Brandon Shepherd

Long-form creators on YouTube need to know which metrics are important to pay attention to and which ones aren’t. Here’s a long list of the most common metrics that have been talked about so far and which ones I believe are important when trying to judge a piece of content in an objective way. Impressions: With anywhere from 15% to 80% of impressions not being counted or being counted wrong, this number is not accurate enough to rely on. Views: Views, or views within a certain timeframe, are your most reliable metric on YouTube. Even though what YouTube counts as a view is still up to them, this number gives you the best idea of how well your video did. Click-Through Rate (CTR): We can't make decisions about individual content based on this metric because it has a low correlation with views as a performance metric. Average View Duration (AVD): AVD depends too much on how long your content is, but even when we compare videos that are about the same length, we don't find the correlation we need to make good decisions. Average View Percentage (AVP): Like CTR and AVD, this metric has a low correlation with views as a performance metric. Because of this, I would never suggest that someone use this metric to judge how well a piece of content is doing. However, when trying to read and understand retention charts, relative average view percentage becomes much more important when interpreting the significance of a drop in the middle of your retention charts. Watch Time: The watch time is affected by both the number of views and the average length of each view. It's a metric we can use in part to try to figure out how well brand integrations convert. But there are better ways to figure this out because it doesn't tell us how many people didn't listen to the ads. In this case, it's best to use both views and percentage of views in your retention charts. Subscribers: Even though the value of a subscriber has dropped a lot, it can still tell us about the growth of our community and how well a video connects with our audience. But it's also important to look at how many subscribers were added and how many were lost. Overall, it's not a great metric because conversion also depends on the content, which is why I don't look at it that much. Unique viewers: It’s a great metric to understand true reach, especially when used in combination with views. New VS Returning Viewers: Even though YouTube puts a lot of weight on this metric, I haven't found it to be too useful. It's not clear what makes someone a new or returning viewer, which makes it hard to use this metric for analysis. I do think this metric will be helpful in the future, but it only tells us a very small amount of information. We still don't know which parts of our content new viewers like and which parts return viewers like. This metric would become invaluable when integrated with retention charts. Likes & Dislikes: Unfortunately, as creators and analysts, when people like and dislike a video, it’s hard to interpret this data without further insights. It depends a lot on how many people see it, and a simple call-to-action can make a big difference. On top of that, it only shows how a certain type of audience feels, which is why you'll see either a very high ratio of "likes" to "dislikes" or a very low ratio of "likes" to "dislikes." As users, we don't know enough about a single user's likes and dislikes and how it relates to our content. Revenue: This is a good metric to keep an eye on. The problem with views is that it doesn't take into account who we want to reach. Some types of content aren't likely to reach a large audience, and views don't show this. Revenue on the other hand, when combined with RPM, does. The amount of money we make from our channel also helps us figure out how healthy our business is. It shows us potential plans for growth and gives us ideas about how to grow our team. RPM: RPM can tell us a lot about our users' ages and where they live, but we have to be careful about how we use this metric. It’s content dependent and gets influenced by factors such as seasonality. However, it can show us problems with how we place ads. Too many creators still let YouTube place their ads automatically, even though it has been shown that putting them there by hand increases RPMs on channels. CPM: This isn't a useful metric because it’s not truly valuable for the improvement of content to know what advertisers are paying. RPM is a significantly more useful metric. However, I do like that this metric is shown because it shows how much of YouTube's income creators get. It helps build trust between YouTube and people who make videos. Browse Traffic: Insights from our browse traffic can show how well we are doing with people who have already interacted with our content or who YouTube thinks are a good fit for it. It also shows how people interact with our content when they are trying to find content to watch. Search Traffic: I don't like to use search traffic to figure out how good a keyword might be, but I do like to use this information to make user profiles for viewers with specific interests. Suggested Traffic: Suggested traffic shows us our ability to reach a wider audience, but it also tells us more about how well our video attracts people who are watching other content pieces on YouTube. Other Traffic Source: Other traffic sources don't give us a lot of useful information that we can use, unfortunately. But end screen traffic can tell us how well our end screen strategies are working. End screen strategies are often undervalued for how well they work. Geography: Geography is a useful metric because it shows how people from different cultures watch our content. It also helps us see how important it is to use the metric system instead of the imperial system in a lot of cases. Or to use neither system to avoid any type of confusion. Cities: This metric is only useful when integrating brands. Other than that, it isn't very useful. Viewer Age: This metric isn't very useful in a lot of situations, which is a shame. We can only use this metric to some degree when it's clear that the audience is in the older age ranges. As soon as our audience gets a little bit younger, we'll realize that this metric is no longer useful. Viewer Gender: This metric is a great way to figure out who we're reaching and if we need to make changes because we're turning away the other gender. This is important sometimes, especially when our content is meant for both men and women. Sometimes it's important to remember that 95% of our audience is one gender, so we don't try to reach the other gender when our content isn't really meant for them. Device type: This seems like a metric that isn't very useful, but it really helps us figure out how our audience consumes our content. Retention Chart: Together with views, traffic sources, and interest insights, I think this is the most useful information on the YouTube dashboard because it helps us figure out what people found interesting and what they didn't. It also shows how well we can keep their attention and lets us know when we're having trouble appealing to a wide audience. Even though most of this information is current as of the time I wrote this tweet, I think YouTube will change their dashboard in ways that will change how we can use some of these metrics. In the end, I think that views, views over time, revenue, and retention charts are the most important metrics. I think that these metrics can almost tell you everything you need to know about a successful channel.