NEW in YouTube Analytics: Impressions for New & Returning Viewers

New Analytics and Live Streaming Updates from YouTube

YouTube has introduced several updates aimed at providing creators with deeper insights into their audience behavior and enhancing the real-time interaction experience during live streams. From more granular impression data to technical improvements in stream latency, these changes offer new ways for creators to optimize their content and engagement strategies.

Deeper Insights into Live Reactions

Live Reactions have long been a polarizing feature—some creators find them an engaging way for audiences to express themselves, while others find the visual noise distracting. Regardless of preference, YouTube is now introducing analytics for this feature.

For creators who keep Live Reactions enabled, you can now access data showing which specific reactions from the reaction wheel were used most frequently. More importantly, these analytics allow you to see exactly where in your stream these reactions peaked. By identifying the moments that triggered a surge of reactions, you can better understand which segments of your live content resonate most with your audience, providing a roadmap for what to do more of in future broadcasts. It is worth noting, however, that this feature is currently slated for an overhaul, so creators should keep an eye on how these tools evolve.

Segmenting Impressions: New vs. Returning Viewers

One of the most significant additions to YouTube Analytics is the ability to segment impressions by new versus returning viewers. While this feature is rolling out gradually and may not be available to all channels yet, it provides a powerful tool for A/B testing and thumbnail optimization.

Impressions represent how many times your thumbnail was shown to potential viewers. By splitting this data between new and returning viewers, you can draw more accurate conclusions about your packaging:

• Analyzing New Viewers: If you see a high number of impressions among new viewers but a low click-through rate (CTR), it suggests that while the algorithm is pushing your content to a wider audience, your thumbnail may not be appealing enough to those who aren’t already familiar with your brand.
• Analyzing Returning Viewers: Conversely, if returning viewers are clicking at a high rate but new viewers are not, your current visual style might be tailored too specifically to your existing community and may need more broad appeal to attract growth.

Creators are encouraged to check their YouTube Studio analytics to see if this segmentation is available for their channel, as it allows for much more targeted adjustments to thumbnails and titles.

Enhancing Engagement with Super Chat Likes

To further incentivize and acknowledge financial support during live streams, YouTube is rolling out “likes” for Super Chats. This update ensures that the community can interact not only with the creator but also with the supporters themselves.

The ability to like a Super Chat will be available both in the main chat feed and within Super Chat threads (the replies to specific Super Chats). This adds a layer of social validation for viewers who contribute financially, potentially increasing the visibility and appeal of these messages. According to the rollout schedule, this feature is being implemented from mid-April to early May.

Reducing Latency in Ultra Low Latency Streams

For creators who rely on real-time interaction—such as Q&As or interactive gaming—latency (the delay between when you speak and when the viewer hears it) is a critical factor. YouTube is currently conducting an experiment in Korea to further reduce this gap.

The goal of this experiment is to reduce latency in “Ultra Low Latency” settings by up to 30%. To put this into perspective, a one-second delay could be reduced to approximately 0.6 seconds. While a fraction of a second may seem negligible, it represents a significant technical step toward instantaneous communication. This stands in stark contrast to “normal” latency, which typically ranges between 20 and 30 seconds—a gap that makes real-time conversation nearly impossible.

While this test is currently limited to the Korean market, such experiments often serve as a precursor to wider global rollouts. If successful, this could eventually lead to a more seamless, “instant” feel for live streamers worldwide.

Original transcript

Transcript

Hey there! We’ve got some exciting updates from YouTube to kick things off. Let’s start with Live Reactions—some love them, some hate them. Personally, I find them a bit distracting, but they’re getting analytics now. This means that if you still use Live Reactions and they haven’t taken over your entire chat, you can now see which analytics wheel was used and where there were a lot of Live Reactions during your stream. I think this is a useful addition if you do use Live Reactions and haven’t disabled them yet, as they’re due for an overhaul.

Staying with analytics, let’s move on to new vs. returning viewers. Now, impressions can be segmented by these two categories, which I find super interesting. My channel doesn’t have this feature yet, but you can check your analytics to see if it’s available for you and maybe draw some conclusions: Do your thumbnails perform better with subscribers or new viewers? You might be able to say, “We’re getting a lot of views or impressions from new viewers, but the thumbnails could be better.” Maybe it’s time to adjust them. I’d love to hear your thoughts on this in the comments! I’m curious to see how this develops.

Next up, we have several updates in the live streaming space. Let’s start with Super Chats: Super Chat likes are now fully rolled out. This means that if you use Super Chats, they can now be liked both in your chats and in Super Chat threads (replies to Super Chats). This feature will be implemented from mid-April to early May. I’m keen to hear what you think about this. Do you use Super Chats? Do you stream a lot on YouTube and enjoy this feature? Share your thoughts with me in the comments!

Still on live streaming, there are significant improvements in latency. An experiment is currently being tested in Korea, aiming to reduce latency in Ultra Low Latency by up to 30%. This means that if we previously had a one-second delay, we now have 0.6 seconds. It might not seem like much, but it’s moving closer to normal latency, which typically ranges between 20 and 30 seconds. I’m curious to see when and if this experiment expands to other markets. I assume it will, and I suspect these experiments will continue to try and squeeze out a few more percentage points. Let me know what you think about this as well!

Catch you all next week for another update. Until then, take care and keep grinding!

Confused about these new YouTube updates? From Live Reactions analytics to impression segmentation, it's a lot to take in. Check out our Creator Services for expert insights and guidance on making the most of these changes.