Viewer Abandonment Rates for Video Infographics

The length of an animated infographic video is a common point of discussion when talking with clients and prospective clients about creating a custom video explanation. Some ask for advice about how long the video should be, others already have a specific duration in mind (and often that duration is unrealistic given the amount of information they want to include in the video, but that’s a topic for future article). My advice is that shorter is usually better – best to keep the duration between 1 and 2 minutes. The reason I say this is largely due to a study reported by Visible Measures in 2010 that reported that you should expect to lose about 20% of your audience within the first 10 seconds of playback and that viewer retention steadily declines throughout the video.

However, I’ve long suspected that there’s more to viewer abandonment rate than what is reported in that article. I’ve thought that there’s probably a strong correlation between the video abandonment rates and length of the videos – the shorter your video, the lower the abandonment rate will be. In other words, that losing 20% of your audience within the first 10 seconds would not hold true across videos of varying durations. For shorter videos, it wouldn’t hold up. The Visible Measures report kept their sampling to videos of less than 5 minutes in length. But there’s got to be a vastly different abandonment rate between say a 30 second video and a 4 minute video. To test this I collected data on some of my own infographic videos that I’ve published myself on YouTube.

YouTube has some nice analytics that will let you analyze viewer retention of any of your videos over the length of the video so you can easily see at what point in your video you lose 10% of your viewers, 50%, etc. You can also see a quick summary of what percentage of your video is seen by your viewers on average. The sample size in my little analysis is pretty small, but to me it suggests a pretty compelling argument that if you keep your video to less than about 1:15, you may be able to keep 80% or more of your viewers. And for those videos I’ve published that are around 30 seconds in length, I only lost 5-7% of viewers in the whole video. So I don’t think we just have to give up hope and admit defeat, that we’ll lose 20% of our viewers within the first 10 seconds. If we keep the video short enough, we may not lose that many viewers at all. For those interested, here’s my data to back this up.

Video Abandonment Data Visualization

First up, here’s the spreadsheet of data:

And here’s a visual representation of that in a bar chart:

With this you can see that both average length viewed and the audience retention at the half way point of the video are pretty good for the top two videos which are both around 30 seconds in length. But let’s see what kind of correlation there is between the video duration and average percentage of the video viewed.

Looks to me like there’s a strong correlation worth considering. After all, you’ve spent time and money creating your video explanation. You want it to get viewed and you want those who do watch it to watch all of it. You’ve probably spent hours honing your script. Do you really want viewers to only see the first 10% or even only 50% of your video? No, I don’t think so. So, if the scatter plot chart above can be extrapolated and generalized, if we keep our video explanations to under 1:15, we can probably get our viewers to watch on average, 80% or more of the video. If we go even shorter, say 30 – 45 seconds, we can probably get 90% or more viewed.

It’s worth noting that there are, no doubt, other factors involved in video abandonment, such as the topic of the video, how niche the audience is, how engaging the video is, etc. But, I still believe that if you standardized these variables you would still see a correlation of abandonment to video duration. The abandonment percentage rate for a highly engaging, funny video explanation may not be as high as it would be for a more matter-of-fact presentation, but there’s no doubt in my mind that you’d still see higher viewer retention in a 1 minute video than a 2 minute video.

The YouTube Data

As I said, this is a pretty small sampling. So I’d encourage all business who have video explanations published on YouTube to do what I’ve done here and analyze their retention. If you log into your YouTube account and go to your Video Manager, each video will have a little chart icon that you can click on to view the analytics data for that video. From there click on Audience Retention and change the date range to suit your needs. For example, here’s a couple of screen shots of a couple of my videos:


So, know when I’m asked about video duration I’ve got a somewhat more refined answer with some data to back it up – ideally, something like 30 – 45 seconds. That’s a pretty tall order to distill your message down to something that succinct, but if you really want ALL of your message to get across, you’ll stand a better chance if keep it short and sweet.



No comments yet.

Leave a Reply