_edited.png)
On August 24, YouTube views go up. Revenue doesn't.
Starting August 24, 2026, YouTube is changing the way it counts views. The numbers displayed under every video will rise mechanically. Only one thing will genuinely shift behind that increase, and it is not the one most people expect.
What changes on August 24
The new rule fits in a single sentence: a view is now counted from the very first frame of the video. All it takes is for playback to start (source: official YouTube blog).
Until now, each format followed its own counting logic and required a minimum watch time. Comparing a long-form video, a live stream, and a Short was little more than guesswork.
The new system unifies everything. 📊 Shorts had already worked this way since spring 2025, so it is really the rest of the catalog that is falling into line with them. TikTok and Instagram use the same method, which finally makes the numbers comparable across platforms.
Why the counters will inflate mechanically
Part of the audience simply was not being counted. The viewer who opens a video, realizes within two seconds that it is not what they were looking for, and leaves did not exist in the statistics. Starting August 24, that viewer counts.
How much the numbers rise will depend entirely on the content. A video whose thumbnail promises more than it delivers will see its figures jump, precisely because it generates a high volume of immediate exits. A video that holds attention from the opening seconds will barely move.
By exactly how much? Nobody can say. Neither YouTube nor any independent study has published an estimate, and it will take until September before real figures are available.
More views, but no more revenue
YouTube has been explicit: the change will affect neither creator earnings nor eligibility for the Partner Program (source: official YouTube blog). Monetization continues to rest on engaged views and engaged watch hours, never on the public counter.
The engaged view is nothing other than the previous counting method, renamed: the viewer who watches past that first frame, or clicks to watch. It remains available in YouTube Analytics.
💰 A channel whose view count climbs sharply in September will therefore not earn a single cent more. The public counter measures exposure, not revenue.
What YouTube values is not the number of playbacks triggered, it is the time genuinely spent watching. One person who watches 80 % of a video carries far more weight, in the platform's economy as in its recommendations, than five people who drop off after two seconds.
Engagement rates, on the other hand, will genuinely fall
Engagement rate, as brands, agencies, and third-party tools calculate it, divides likes and comments by the number of views. If the denominator inflates while the numerator stays put, the result falls. Mechanically.
This is not a hypothesis. Both counting systems have coexisted for a year, Shorts on one side and long-form on the other. A study covering 1.46 million channels shows that within a single channel, the engagement rate on Shorts reaches a median of 78 % of the long-form rate (source: CreatorDB study).
⚠️ For a brand, the consequence is concrete. An advertiser who ties a creator's fee to public view counts will pay more for exactly the same audience. An influencer campaign measured on engagement rate will appear to underperform even though nothing has changed in its execution. Contracts signed before the summer are worth a second look.
Inside YouTube Analytics, by contrast, nothing degrades: average view duration, retention curves, and RPM are still calculated on engagement-based figures. The discrepancy only shows up outside the tool.
A clean break in the data
One final point. Views already accumulated are not recalculated: a video uploaded in June keeps the counter it displays today. However, every new view recorded from August 24 onward follows the new rule, including on those older videos.
August 24 is therefore a hard boundary in the data. Setting a figure from before against a figure from after means comparing two different definitions, and month-over-month or year-over-year trends will be distorted across it. For a comparison that holds up, engaged views are the metric to use.
Key takeaways
August 24, 2026: a view is counted from the first frame, across every format.
Public counters rise mechanically, with no real increase in audience.
Revenue does not move: monetization is still calculated on engaged views.
Engagement rates fall: around 78 % of the long-form level.
Inside YouTube Analytics, nothing changes.