If you make developer videos, or you're deciding whether to, it helps to know what a normal month looks like in your niche. Not the outliers you see shared everywhere, the whole spread.
So we pulled one month of it. One run of our YouTube Scraper Actor on Apify, searching "web scraping python", videos uploaded in the last month, sorted by popularity, 100 results:
- Time: 64 seconds from request to rows (55 s of run time).
- Cost: under a fifth of a cent of compute on our side. A user pays $0.50 per 1,000 videos on the Starter tier, so these 100 rows are about 5 cents.
- Coverage: 100 videos from 64 channels, uploaded between 10 October and 9 November 2026.
Every number below comes from that one run.
The input
{ "searchTerms": ["web scraping python"], "searchType": "videos", "uploadDate": "month", "sortBy": "popularity", "maxResults": 100, "videoDetails": true } videoDetails opens each video page, which adds the exact view and like counts, the publish date, chapters, tags and the channel's subscriber count. Turn it off and you get the search-page fields only, faster.
The Python
from apify_client import ApifyClient from statistics import median client = ApifyClient("YOUR_APIFY_TOKEN") run = client.actor("locaihost/youtube-scraper").call(run_input={ "searchTerms": ["web scraping python"], "searchType": "videos", "uploadDate": "month", "sortBy": "popularity", "maxResults": 100, "videoDetails": True, }) videos = list(client.dataset(run["defaultDatasetId"]).iterate_items()) views = sorted((v["viewCount"] or 0 for v in videos), reverse=True) print("median views:", median(views)) print("top 10 share:", sum(views[:10]) / sum(views)) Each row has the title, URL, views, likes, comment count, duration in seconds, publish date, chapters, tags, hashtags, links in the description, and the channel's name, handle and subscriber count.
1. One video took 55% of the month's views
The 100 videos had 64,452 views between them. One of them, a 3-minute tutorial from a channel with 274,000 subscribers, had 35,398 of those. The top 10 videos had 83%.
| Views | Videos |
|---|---|
| Under 1,000 | 92 |
| 1,000 to 9,999 | 7 |
| 10,000 or more | 1 |
The median video had 92 views, and that's with the results sorted by popularity. A month-old video in this niche with a few hundred views is doing better than most of its peers.
2. Small channels stay small, for now
97 of the 100 rows came back with a subscriber count. The median channel had 590 subscribers.
| Channel subscribers | Videos |
|---|---|
| Under 1,000 | 53 |
| 1,000 to 9,999 | 21 |
| 10,000 to 99,999 | 19 |
| 100,000 or more | 4 |
None of the 53 videos from channels under 1,000 subscribers reached 1,000 views in their first month. Their median was 73.
3. Posting more is not the same as reach
One company channel, Scrapfly, published 16 of the 100 videos, more than anyone else. Together they had 2,369 views, a median of 97 each. That's about the same as everyone else's median: volume got them into the list, not into the top of it.
4. Length, chapters and "AI" in the title
| Length | Videos |
|---|---|
| Under 3 min | 13 |
| 3 to 10 min | 50 |
| 10 to 20 min | 23 |
| 20 to 60 min | 13 |
| Over an hour | 1 |
The median video ran just under 7 minutes. 45 of the 100 had chapters, and those did slightly better (median 99 views against 85). 14 titles mentioned AI, Claude, GPT, LLMs or agents, and they did slightly worse (median 83 against 92). With 100 videos these are small gaps, so read them as "no clear lift" rather than a rule.
The most common title words after the search terms themselves were scrape (16), data (13), API (12), tutorial and Google (11 each), and 2026 (8).
Honest limits
- This is YouTube's ranking, not ours. We return what YouTube's search shows for that phrase and filter. A different phrase ("python scraper", "beautifulsoup") gives a different 100.
- Views at one moment. Counts are as of the run. A video uploaded yesterday hasn't had its month yet; filter by publish date if you want a fair comparison.
- Subscriber counts are YouTube's rounded figures (274K, not 274,113).
- No watch time or revenue. Only the public counts any visitor can see.
Price
YouTube Scraper costs $0.50 per 1,000 videos or channels on the Starter tier, less on Scale and Business, plus a one-time start fee of $0.00005. Comments are an optional extra event. It needs no YouTube API key and has no daily quota. Try it here: apify.com/locaihost/youtube-scraper.
