YouTube Prices 10M Shorts Views the Same as 4,000 Watch Hours
YouTube Shorts pay creators an RPM of roughly $0.03 to $0.10 per thousand views, while long-form video on the same channel earns $3 or more. The gap is structural, not a niche problem: Shorts ad money goes into a shared Creator Pool, gets split by engaged views, and creators keep 45% of whatever lands in their bucket. Judge Shorts on the subscribers they feed into long-form, not on the payout line.
Most of the Shorts versus long-form debate gets argued as a revenue question, which is the least interesting version of it. If you run a brand channel you are not monetizing anything. You are spending. So the RPM spread only matters as a signal, and it turns out to be a pretty loud one about what YouTube thinks a swipe is worth compared to a watch.
YouTube already priced a Shorts view, and it isn't flattering
The clearest number nobody quotes in these arguments is sitting in the Partner Program eligibility rules. There are two doors into monetization, and vidIQ lays both out: the long-form path needs 1,000 subscribers plus 4,000 valid public watch hours over 12 months, and the Shorts path needs 1,000 subscribers plus 10 million valid Shorts views over 90 days.
Those two doors are supposed to be equivalent. YouTube built them that way. So run the arithmetic and you get YouTube's own exchange rate.
4,000 watch hours is 240,000 minutes. If your long-form audience averages four minutes of view duration, that's 60,000 views to clear the bar. If they average eight minutes, it's 30,000. Against 10 million Shorts views, you land somewhere between 165 and 335 Shorts views for every long-form view YouTube considers equally valuable.
Ten million Shorts views and thirty thousand long-form views open the exact same door.
I want to be careful here, because that ratio depends entirely on your average view duration and I'm estimating it, not measuring yours. But even at the friendliest assumption the order of magnitude holds. YouTube is not being coy about this. It published the conversion rate in the eligibility docs and then let everyone argue about RPM instead.
What to do with that: pull your last 90 days in Studio and compute subscribers gained per 10,000 engaged views, separately for Shorts and for long-form. In most channels I've seen written up, long-form wins that comparison by a wide margin, and if yours comes back inside 2x of each other you have an unusually good Shorts funnel and should be pouring more into it. If Shorts is more than 10x worse on subscribers per engaged view, the format is buying you impressions and very little else.
The Creator Pool is the reason your Shorts RPM won't move
Long-form ad revenue works the way media buyers expect. An auction happens, advertisers bid against your inventory, you get a cut of the winning bid. Your RPM is a function of who wants your audience.
Shorts doesn't work like that at all. Per YouTube's own monetization policy, revenue from ads running between videos in the Shorts feed is added together each month into a Creator Pool, then distributed based on "their share of total engaged views from monetizing creators' Shorts in each country." Get 5% of eligible engaged views in a market, get 5% of that market's pool. Monetizing creators then "keep 45% of their allocated revenue, regardless if music was used or not."
Sit with the mechanism for a second. Your Shorts RPM is not really yours. It's the platform average, adjusted for your country mix. You cannot out-niche it the way a finance channel out-niches a gaming channel on long-form CPMs, because nobody bid on your Short specifically. Making better Shorts gets you a bigger slice of the pool. It does not make the pool richer.
There is one lever, and it's smaller than it looks. The same docs spell out the music split: a Short with one licensed track sends half its associated revenue to the Creator Pool and half to music licensing, and a Short with two tracks sends a third to the pool and two thirds to licensing. Your 45% rate doesn't change. The thing being multiplied by 45% does.
So if you're stacking two licensed tracks on Shorts out of habit, audit the last 30 uploads and count how many have more than one. Getting that number to zero is maybe a 20 minute cleanup in your editing template.
And honestly, for most brand channels that cleanup is worth almost nothing in dollars. vidIQ's example is a Short with 468,500 views that earned $16.61. Half of nothing is still nothing. I'm flagging it because creator-side advice keeps presenting the music split as a real optimization and it mostly isn't, unless Shorts revenue is genuinely paying someone's salary.
March 2025 inflated every Shorts number in your deck
This one still trips up reporting a year later. On March 31, 2025, YouTube changed what counts as a Shorts view. TubeBuddy's writeup of the change is direct about it: every play, replay, and scroll-by now counts, with no minimum watch time. The old, stricter definition didn't disappear. It got renamed to "engaged views," and engaged views is still what drives monetization and Partner Program eligibility.
The practical effect is that a lot of channels booked a step-change in "reach" in early 2025 that was purely definitional. If your year-over-year Shorts deck crosses that boundary and uses the Views column, the comparison is broken. Not slightly noisy. Broken.
Swap Views for Engaged Views in every recurring YouTube report you own. It's a column toggle in Studio's analytics, so this is a ten minute fix, and it will make your numbers look worse, which is the point.
For a benchmark I'd hold loosely: if engaged views come in under about a third of raw views, your first second is losing people before the content starts. That's my rule of thumb from watching the two columns diverge, not a published figure, so treat it as a starting line rather than a standard. The useful part isn't the threshold anyway. It's that you now have a hook metric that isn't contaminated by scroll-bys.
The same discipline applies across formats, incidentally. We went through this with a million TikTok views converting to about 2,000 follows, and the pattern repeats on every short-form surface: the top-line number is generous and the downstream number is brutal.
Where I'd put the next 90 days of production budget
The fear that posting Shorts poisons your long-form distribution is the most common reason teams stall on this, and the evidence doesn't really support it. Search Engine Journal's interview with YouTube's growth and discovery team makes the point that the recommendation system evaluates each video on its own rather than penalizing a channel for its average. Todd Beaupré, who leads that team, framed the platform's Shorts push as following the audience rather than pushing them: "YouTube is focused on Shorts because audiences have let us there."
Beaupré's other line is the one I'd actually tape to the wall, and vidIQ has the cleanest version of it: same audience, same channel; different audience, different channel. That is the whole decision. If your Shorts pull the same people your long-form serves, one channel, and the recommendation system can bridge them. If your Shorts are chasing a broader, cheaper audience because short-form is where the volume is, you've built a second audience on the same URL and you should expect the bridge to carry almost nobody.
To be fair, this isn't entirely new advice. It's the same reach-versus-conversion split we looked at with Reels reaching strangers while feed posts convince them. It just feels sharper on YouTube because the revenue data makes the gap legible in dollars instead of vibes.
For a brand channel specifically, I'd run it like this. Long-form carries the offer, the demo, the comparison content, anything a buyer searches for on purpose. Shorts carry the top of the funnel and get judged on subscribers and engaged views, never on revenue. And the paid layer is where Shorts inventory actually earns its keep: agency-side reporting has put in-feed Shorts ad CTRs in the 1.0% to 3.0% range against a broader YouTube average nearer 0.65%, which is a vendor figure and worth verifying against your own account before anyone builds a forecast on it, but the direction matches what I keep hearing from buyers.
My prediction, with stakes: within about 18 months YouTube collapses Shorts and long-form into a single revenue line in Studio and the RPM comparison stops being possible to run at all. If Shorts RPM hasn't cleared $0.25 by the time that happens, the pooled model was never going to get there and the format's value stays permanently indirect.
Questions that keep coming up
Should Shorts and long-form live on separate channels? Only if they serve different audiences. YouTube's own guidance is same audience, same channel. Splitting a single audience across two channels means both of them look inconsistent to the recommendation system and neither builds a watch history worth much.
Do Shorts views count toward the 4,000 watch hour threshold? No. They're separate doors into the Partner Program, and Shorts views only count on the Shorts path, which needs 10 million valid views in 90 days. Mixing the two doesn't get you there faster.
Is a low Shorts RPM a sign something is wrong with my channel? Usually not. Because payouts come from a pooled distribution split by engaged view share rather than an auction on your inventory, most channels cluster within a narrow band regardless of niche. Country mix moves it more than content quality does.
What I'd stop arguing about
The revenue comparison is settled and it was never that close. What's still open is whether your Shorts are recruiting for your long-form or just running in parallel, and the eligibility math gives you a cheap way to check: if it takes hundreds of Shorts views to equal one long-form view in YouTube's own accounting, your Shorts need to be handing something off. Subscribers, a search, a returning viewer, anything.
I don't think the channels that win here are the ones posting the most Shorts. From what I've seen it's closer to the ones who figured out which long-form video each Short is supposed to be selling, and then went and measured whether it did.
