Last year we started vlogging and overall using videos a lot more in our communication. It’s an easy way to communicate and a lot more personal than most online communication. We like it a lot and will obviously continue to focus on video this year. Following our recent posts on social media analytics (Twitter, Facebook), it only seems to make sense to go over YouTube Analytics as well. YouTube continues to be the second largest search engine in the world, and of course, you want to monitor your efforts there as well.
When you visit youtube.com/analytics (be sure to pick the right profile in the upper right corner), you’ll find a number of ‘channels’ in the left menu. For this post, I’ll mainly focus on the watch time section. I’ll write a second post on the engagement reports section of YouTube Analytics soon.
YouTube Analytics Overview
YouTube Analytics starts with a nice overview of watch time, average view duration and things like your top 10 videos. The page bottom shows some graphs showing countries, gender, traffic sources, and playback locations. Filters are available for uploads/playlists, subscribed/not subscribed, and live/on demand. My 2 cents? Just a nice overview, but it tells me little. I do like the top 10, but for the rest I’d like to see more details, and that is probably why Youtube Analytics is divided into many subsections, that can easily be accessed through the left sub menu.
Realtime overviews are just available for uploads, not for playlists. Makes sense, right. You can see stats for the last 48 hours and last 60 minutes:
Hovering elements will give you more details. I already mentioned in an earlier post on Google Search Console that some graphs are better used for trend analysis than anything else, and that seems to be the case here as well, that is why these overviews only really matter at an event, like the start of a new campaign or when you’ve just sent out your email newsletter.
There’s also a convenient overview of these stats for the last five videos you uploaded. So if traffic on that important video is low for instance, you could consider giving it some extra social media exposure or mentioning it in your upcoming newsletter. Clicking one of these specific videos will get you to a page where you’d expect more details on the video, but that isn’t the case in this Realtime section. It just shows the stats a bit larger 🙂 The hover state in the overview shows you a lot more – see image on the right (channel, date of creation, when it’s published, duration and privacy setting).
I’m going to skip this section, as this only works when Adsense is associated with your YouTube channel. We didn’t, as our earnings come primarily from this website and not per se from our videos on Youtube.com. If you have associated Adsense, this section will show estimated earnings and ad performance.
Watch time reports
Now this is an interesting section. This is about how people watch your videos and analyzing this can give you new ideas for promotion and on how to organize your videos.
This section starts with this graph:
That’s a pretty dull graph, I know. The thing is that it’s packed with extras that YouTube Analytics decided to hide in the icons on the left. There is a line graph for watch time and views (filter at the top), which is what is in the screenshot, and then we have from top to bottom:
- A multi-line chart for the videos that were viewed in the selected period (which can be selected using the slider at the bottom of the graph). This will show you your best performing videos and your underachievers. There is a filter for growth, totals and % of totals.
- A stacked area chart for totals and how that total is divided. This isn’t my kind of chart, but hey, there’s a chart for everyone.
- The next one is a pie chart showing basically the same information. Big difference is that this chart can’t be filtered to show data per day, rolling x days total, week, month etc. That makes this chart less useful in my opinion.
- Then there is a bar chart, adding nothing new.
- And then, as a total surprise, a map of the world. This maps shows where your views come from.
Next to watch time, all charts can also show average view duration, average percentage viewed, watch time in hours instead of minutes, subscriber views and/or subscriber minutes watched in that same chart.
Audience retention in YouTube Analytics is about how loyal your viewers are. These are actually nice graphs to play with and see how things compare to each other. It will give you a better view about the performance of certain videos.
That’s last year on our Yoast channel. I can clearly see when we started focusing on video a lot more. The spike just before week 23 (2015) was YoastCon. In our case, that’s also reflected in the table below these charts:
The details of the live streams (just click live in the table) show that these indeed only occurred during YoastCon, conveniently the only day we offered live streams in 2015 😉
You can see there are a few more tables to be analyzed: which country had the most watch time (including average view duration), which day had the most watch time and whether most watch time was done by subscribers or people that still have to subscribe to our Yoast Channel. Nice to see that subscribers watched a bit longer on average (even about 50% longer!).
You should definitely play around with the data in this section and see what comes in handy regarding your own videos.
It almost feels like overkill, but next to the geographic information in the previous sections, YouTube Analytics has some more geographical data in store for us:
Our main YouTube audience is from the US, UK, India and the Netherlands, and these are mostly male viewers in the age of 25-44. I can think of companies where this information is very valuable, like (online) magazines and other companies that try to target a specific audience.
This section gives you a nice overview of where your videos are watched:
- Embedded in external websites and apps. This is the main location for our videos, as we share these on our website and promote them via Facebook, for instance.
- YouTube watch page. Another important one, as a lot of viewers prefer youtube, or click from one video to the other or use the app to view videos.
- There are two minor locations in our case, being YouTube channel page (no need to optimize that, as far as I can see that one gets little traffic), and the leftover location ‘other’.
This section starts with the alert that “Data in this report may be incomplete or missing.”(in our case, not sure whether this is a general alert), and it mentions as the main traffic sources External, Direct or unknown and Unknown – embedded player. Luckily, as with most items displayed in tables in YouTube Analytics, there is an easy click to more detailed information:
Unfortunately, this is as detailed as it goes, clicking the links in this table will give you (drumroll) geographical information, per country per traffic source.
I do like the devices sections, which tells me that a whopping 87% of our watch time is done on a computer, 64% on Windows and 33% on Mac. Just about 6% is done on a phone, and even less on a tablet.
This section just applies to the live streams you had in a certain period. For us, this section holds little information as we only streamed videos for a day. I can imagine this section might also be interesting for certain companies.
That concludes our first post on YouTube Analytics. In an upcoming post, I will dive a bit deeper in the engagement stats YouTube provides. Keep a keen eye on our website for updates!