YouTube Changed What a View Is and Tracking It Is Essentially Useless Now.

TL;DR

• YouTube's new "view" counts from the very first frame, including when a video autoplays because someone hovered over or scrolled past a thumbnail.

• What used to be called a "view" (watched past the first frame by clicking to watch) is now called an "engaged view" and it's still what the algorithm actually cares about.

• CTR, average view duration, and retention calculations haven't changed. They're still calculated using engaged views.

A lot happened after I posted that first piece less than two weeks ago. You may have heard that YouTube made a big announcement on changes to what a "view" means.

If you recall in my last post about the dangers of YouTube paid promotion, I hammered on how "views" are far overvalued already and should rather be regarded as "impressions". Allow me to quote myself here.

Somewhere along the way, we started equating raw numbers like subscribers and views as the qualifying metrics for any goal. And whether you agree with it or you also believe that metrics are kinda broken, you live in this reality where too many people do agree. An executive asks your team to make a video and just wants to see a high view count, so you post it to a dead channel, pay for promotion, get the big number, send the result, and everyone's happy and then it's done. Afterwards, you question yourself: What impact did that actually have? Often the answer is… none.
— Jackie Lamport, Aug. 10th, 2026

You can read the full post here: Zombie Channels: How Paid Promotion Kills YouTube Audience Growth. I will still, as promised, explain the solution to how you can actually track a YouTube Paid Promotion campaign's impact with the right metrics, and provide a custom-built calculator for paid promotion channel impact specifically. That will, however, have to be in a separate blog that I will post in a few days time as I first needed to digest and respond to the recent changes.

You see, YouTube has done me a favour. In talking about this, my intention was to educate brand channels on the misadventures of prioritizing the wrong metrics, and more importantly, misunderstanding them, as well as the lasting compounding effects that can have on a channel. This update to how a "view" is actually measured greatly emphasizes the point I was making.

Let me again point to the language I used when I called a "view" an "impression". Funny enough, YouTube has made the distinction so thin that they had to write an article explaining themselves. I kid you not, I let out an audible chuckle when I read this line:

"What's the difference between a view and a thumbnail impression?"
— YouTube Official Blog, Aug. 19th 2026

Okay, I'll stop being sassy about this and explain. I am still just appreciating the comedic timing of this update falling in between the two blog posts I planned.

What Actually Changed in YouTube's View Definition?

In YouTube's words, a view used to mean someone who watched past the first frame of a video by clicking to watch it. That's now called an Engaged View. The new view counts from the very first frame, regardless of how the person was exposed to that content.

The big distinction here is that in one case, someone clicks to be exposed to the content, and in the other case, they don't need to actively choose to watch your video.

In my previous article, I argued that by nature of the discovery behaviour of the platform, along with what a meaningful exposure should entail (not a few seconds or single frame), a view is already a vanity stat that shows little to no real intent, exposure, or audience nurturing and growth. The first few seconds of a video should be produced with the idea that it is a "teaser" for the actual video. A person who has clicked has not yet decided to stay. If content gets ton of clicks but no one stays, and you care about people watching your videos, it doesn't matter how big the number under that video frame is.

The length, or lack thereof, is why I argued that views should be seen more as impressions in the first place. That is still my take, and this new version of views only makes the need to change perceptions on views more clear.

The most important thing to understand here is YouTube's autoplay feature. When you are on YouTube, scrolling, searching or what not, if you hover over a thumbnail, the video will start to autoplay. That is now counted as a view.

Is now a "view".

Because of how slim the new difference between a view and an impression is, in its article about the update, YouTube had to make it clear that they are in fact different.

Basically, it counts as a view when a person sees a thumbnail and "engages" with it by hovering over it. Even if that view is as short as a single frame. So if someone moves their mouse, or scrolls by, with no intention of actually seeing the video, but it autoplays at all, it is a view.

It counts as a thumbnail impression when a person sees a thumbnail, and the autoplay is not activated.

Regardless of what the metrics are called, this now effectively makes it official that a view is really an impression. And impressions are easier to get than views. The line between a view as a useful stat and a vanity metric has always been a bit blurry, and it seems that YouTube is leaning into that. Ya, YouTube may be updating what a view means externally, but the telling part is how they view it internally.

YouTube Doesn't Value the New "Views"

The new "view" doesn't move your algorithm. That much YouTube's confirmed. Whether it moves anything else, I couldn't tell you.

The way clicks and taps are counted hasn't changed, and the vast majority of analytics remain anchored on engaged views, so the way metrics like CTR, AVD, and retention are calculated hasn't changed.

The way YouTube determines if a video is worth pushing is already based on these engagement style stats. This is why paid promotion campaigns can negatively impact your channel or growth, because bad/unfit views = bad engagement stats, and bad engagement stats = algorithm death. So, high view count but poor engagement and performance naturally indicates either bad views (the wrong audience), or content that isn't engaging enough. This update skews that though.

Now, since a view that comes from an autoplay still counts, the relationship isn't as clear. The good news is that this means the inflation in views that is inevitable shouldn't have negative impacts on the algorithm. In the same direct way, at least. But this shows how little YouTube values this kind of impression. They are still prioritizing the "engaged view" (old view), because it shows clear intent, and gives real signals.

Developing an audience, building trust, and having meaningful impact derived from your branded content won't come from, frankly, superficial and vague metrics. Analytics are indicators. They don't build relationships. Making content should never come from a perspective of "we need big numbers". It should come from you understanding what your audience wants, what you can give them, and how you can present that to them in the most engaging way so you can build a continuing relationship.

If you're reading this today, then in a few days time I will have the next instalment with the interactive calculator. Sign up to the newsletter if you don't want to miss that.

FAQ

Does YouTube's new view count affect my channel's algorithm ranking?

No. YouTube made it clear in their article that there is no change to recommendations and no change to the algorithm. Engaged views are still being used to calculate CTR, average view duration, and retention in the same way they were before.

What's the difference between a view and a thumbnail impression now?

Almost nothing. A view counts if a thumbnail autoplays because someone hovered over or scrolled past it, even if it was just for a single frame. A thumbnail impression counts if autoplay never triggers.

Should I still track views as a KPI?

Not on their own. Views just got easier to inflate, and YouTube itself is signalling that it doesn't weight them the way it weights engaged views. Track CTR, average view duration, and retention instead. Those are what actually drive whether your content gets pushed. I go deeper on what these mean specifically for branded podcasts in this article: YouTube Analytics for Branded Podcasts.