Designs Valley

From Data to Strategy: Advanced Analytics for Video Marketers

Illustration of advanced video analytics displayed on a computer monitor.

Video is arguably the best and certainly the most popular medium for sharing information compellingly with your audience. Video content has a wide reach and can be offered in multiple ways, including television, social media, websites, advertisements, or even slide decks.

Thanks to technological improvements, the barrier to entry for creating video content is lower than ever—many smartphones can capture high-definition video. To effectively optimize video content for your audience, you need to keep an eye on how you can optimize and otherwise improve your videos. One way this is done is by analyzing the data from views of your videos. This might include user engagement, user experience, or conversions attributed to your video.

In this article, you’ll learn more about video analytics. You’ll look at some traditional analytics and see how they might not be the most useful tools for many modern use cases. You’ll also learn more about more useful alternatives to these analytics.

Traditional Video Analytics

Platforms that host video content, such as YouTube, usually provide analytics based on your audience, views, and user engagement. Data on your audience could include the number of impressions, which is how many users were presented with the video, and reads, which is how many people watched it. Other information might include how the user arrived at your content, such as through a marketing campaign or playlist.

Data on your views could include:

  • View count, which is the total number of times your video was watched.
  • Play rate is a measurement of how many people were presented with an impression of your video compared to the number who watched it.
  • Completion rate, or the number of people who watched the whole video.
  • Play time, or how long users spent watching the content.
  • Drop-off points are the points at which people commonly stop watching the video.

Common metrics for assessing user engagement include your video’s number of likes, comments, and shares. For videos that include links or call-to-action buttons, user engagement metrics could also include a click-through rate, which is the percentage of viewers who performed your desired action.

All these metrics could help optimize your marketing and video presentation and conversions from your video content. However, if you’re trying to perform an in-depth analysis of your video or trying to optimize for some of the issues that end users of your video content might face, they might not give you the data you need.

Common Video Problems

While it may seem like video is everywhere, many issues may disrupt the user experience. For example, when streaming videos, issues with the video resolution may lead to blurry, unfocused content. On the other hand, if your videos are excessively large, they might take up too much of the user’s bandwidth, leading to excessive buffering periods, slow starts, and lag when seeking. All this can contribute to a frustrating user experience and customer churn—and traditional analytics won’t tell you anything about it.

Product managers and other professionals interested in optimizing user experience would benefit from more quantitative information on this problem, which would allow them to understand better if these problems are caused by the video’s encoding or protocols, the user’s internet connection, or their device. This would enable in-depth root-cause analysis and A/B testing to optimize the video pipeline process.

There’s also room for significant improvement in understanding your video’s audience and their behavior. For example, looking at traditional view metrics, your video could have been auto-played due to the user’s device settings and immediately shut off by the user, but it could still be added to your view count. The completion rate is another metric prone to inaccuracies: Is completion measured if the whole video is watched or if the user reaches the end even if they skipped half of the video? Another consideration is playback time. If users are consistently replaying or rewinding your video, this could point to issues in the content, but this metric is often tracked inaccurately—or not at all.

Apart from technical video concerns, content creators and video catalog owners should be able to cater to a diverse audience by understanding their demographic differences and using this information to improve the user experience.

Advanced Video Analytics

Advanced video analytics allow you to understand your video’s performance better and return on investment. They provide more granular data on your audience’s preferences and their viewing habits, allowing you to create customer segments across device types, geographic locations, and personal interests.

Latency Metrics

Latency, measured in seconds or milliseconds, is how long the video is broadcast to reach the user. Specific types of latency can include how long the player takes to initialize, how long it takes for a video to start playing, and how long it takes for playback to resume if the viewer moves forward or backward in the video. High latency is frustrating for end users and detracts from the viewing experience. In some cases, it can cause you to lose viewers entirely. 

One common cause of latency is an insufficiently widespread content delivery network. If you’re only serving content from one or two locations, users further from those locations will experience higher latency. Another cause is video that isn’t optimized for delivery, such as inappropriate encoding settings or video protocols. 

Rebuffering Metrics

Most people have experienced video buffering, which is when a video partially loads before it begins playing. This is a crucial part of online video delivery and helps enable uninterrupted playback, even if the viewer’s connection experiences a brief interruption. As the video plays, the player will retain a buffer, downloading video segments slightly ahead of the user’s present position in the video. When new segments can’t be downloaded as fast as the video is playing, this buffer can run out. If this happens, playback will stall as more of the video downloads, a process called rebuffering. 

Rebuffering interrupts the viewing experience and is a strong contributor to video abandonment. It can indicate problems with the delivery network, content that isn’t optimized for viewers on lower bandwidth connections, or several other problems.

Rebuffer metrics can include information such as how many instances of rebuffering there were across all playbacks in a given timeframe, how long viewers spent waiting for videos to rebuffer, and the number of times videos rebuffed across all video views.

Playback Metrics

Playback metrics measure how frequently users do—or don’t—finish watching your videos. They can also include information about the cumulative amount of time viewers spent watching your videos, how often users click play and then abandon the video before the player even initializes, and the percentage of users who experienced an error during video playback.

Another especially useful playback metric is drop-off points, the point at which viewers stop watching the video. While users often don’t watch the entire video, if half of your viewers stop watching at the twenty-percent mark, it’s a clear sign that your content needs to be optimized for audience retention and user engagement. 

Video Quality Metrics

The last group of advanced metrics to be discussed is video quality metrics. Ideally, the video data sent to your users should be optimized for the screen on which they view your content, but this isn’t always true. Bandwidth wastage represents the bandwidth difference between the content you’re sending and what your user receives and indicates that the video size is larger than necessary for the user’s screen, which can create additional expenses for both the content provider and the user. 

Upsampling, on the other hand, is effectively the inverse of bandwidth wastage. Upscaling indicates that the video being delivered is smaller than optimal for the user’s screen, and the video is being automatically enlarged. While this doesn’t have the cost implications of bandwidth wastage, upsampled videos will be blurry and pixelated, resulting in a poor user experience. 

Why You Need Better Video Analytics

Traditional systems for video metrics can be difficult to set up, and extracting meaningful data to generate advanced analytics is another hurdle. Traditional metrics often fall short in helping you understand your users or viewers, optimize their experience, and retain them as users.

With more information about your user’s viewing experience, you can explore how your video is experienced by the user, including the video quality and any stops or pauses due to streaming problems or user action. This knowledge allows you to fine-tune your delivery to create a seamless streaming experience for all your users, ensuring they see your content at the quality you intended.

Advanced video analytics also offer you valuable information about your content itself. Your completion rate could be low not because of a lack of engaging content but because your video takes too long to start and buffers for extended periods. Information about drop-off points can demonstrate where content could be clarified or show what the viewers are looking for in your video. Similarly, information about video replay can point to content that could need to be simplified or expanded. It’s possible to perform A/B testing targeting user engagement metrics, completion rate, or playback time to optimize viewer experience and present the best version of your content to your viewers.

Video delivery platforms don’t offer these metrics by default, so to get this information for content hosted by a third party, you’d need to set up secondary platforms in your video pipeline to serve and optimize these features. Independently hosted videos, on the other hand, can be served by a video delivery platform such as Gumlet, which integrates advanced metrics into the platform.

Conclusion

Better video analytics provide you with data that directly quantifies the user experience. This information can be used to optimize the content and delivery of your videos, ensuring that the end user sees the more optimized version of your content, no matter what sort of device or connection they’re using. 

Scroll to Top