YouTube Research Reports
Most YouTube statistics you find online have no source, no date, and no method. These reports have all three. Each one is built from live OutlierKit API pulls, states the exact query and the day it ran, and links every outside claim to the page it came from.
Published reports
YouTube RPM Benchmarks by Niche: Every Published Range, Compared and Sourced
Published RPM figures for the same niche disagree by up to 7x, and most sources quote CPM as if it were RPM. Every benchmark table side by side, what YouTube officially defines, and the view-supply data that decides earnings.
YouTube Trending Niches Report: Where the Outliers Actually Are, August 2026
We measured how often videos in six niches beat their own channel by 10x or more across 677 live videos. Faceless automation hit 3.1 percent. Personal finance hit 16.7 percent.
How Many YouTube Views You Need to Make $1,000, $2,000 or $5,000 a Month
The honest answer is a range, not a number. Views needed at every published RPM, the Partner Program bar you clear first, and how many videos a month that means in four measured niches.
The Sleep and Ambient Audit: A 15.8 Percent Outlier Rate That Means Almost Nothing
Sleep and rain sounds scored the second highest breakout rate of any niche we measured. 58 percent of those breakouts came from channels under 1,000 subscribers, one with 7.
How these reports are built
Every report follows the same four rules, because a number you cannot check is worth nothing in a client deck or a board pack.
- Dated at the source. Every first-party figure carries the date it was pulled. View counts move, so an undated view count is a guess.
- Query published. The exact search that produced a sample is printed next to the sample, along with how many channels it returned.
- Outside claims are linked and attributed. Where another site publishes a figure, we name the site, link the page, and say it is their number, not ours.
- Limits stated. Each report ends with what it cannot tell you. Watch time, revenue, and audience demographics are private to the creator, so no outside tool can report them as fact.
The data comes from the OutlierKit API, the same endpoints any Pro or Max account can query directly. If you want to reproduce a table in one of these reports, you can run the same query yourself.
Who these are for
Anyone who has to defend a number to someone else. Agencies building a trend deck, research teams sizing a category, investors looking at a creator business, and creators deciding where to spend the next six months of their life.
If you want the underlying data rather than the write-up, the same figures are available through the API and the OutlierKit MCP server, so you can pull them straight into your own model or notebook.