YouTube long-form videos are quietly turning into Shorts. The data isn't subtle.
A Metricool study published August 3, 2026, covering 799,718 videos across 71,177 US accounts, found something that looks good on the surface and isn't: long-form views are up 76% year over year. Underneath that headline number, average view duration on those same long-form videos fell 37%, from just under four minutes to about two and a half, engagement dropped 45%, and estimated ad revenue fell 55%. More people are clicking long-form videos than ever. They're leaving faster than ever too.
Views up, everything else down
| Metric | Year-over-year change |
|---|---|
| Long-form views | +76% |
| Long-form watch time | +11% |
| Average view duration | -37% (3.98 min → 2.51 min) |
| Engagement | -45% |
| Ad impressions / monetized playbacks / est. ad revenue | -51% / -59% / -55% |
Source: Metricool, August 3, 2026, as reported by Tubefilter.
Watch time only grew 11% against a 76% jump in views, which is the whole story compressed into two numbers: far more people are starting a long-form video, and far fewer of them are staying anywhere close to the whole way through. A video can post record view counts and still earn meaningfully less than it used to.
Long-form hasn't lost ground to Shorts. It's absorbed Shorts' behavior.
Shorts traffic more than doubled over the same period and now accounts for 61% of all measured views platform-wide, which points at the likely cause: the Shorts feed has become the dominant way a viewer discovers a channel in the first place. A viewer who arrives at a long-form video from a scroll-based feed brings scroll-based patience with them, clicking in and bailing within the first stretch if nothing grabs them immediately, the exact behavior our hooks guide already treats as decisive for retention. That behavior used to be a Shorts-specific problem. The data now says it isn't.
What actually holds up against this
Front-loading payoff matters more now, not less, and a weak opening costs more than it used to given how much faster the average viewer is leaving. This doesn't overturn the reasoning in our Shorts vs long-form guide, long-form still carries the ad revenue and watch-hour math that matters for monetization, but it does mean the retention discipline that guide already argues for isn't optional anymore. Check your own Audience Retention graph against this pattern before assuming a specific video's underperformance is about the topic rather than the pacing.
Where Thothium fits
Thothium keeps a video's opening scenes as editable as everything else, so tightening a hook after checking your retention graph is a targeted fix rather than a full re-cut. It is in free alpha, and the form below gets you a key.
Frequently asked questions
Does this mean I should stop making long-form videos?
No. Long-form views and watch time are both still up year over year, and the ad-revenue split still favors long-form. What's changed is the shape of a typical view: shorter, more scroll-like, closer to how a viewer treats a Short. That argues for tightening pacing and front-loading payoff, not abandoning the format.
What's actually driving the drop in average view duration?
The most likely explanation is discovery: Shorts now account for 61% of measured views platform-wide, and the Shorts feed has become the main way viewers stumble into a channel. A viewer arriving from a scroll-based feed carries scroll-based habits into a long-form video, clicking in and bailing fast if the opening doesn't hold, the same behavior that made hooks matter in Shorts now showing up in long-form numbers.
How does this affect ad revenue specifically?
Directly and negatively, per the same study: ad impressions fell 51%, monetized playbacks fell 59%, and estimated ad revenue fell 55% even as raw views climbed. Mid-roll ads depend on a viewer staying well past the opening, and a viewer who leaves at the two-and-a-half-minute mark on average is seeing fewer of them than one who used to stay closer to four minutes.
What should a faceless channel do differently in response?
Treat the first thirty to sixty seconds with the same urgency Shorts creators already do, since that's now effectively competing with scroll-away behavior even inside a long-form upload. Beyond that, watch your own Audience Retention graph rather than reacting to industry-wide averages; the shape of the drop-off, not just its existence, tells you whether the fix is the hook, the pacing, or the topic itself.
Last updated August 31, 2026. Statistics sourced from Metricool's August 3, 2026 study of the US creator economy, as reported by Tubefilter. Aggregate platform trends may not match any individual channel's performance; check your own Analytics for a current picture.