | Outbound Lab |
| channel | Indie Founders | 128,000 | — | — | — |
| channel | Outbound Lab | 54,300 | — | — | — |
How it works
The YouTube scraper fetches each video or channel page live and reads the public metadata YouTube exposes — counts, dates, titles and descriptions. Channel inputs also have their recent upload list collected so the result reflects current activity, not a stale snapshot.
Every run is live, so view and subscriber counts are current as of the job. Requests are paced and routed sensibly so a long list of videos or channels completes in one job and downloads as a single file.
Frequently asked questions
Can I mix video and channel links in one job?
Yes. The YouTube scraper accepts video URLs, channel URLs and @handles together, detects each type and returns the right fields for it in one combined file.
What data does it return for a channel?
For a channel it returns the subscriber count, total video count and a list of recent uploads, so you can judge both the size of an account and how active it is.
Are the view and subscriber counts current?
Yes. Every job is a live fetch, so counts reflect YouTube at the moment the job runs rather than a cached figure from a database.
How do results export?
Results download as a CSV with one row per video or channel and columns for every metadata field, ready to sort or chart in a spreadsheet or imported into a wider research dataset.
Try it free — 1000 credits on us
Pay per result — no subscription, no seats. New accounts start with 1,000 free credits.