Quantcast
Channel: Data-Driven Marketing Archives | Rival IQ
Viewing all articles
Browse latest Browse all 138

Master Your Facebook Video Metrics

$
0
0

We’ve all heard it: organic reach on Facebook has declined, and sharing native video is your best bet for getting more organic reach.

The extent to which this is true doesn’t really matter.

If you’re publishing native video on Facebook and you’re not taking advantage of the full set of video metrics they make available, you’re leaving money on the table.

In this post, we’ll introduce you to Facebook’s native video engagement metrics and how to compute several key video engagement rates. From there, you’ll learn how to measure your video performance and how to benchmark any of your videos against your average performance.

header

The Basics of Native Facebook Video Posts

The ability to share native video on Facebook has been around for years, and in May of 2014, Facebook expanded their video engagement metrics. This data gives you the info you need to build an understanding of how your audience responds to your video content.

We’re going to dig into all of this video-metric goodness, but before we do, let me introduce you to the example video post we’ll use throughout this article. This video of insane manual dexterity and focus comes to us courtesy of CLYW, a high-end yo-yo manufacturer in Edmonton, Canada.

The post from October 14th features native video and encourages sharing as a mechanism to enter a contest – a drawing to win a yo-yo from the collection of a Chezh yo-yo champion.

I’ll use Rival IQ’s Facebook Insights dashboards to walk through the data. Of course, the raw data you will need to compute these metrics yourself are available directly in Facebook Insights, too.

For reference, here’s the example post as shown in Rival IQ.

Facebook Video Post Details

Post Engagement Metrics for Video (And All Posts, Really)

The standard Facebook post metrics are incredibly useful, particularly for comparing the performance of your video content to your other posts.

You’re probably very comfortable with the standard Facebook Post engagement metrics, but let’s review anyway.

Reach and Impressions

Reach is the measure of the unique number of Facebook users that saw your post. Impressions is the number of times your post was seen. In other words, Reach is really just unique Impressions.

Facebook breaks down each of these concepts into three segments: organic, viral, and paid.

Both Reach and Impressions are metrics that are private to your page admin account. You cannot access these metrics for anyone’s posts other than your own.

In the example post, the total reach is 84k users, and there were 145k total impressions. That means that many of the people who saw this post saw it more than one time.

Facebook Video Reach

Likes, Comments, and Shares

Likes, Comments, and Shares are the three most basic Facebook engagement metrics. They help measure the audience resonance of your content, and they’re publicly available data for both you and other pages on Facebook.

Being publicly available, they are a great way to benchmark your post engagement against other pages.

In our example post, we can see the sum of Likes, Comments, and Shares is 1.1k.

Facebook Video Likes Comments Shares

Engagement Rate

Engagement rate is a measure of the audience engagement with of your post, relative to some other metric. Typically, you calculate engagement rates to help you compare two different posts.

For example, if you have one post with 100k total reach and another with only 20k total reach, you can divide the engagement of each post by its reach to adjust for the fact that five times the people saw the first post.

In the example post, we can compute the total engagement as the sum of all clicks, likes, comments, and shares. To get the engagement rate (i.e., the percentage of people who saw the post and engaged), we divide the total engagement by total reach.

The engagement rate is: (7.8k + 1.1k) / 84k = 10.6%.

Facebook Video Engagement Rate

At Rival IQ, we most often talk about engagement rates computed only using publicly available data. That is, the sum of likes, comments, and shares divided by the page audience. Obviously, this number is much different that the rates we can compute when we have more data. However, when you’re looking to benchmark yourself against the competition, you go with what you can get.

Native Facebook Video’s Engagement Metrics

Now that we’ve reviewed the basic metrics that are available to you in Facebook Insights for posts of all types, let’s move on to the good stuff, video.

When you post native video to Facebook, you’ll get access to rich video engagement metrics through Facebook Insights. With this additional data, you can breakdown the aggregate viewing behavior of users who saw your video post, including how well your post captured users and then engaged them throughout the video.

Let’s get into the metrics and talk about what each of them can tell you.

Total Video Views

Total Video Views is the number of times your video was viewed for 3 seconds or more. Viewers who watch less than 3 seconds of your video are not counted in this number.

Facebook also provides breakdowns of Total Video Views into Paid and Organic segments.

In our example post, the Total Views are about 26k with 15.5k Paid and 10.5k Organic.

Facebook Video Total Views

30-Second Views

30-second Views is the number of views of your video to 30 seconds, or to the end if your video is shorter than 30 seconds. Again, Facebook provides breakdowns of these metrics into Paid and Organic segments.

In our example post, the 30-second Views are about 7.8k with 4k Paid views and 3.7k Organic views.

Facebook Video 30-second views

Complete Views aka Views to 95%

Complete Views is the number of views of your video where the user viewed from the beginning to 95% of the length. Just as with our two previous metrics, Facebook provides breakdowns of Complete Views for Paid and Organic segments.

In our example post, the Complete Views are about 7.7k with 4.0k Paid views and 3.7k Organic views.

Facebook Video Complete Views

One additional note about Complete Views:

At times with videos less than 30 seconds in length, you may see situations where this count is higher than your 30-second Views. This is explainable: for a 25-second video, a user must watch the entirety of the video to be counted as a 30-second view; however, watching 24 seconds of the video is enough to be counted in the 95% category.

Calculating Video Engagement Rates Unlocks Benchmarking

Facebook’s video engagement metrics give us the ability to measure the absolute results of our native video posts, but they don’t really give us the ability to compare performance across our video posts.

Without the ability to compare your posts, you’re stuck. Fortunately, math is your friend, and we can calculate a few handy video engagement ratios rates that will unlock your benchmarking activities.

Video View Rate (Views to Impressions)

The Video View Rate is the percentage of users who viewed your post that actually watched your video for more than 3 seconds.

To get the Video View Rate, you calculate the ratio of Total Views to Total Impressions, both metrics available in Facebook Insights.

Video View Rate = Total Views / Impressions

For our example post:

26k Total Views / 145k Impressions = 18%.

Facebook Video View Rate

30-second View Rate (30-second Views to Total Views)

Now that we understand how to answer how enticing your video was to watch, let’s measure how engaging viewers found your actual video content.

Let’s answer the question: what percentage of users who viewed my video watched for 30 seconds? (Remember, if the video is less than 30 seconds, watching to the end counts here, too).

To answer the question, you can calculate the ratio of 30-second views to Total Views.

30-Second View Rate = 30-Second Views / Total Views

Using our example post, 7.8k 30-second Views / 26k Total Views = 29%.

Facebook Video 30-second view rate

Complete View Rate (Complete Views to Total Views)

Similar to our 30-second View Rate, the Complete View rate can answer: what percentage of my viewers watch my entire video?

To get the answer, we’re going to compute one more ratio that compares Complete Views to Total Views (you guessed it!).

Complete View Rate = Complete Views / Total Views

Using our example post, 7.7k Complete Views / 26k Total Views = 29%.

Facebook Video Complete View Rate

Putting It All Together!

Now that we’ve gone through the exercise of defining our video engagement rates, we can put them together with our other video metrics to build a few performance summaries. We’re going to build up a single-post engagement funnel, including breaking down paid vs. organic segments.

Build a Single-Post Engagement Funnel

Facebook users who see your video post (your reach) are all people who might watch your video. How many of them actually watch? Once you get them watching, how many hang around for 30 seconds? Until the end?

You know the answer that’s coming: the Facebook video engagement funnel.

Here is the engagement funnel for our example post. We’ll step through how to build this funnel after the graphic.

Facebook Video Engagement Funnel

Stage 1: How many Impressions?

The sum of your Organic, Viral, and Paid impressions sum to your Total Impressions. The more people you reach (and the more times you reach them), the more potential audience for your video you earn.

Our example post had 145k Impressions.

Stage 2: How many of our Impressions turned into Viewers?

Here, we can understand how the crafting of our post including text, title, preview thumbnail or video beginning (in the case of auto play) all impact the percentage of users who decide to watch our video.

Of our 145k post impressions, the example post gathered 26k Total Video Views.

26k / 145k = 18%.

Stage 3: How many of our Viewers watched for 30 seconds?

Moving down the funnel, we can gain an understanding of how engaging the video content is, as the viewer gets further into the video.

Of the post’s 26k views, 7.8k of them went to 30 seconds.

7.8k / 26k = 29.5% watched through 30 seconds.

Stage 4: How many of our 30-second Viewers watched to the end?

Let’s finish up the funnel by computing how many of the viewers that watched to the 30-second mark stuck around until the end. This will give you a general idea of how engaging the rest of the content was after the 30-second point.

For our example post, this won’t be the most exciting calculation because the video was only 31 seconds long. Still, enough people scrolled away or didn’t finish watching to give us the following numbers.

7.7k Complete Views / 7.8k 30-Second Views = 99% of the 30-second viewers watched until the end.

Don’t you wish you’d see more 99% results in all of your funnels?!

Comparing Paid Versus Organic

If you’re doing paid promotion of your videos to gain more reach, you should be evaluating the video engagement of your paid audience. Facebook gives you separate metrics of Total Views, Views to 30 seconds, and Views to 95% for paid and organic. With this data and the ratios we just defined, you can compare the video engagement from our paid impressions to that of your organic impressions, even with different total impressions for each.

To compute the funnel for your paid traffic, you’ll use the corresponding paid number for each of the metrics in your funnel. You can do the same for your organic views.

Here is the paid vs. organic comparison funnel for the example post.

Facebook Video Paid vs Organic

As you can see, the performance for the paid traffic was higher for complete views, but had a much lower 30-second view rate. My guess is that the targeting for the paid traffic was pretty good, and the organic performance was dragged down by the large number of shares to a less relevant audience. Do you have a hypothesis? I’d love to see them in the comments.

In general, once we’ve done this analysis, we can start to answer questions about the accuracy of our paid audience targeting. Using our organic video engagement as a benchmark, we can evaluate the similarity of our targeted audience to the audience we earned organically.

Obviously this comparison depends on your paid targeting, so it is hard to make sweeping conclusions without considering the objective. If you were targeting your page fans with your paid activity, it is reasonable to expect similar or even better engagement numbers from your paid viewers.

One More Thing: High-Resolution Retained Viewer Data

For any Facebook video post, you can find a nice visualization of retained viewers versus time in the Insights interface. Thanks Facebook!

Facebook Video Retention

If you lose lots of visitors at a particular point in the video, that gives you an idea of where you could improve the content of your video.

Benchmarking Your Facebook Videos Against One Another

Now that we have established funnel metrics that enable us to compare videos, let’s apply what we’ve learned to compare our example post to other CLYW video posts.

Finding Top Performers

Understanding which pieces of content performed best can give us insights into improving future efforts. From video creation to distribution strategy, learning what works best for you can help maximize your efforts.

To find your top performers, you can compute your various video engagement rate metrics for a set of video posts and sort the thing you’re trying to maximize.

For example, to find the video post that did a best job of getting people to watch, sort your data by the views to impressions ratio.

To find out which video content kept people engaged to the end, sort by the views to 95% to Total Views ratio.

Computing Averages Across Videos

Keeping track of your average video post engagement will give you a baseline for evaluating each new video  you produce and share. It turns out that computing the averages for each of your video engagement rates is a little tricky, so you might want to check out the inset below, Computing Averages Across Videos.

Tutorial: Computing Engagement Rate Averages Across Videos

Computing the average of your engagement rates is not as simple as averaging the rates themselves. If you were to do that, you’d be giving each video equal weight in your calculation. Most likely, you would like to give each video weight in proportion to how many users saw each post.

For example, say you have two videos in your set.

  •      Video one has 500 views, and 50 people watched to 30 seconds. The 30-second view to Total Views rate is 10% = 50 / 500.
  •      Video two has only 100 views, but 40 of those people watched to 30 seconds. The 30-second view to Total Views rate is 40% = 40 / 100. This is a much higher rate, but on a post with many fewer viewers.

The average of your 30-Second View to Total Views rate is (50 + 40) / (500 + 100) = 90 / 600 = 15%.

If you just averaged the individual rates, you’d get the incorrect value of (10% + 40%) / 2 = 25%.

Let’s work through the example with CLYW’s data. We’re analyzing 14 native video posts over the last 90 days.

Here are the posts:

full-posts-table

 

Average Video View Rate

The sum of the Impressions for these 14 posts is 405k. The sum of the Video Views for these 14 posts is 55.5k. With these two sums, we can compute the average Video View Rate as 55.5k / 405k = 13.7%.

Hurray! Now that we have a benchmark, how does our example video compare?

The example post video had 26k views and 145k impressions, an 18% Video View rate. Nice work, CLYW! Your view rate on that one is a 40% improvement over your average.

Average 30-Second View Rate

The total Video Views for these 14 posts is 55.5k. The total 30-Second Views is 18k. Together they yield an average 30-second view rate of 18k / 55.5k = 33%.

Let’s compare to our example video which had 7.8k 30-second views and  26k total views equalling an average 30-second view rate of 29.5%. That’s a 10% drop from the average. Somebody has some ‘splainin to do.

At this point, it’s time to break this one down into paid and organic segments to see if we can get a better idea of what is happening. To compute this number for paid, divide the sum of the paid 30-second paid views by the sum of the paid views. Now, repeat the formula for the organic numbers.

Once we break it down, we will see that the average 30-second view rate for organic is 42.7%. The average rate for paid views is 25.2%. This video has lower 30-second view rates for both segments!

It is time to revisit the original premise of the post: a contest where sharing the post entered the sharer into a drawing. My hypothesis is that much of the viewing audience for this video was not comparable to the typical audience for CLYW. Many of the viewers were friends of the sharers. While they were intrigued enough to start watching, their interest waned quickly, and they didn’t stick around until the end. Have other theories? I’d love to see them in the comments.

So Now What?

You’ve made it to the end! I hope you’ve learned how to take your analysis of Facebook native video posts to the next level. With a little bit of division and your data, you can produce an in-depth analysis of Facebook native video posts for your clients or company.

What are you waiting for, get after it!

Before I let you go, I have to mention that Rival IQ makes doing all of this Facebook video analysis so easy! We have the data you need, and we compute most of the engagement rates you’ll need as well. Go read about our Facebook Insights dashboard.

Thanks again to Chris Mikulin from CLYW for letting us use his Facebook data as an example for this post! It was a fun one to write!

The post Master Your Facebook Video Metrics appeared first on Rival IQ.


Viewing all articles
Browse latest Browse all 138

Trending Articles