Getting YouTube Views Is Changing: What Marketers Need to Know

Wondering why your YouTube view counts jumped even though your leads and sales didn’t move? Want to know which YouTube metrics actually matter now and how to get your videos found in Ask YouTube, ChatGPT, and other AI search tools?

​In this article, you’ll discover how YouTube’s new view count and engaged views work, which YouTube analytics to track instead, and how to optimize your videos, chapters, and Shorts for search and AI discovery.

#1: How YouTube Now Counts Views vs Engaged Views

YouTube’s view metric is now close to an impression. If someone scrolls past a long-form video and it autoplays for a split second, YouTube counts that as a view. What YouTube used to call a view is now called an engaged view, meaning the person showed some interest in the video rather than idly swiping past.

​According to Justin Brown, the change is YouTube playing catch-up with other platforms. For a long time, creators who published the same video on several platforms noticed that YouTube reported fewer views than TikTok or Facebook, because the platforms count views differently. On some platforms, swiping past a video has counted as a view for years.

​Under the old system, YouTube never published a precise definition of a view. Rumors put the threshold at 30 seconds, but Justin says that was never a good metric because it couldn’t work for short videos. A view was simply someone who had watched enough of the video to be considered significant, which might mean a few seconds on one video and a large portion of another.

What Counts as an Engaged View on YouTube

YouTube hasn’t laid out exactly what an engaged view is either, and Justin suspects the platform deliberately left the definition vague. Google owns a web browser, operating systems, Chromebooks, and Android phones, so it can draw on signals such as whether the browser tab is open or whether the person watched a previous video and is more likely to watch the next one when it autoplays. Justin believes those behavioral signals now carry more weight than the percentage of the video watched.

​His best guess is that an engaged view depends on video length and starts at a couple of seconds, enough for someone to decide whether the video is for them. From there, the count should depend on what happens next: a viewer who leaves immediately probably shouldn’t count, while one who sticks around should. He also allows that YouTube’s own AI systems may be making these calls in ways the company itself can’t fully explain.

​Autoplay is also changing how people watch. Videos that didn’t autoplay before now autoplay on TV, mobile, and desktop. Justin has caught himself halfway through videos in the mobile app that he never tapped on because they started playing as he scrolled. 

YouTube now plays those previews from the start of the video, and if he scrolls away and comes back, playback resumes where he left off. Because YouTube’s algorithm understands a viewer’s behavior and preferences, Justin sees this autoplay experience driving more engaged views, not just more views.

What the New View Counts Look Like in Practice

On the Primal Video channel, view counts jumped noticeably the day the change took effect, and the baseline is now higher. 

A video targeting search without a clickbait title used to earn a few thousand views in its first 24 hours. That number is now closer to 5,000-7,000 views. Engaged views, by contrast, are roughly on par with where they were or up slightly. Justin treats the higher view counts as a new baseline rather than a performance gain, because the backend tracking his business relies on hasn’t moved much.

​The rollout also appears uneven. Justin sees different view counts on desktop than in the mobile app, particularly for live streams, which he notes is typical for how YouTube rolls out changes.

#2: Which YouTube Metrics Matter for Business Growth

For a business, the most important YouTube metric is the one tied to the video’s goal. Justin says views are only one small piece. Before judging a video, he recommends deciding whether its job is to build an audience, grow the email list, or drive sales. For businesses building a brand, the goal is generating leads, sales, and customers, and Justin works backward from that.

​Views feel good, but they don’t say much about business results on their own. Among the available YouTube analytics, engaged views (the equivalent of the old view count), watch time, and click-through rate provide a clearer picture of whether the right people are clicking and watching.

How to Read Watch Time and Retention Graphs

Justin considers every metric worth looking at, but says each has to be read on a case-by-case basis. 

Comparing average view duration works when the videos are similar, such as a recurring series. On a channel like Primal Video, where one week’s video covers editing software and the next covers filming with an iPhone, a longer watch time on one video doesn’t mean viewers got more value from it. A viewer who watched for a shorter time may have gotten exactly what they needed, solved their problem, and become a loyal returning viewer.

​Rather than spending hours in YouTube analytics, Justin recommends a forward-focused review built around outliers. He identifies the top- and lowest-performing videos over a period, whether measured by engaged views, views, or watch time, then compares their retention graphs to find where people drop off.

​A video that underperformed might have lost a large share of viewers at the start, while the people who stayed watched the rest. That pattern points to one of two problems: a weak hook, or YouTube showing the video to the wrong audience, who landed on it and left right away. No single metric explains performance in isolation, so Justin looks at average view duration, watch time, and retention together to understand what’s happening for people who watch past the first 30 seconds.

Why Day-One Performance Doesn’t Predict Long-Term Results

How a video performs initially doesn’t represent how it can perform over time, especially for channels targeting YouTube search. Primal Video doesn’t get big numbers on day one or in week one, even with its channel size. Over time, videos filter into YouTube search and, increasingly, AI-driven traffic, and then they snowball. Some of the channel’s biggest videos tanked at launch.

​Because of this, Justin usually leaves an underperforming video alone and moves on. Re-editing has also become riskier. The Primal Video team used to shoot multiple hooks, publish the video, and use the YouTube editor to trim out the versions that didn’t work, with the option to restore them later. Now, edits to a published video can only remove footage. Nothing can be added back, and once an edit is saved, there’s no undo.

​Pro Tip: The YouTube editor can still be used to test a longer intro. Justin’s approach is to publish the video with the extended section and, if it doesn’t work, trim it out with a clean cut. Justin says viewers watching free content rarely notice a jump cut or a skip in the music.

What the Views (Co-Viewed) Metric Means for Marketers

YouTube also added a metric called views (co-viewed), which estimates when more than one person is watching a video together, such as a family watching on a living room TV. YouTube bases the estimate on factors such as demographics, video genre, and time of day.

​Justin doesn’t recommend changing content to drive group viewing, such as asking viewers to invite a friend or spouse. From everything he’s seen, the metric is still an estimate on YouTube’s end, and shared family accounts on TVs already skew the data. Until YouTube has a more reliable way to track co-viewing, he doesn’t see it as a meaningful lever for content strategy. 

Where co-viewed data has value is in brand deals and sponsorships, because it can show that a creator’s reach is larger than the view count alone suggests, especially for family-oriented content.

#3: How to Turn More YouTube Views Into Engaged Views

Justin expects the ratio of views to engaged views to become the metric that creators compare, replacing the traditional comparison of impressions to click-through rate. Because the view metric is so new, it’s too early to set benchmarks. But converting a view into an engaged view still comes down to packaging: the title, the thumbnail, and especially the start of the video.

​When a video autoplays, the thumbnail matters less because the viewer may never see it. What happens in the opening seconds matters more. This isn’t entirely new. When YouTube first introduced autoplay in the feed, the Primal Video team saw that people were watching silently without clicking, a behavior Justin calls “toilet TV.”

How to Add a Baked-In Title to the Start of a Video

Justin’s response was to put text on screen at the start of every video saying what the video is about. This isn’t the closed captions viewers can turn on or off. It’s a title burned into the video, such as “Best Editing Software Mac 2026,” so anyone who sees the video autoplay knows exactly what it covers, even if they never saw the thumbnail or title.

​The text appears for roughly the first 3 seconds as the action starts on screen, stating the video’s promise: the outcome or transformation the viewer will experience. Justin says this made a significant difference for Primal Video when autoplay rolled out because it moved viewers into an engaged state. A talking head without on-screen text makes it hard for a silent viewer to work out what the video is about.

​Pro Tip: On-screen title text doesn’t require a specialized video editor. Justin describes it as simple on-screen text stating the video’s promise, which any editor can add.

#4: How to Optimize YouTube Videos for Ask YouTube and AI Search

YouTube has been rolling out a feature called Ask YouTube, starting on TV and expanding to mobile and desktop. 

It lets users have a conversation with YouTube to explore complex topics beyond traditional keyword search, and responses can include text along with relevant videos, Shorts, and video clips. YouTube says creators can improve their chances of appearing in Ask YouTube results by continuing to create unique, high-quality content and by adding clear video chapters and descriptive titles.

​Justin sees Ask YouTube as part of a broader shift. People expect a faster time to outcome, whether they want to learn something or be entertained, and AI tools have raised those expectations. He expects viewers to increasingly jump to specific sections of videos instead of watching the whole thing. The hope is that a short clip or summary leads viewers to watch the full video, the same way Google’s search snippets help some searchers on the spot while still driving traffic to videos.

​AI search is already a meaningful traffic source for Primal Video. The channel’s number one traffic source is YouTube search, and its second is Google search, from people who never started on YouTube. The third, and the fastest growing, is AI tools. ChatGPT and Claude are sending viewers to Primal Video’s YouTube videos, and the channel regularly receives comments from people saying ChatGPT sent them there.

Why the Content Itself Is Now the Main YouTube SEO Signal

Keyword research hasn’t disappeared. Primal Video still looks at search volume to gauge demand before making a video rather than publishing and hoping. But getting keywords in the right order matters less than it used to. 

YouTube has transcribed what’s said in videos for a long time, and now AI lets it understand what’s actually happening in the video too. The longer a video is live, the more YouTube learns from watch time which parts viewers find compelling, and Justin expects YouTube to direct viewers to those sections.

​That means YouTube optimization now comes down to the viewer experience. Justin focuses on what viewers want, what they’d type into Google or YouTube to find the video, and what would grab their attention even if they weren’t searching for it.

​As quick answers increasingly come from AI, Justin believes that the part of a video that AI can’t replace is the human part. A viewer can get a banner size or a setting from AI, but they want to hear how a specific person would approach it, what choices that person would make, and why. Sharing opinions, experience, beliefs, and frameworks gives viewers something no one else can offer. 

Justin says people are already burning out on AI slop and that he leaves as soon as he hears an AI voiceover, which is why he expects human-led content to become more valuable, not less.

How to Write YouTube Chapters for Search and AI Discovery

Primal Video treats chapters the way it used to treat titles. For each chapter, the team asks what someone would type to find that section if it were a stand-alone video, researches the keywords and search volume, and titles the chapter accordingly. The timestamps and chapter markers in the video description remain a place to optimize for both YouTube and Google search, and each one is a direct link to that moment in the video.

​Instead of writing chapter titles as hooks that tease what’s coming, Justin recommends titles that state what each section actually covers. For a video on five things AI can edit well, each chapter would name the specific thing. Revealing the answer in the timestamps helps YouTube surface that section, gives AI tools like ChatGPT a direct path to the answer, and keeps it from being buried in a cryptic title viewers may or may not click.

​Justin frames search-optimized chapters as extra fishing lines that help viewers find the content. That approach fits channels built for search. Primal Video teaches rather than entertains, so instead of competing with entertainment creators like MrBeast for home page placement, it aims to show up at the top when someone has a specific pain point. He sees the same opportunity for most business owners, who can share their opinions and talk about their products without competing for entertainment views.

​AI tools make chapter optimization much faster than doing it manually. Justin points to vidIQ, whose AI tools generate titles and chapter markers based on search volume. The Primal Video team has connected vidIQ to its AI assistant through an MCP connection. Once a video is finished, the team feeds in its VTT or SRT caption file, which includes timing for everything said, and asks the AI to generate chapters based on the keyword phrases around each section while keeping the titles human-readable instead of stuffing in keywords.

​Pro Tip: Even with YouTube’s automatic chapters enabled, Justin overrides them with his own optimized chapters. The result can be a mix of the two, but he doesn’t rely on YouTube’s automatic chapters alone.

​Justin acknowledges there’s no stable ground here and that his approach is the result of long-term testing. For any individual channel, he suggests testing a handful of videos with more engagement-based chapter titles against search-based ones to see what works.

#5: Why Custom YouTube Shorts Thumbnails May Not Be Worth the Time

YouTube now offers a built-in AI thumbnail generator and lets creators upload custom thumbnails for YouTube Shorts. Justin isn’t creating custom Shorts thumbnails. Most Shorts views come from people already watching Shorts who swipe up to the next one, where a thumbnail never appears. Views from channel pages are also lower than they’ve ever been, because most people find creators through search and recommendations. Given that, he doesn’t think the time spent crafting Shorts thumbnails pays off yet.

​Primal Video doesn’t generate many views from Shorts. The channel has published one long-form video a week for 14 years and repurposes short nuggets of value from those videos into Shorts, since that fits its workflow.

How to Choose a Thumbnail Frame for YouTube Shorts

Instead of a custom design, Justin picks a frame from the video that’s good-enough, avoiding frames where he’s mid-sentence or mid-word. He also makes sure the frame includes captions or text so viewers know what the Short is about. Shorts titles are either missing in some places or truncated so much that they can’t say much, so the frame has to communicate the topic on its own. He’d also rather see Shorts that feel raw and real than polished, especially with ads and AI-generated content now appearing in the Shorts feed.

​Shorts thumbnails may become more relevant for businesses that prioritize Shorts, since Shorts now appear in a shelf on the YouTube home page and in search results. One middle-ground approach discussed in the interview is having AI select the best shot from the video and add a short baked-in title at the top. Justin supports this approach because it’s still a real frame from the video rather than a highly polished bait-and-switch.

Other Notes From This Episode

• Connect with Michael Stelzner @Stelzner on Instagram and @Mike_Stelzner on X.
• Connect with Jerry Potter on LinkedIn and YouTube.
• Watch this interview and other exclusive content from Social Media Examiner on YouTube.

Where to subscribe: Apple Podcasts | Spotify | YouTube Music | YouTube | Amazon Music | RSS

✋🏽 If you enjoyed this episode of the Social Media Marketing podcast, please head over to Apple Podcasts, leave a rating, write a review, and subscribe.

---

Stay Up-to-Date: Get New Marketing Articles Delivered to You!

Don't miss out on upcoming social media marketing insights and strategies! Sign up to receive notifications when we publish new articles on Social Media Examiner. Our expertly crafted content will help you stay ahead of the curve and drive results for your business. Click the link below to sign up now and receive our annual report!