PHC
Ever since Facebook started auto-play video, I’ve promised myself that I would start creating more videos. For one reason or another, that really hasn’t happened, but this remains a high priority.
Recently I found a reason to experiment, and it would be with an unorthodox video that broke many of the rules we’re told to follow when sharing to Facebook. In this post, I wanted to share what I did, the results I’ve seen and what I plan to do next time.
The Video
Last week, I recorded a Social Media Pubcast episode with Andrew Foxwell, one of a monthly series that we schedule. This is typically only audio, published to iTunes and Stitcher. For no real reason, I was inspired to record a video version of the Pubcast that I could upload both to Facebook and YouTube.
When people ask me about Facebook video optimization, I generally recommend a handful of things:
- Keep it relatively short (30-120 seconds)
- Make it VISUAL
- Add text or something that gets the user’s attention
- Don’t just do a “talking head” video
Well, I broke all of these rules. But as an experiment, I really didn’t care.
The video was 45 minutes. In fact, it would have been longer than 50 minutes, but Facebook has a 45 minute maximum on uploaded videos.
The video was boring as hell. Yeah, it wasn’t visual at all. It was just Andrew and me, talking to one another, each with a beer in hand. It was a screenshare of a GoToMeeting. I guess it may have been somewhat interesting to people really into the content, but it wasn’t visual at all.
There was no text. Yeah, like I said, just two guys talking. I didn’t add an intro or anything to attract a user’s attention when the video auto-played without sound.
It was the definition of a boring, talking head video. Just talking. Nothing else.
Why would someone watch this 45 minute video? I don’t know. I fully expected it to fail miserably. The video version offered very little that couldn’t be found in the podcast version.
It was also kind of a hassle. When you have a 45 minute video, it takes a long time to upload. I really didn’t do any editing at all, but that didn’t matter. It first took several minutes to export, and then it took several hours to upload to both Facebook and YouTube. Of course, it also killed my network’s bandwidth during that time.
I also promoted the post. I didn’t shell out a ton of money promoting it because I really didn’t expect it to do very well. I set a $10 per day budget targeting only my website visitors on desktop news feed. I ran that campaign for a week.
The Results
The surface results are far better than I every would have expected. After a little digging, it’s a bit less impressive, but still good.
As of this moment, the video has reached 21,456 people, though only 12,640 of that is organic. As you know, I really don’t care about this metric, but most people do care so I thought I’d share. The organic reach is lower than the typical link share post to my page which tends to average reaching about 20,000 people.
So far, there have been 6,368 views of the video (three seconds or longer), 3,493 of which were organic.
Of the 6,368 views, 1,580 were 30 seconds or longer. This is a bit more humbling. So nearly 5,000 of the views were between three and 30 seconds. Not that impressive.
Of the 6,368 video views, 6,133 were auto-play and only 235 (4%) were click-to-play. I find this a bit odd. What value is there in watching this video without audio?
The average view duration is 2:08. That sounds good and horrible at the same time. The video itself is 45 minutes long, so that’s a small percentage. But I’m seriously surprised the average user watched it that long on Facebook.
The biggest drop-off of views occurred at the 1:07 mark. I really don’t know what was special about that moment, but it’s when Andrew and I were talking about how this video was “RAW!”
Approximately 1% of the views were for 41 minutes or longer. That tells me that of the 235 click-to-plays, about 25% resulted in people watching nearly the entire 45 minute video.
So far, I’ve spent $55.94 promoting this video. That promotion resulted in 2,876 video views and 111 clicks-to-play views. Cost per video view is under $.02 and cost per video play is $.50.
Lessons and Future Plans
This was a video that was destined to fail. Yet, it did far better than I expected. Still, there are several lessons here.
1) Some visuals are necessary. People watched the video. A lot of people watched the video. But I wanted them to click to watch and listen. To get that, adding visuals and text would help.
2) Include text during the first 30 seconds. This is related to #1, but visuals could also mean graphics. Adding text for the auto-play portion to clarify what this is and what it’s about — maybe even a CTA — would help.
3) Include notes that indicate when each new section starts. This was a user request and is a great point. If I’m going to publish a 45-minute video on Facebook — where I don’t expect people to watch longer than two minutes — I should include a guide so that people can skip ahead.
4) Promote on mobile. I promoted the original post to desktop for a reason. I wanted to create separate ads for mobile based on the OS the user was on. That way, I could include a CTA button to download from either iTunes or Stitcher.
Those are the things I’d do to make Pubcast episodes more interesting on Facebook. However, I know that this is probably not the ideal format.
I’m in the early stages of planning some short, visual, educational videos. These are ideal for Facebook, and I expect them to do very well.
Your Turn
So, will Facebook users watch a 45-minute video? Yes and no. More than I expected, but you still need to make it easy for them to watch!
Have you experimented much with Facebook video? What results are you seeing?
Let me know in the comments below!
Category: PHC Blog Uncategorized