A frequently asked question today is also the headline to this AdAge article by Mary Pedersen: “Best Practices: What is the Optimal Length for Video Content?”
For years we’ve heard about our decreasing attention span, now apparently less than that of a goldfish. The proponents of this hogwash blame digital media (and yes, they are the digital media) and are somehow OK with believing human evolution works that fast. Whatever.
So when we meet with clients about video their briefings always includes this advisory. "Make it short. No, make it shorter. Shorter, please. But be sure to get all the messages across." The contraction in this brief is not the issue. The belief that audiences are unable to pay attention for more than a few seconds is the issue. Add to that the belief that viewers/consumers/prospects are unwilling to view more than a few seconds and it makes me wonder why we’re doing video in the first place.
But there's another issue, and it's more strategic. There’s an unspoken objective that everyone who watches the video should watch all of it. I’m here to say, that’s wrong. Think about that for a second while I take you back a bit.
Many years ago, during the mostly-print days, when the advertising world first began researching creative to learn what works and why, there was a basic understanding and acceptance that, of those who view an ad, most only read the headline and look at the image. And a smaller group reads the body copy. David Ogilvy popularized this saying only 10% read the body copy. Importantly, he pointed out that the 10% that are sufficiently interested to read the body copy are your best prospects or customers. Those are the ones who buy.
Now think about these two things: First, is 10% of those who view your ad a good number? (You know the answer is yes.) And second, what if there was only a limited amount of body copy? Or none, as the "no one reads body copy" crowd prefers. Would that 10% be cheated out of important information? Again, the answer is yes.
Ogilvy’s rule was that long copy sells better than short copy. The rationale was simple and believable in that context – that only your best prospects and customers read the body copy, so the more information you can give them the more likely they are to buy. Makes sense if you don’t believe that people are as dumb as goldfish. (And if the ability to pay attention isn’t related to intelligence, then I don’t know what is. So much for bragging about self-diagnosed ADHD as proof of one's creative mind.) It makes even more sense if you ever worked in direct marketing.
Now cut to today, assuming your goldfish-level attention span hasn't been exhausted. What has changed? Have humans really evolved (or devolved) to where we compete with goldfish for who has the better attention abilities? To test this we need only grab a control group, say a primitive tribe in the Amazon or New Guinea and test our attention abilities against theirs, right? Control for distractions and we’ll have the answer. Frankly, that is the answer: distractions, AKA options. Today we have more options for our attention, and those options are at the heart of the matter. With a shorter video you reduce the viewer’s options, but not in a good way. You tell the viewer, "Sorry dude, but that’s all you get. Click through now or go away."
But what if some viewers want more info before investing in a click? (And we know clicks are not that easy to get, the good clicks that is.) Why not give it to them? The answer - make that the wrong answer - goes back to the unsaid objective, that the viewer watch the entire video.
The point (finally!) is that we should be happy with a sufficient percentage/number of viewers watching enough of the video to achieve a proper objective, be it perception change or action. And along the way, why cut the program short on a potential customer by imposing an artificially-derived length, a goldfish-inspired length.
Let me give you two examples of how this works. About a year ago we produced a video that was 1:48. There was no predetermined length in the brief, not even a goal to keep it under two minutes. We have many more that are longer, but I’m using this example because it is an "OK" performer among the videos we’ve produced the past two years - not a great performer and not a loser. It’s also an example of a high-involvement product (master’s degree) with a relatively small market, so every click counts.
Looking only at the TrueView In-stream data (the most interruptive and easiest ads to skip), we lost 73% of the viewers in the first 15 seconds. But we still had 27%, and they would have been lost at that point too had we produced a :15. By 30 seconds we were down to 18%. At 1:23 we dropped to 10%, and by the end we were at 8.5%. For us, the glass was half-full, or rather 8.5% full, and that’s good.
By the way, CTR, CPC and Cost/Conversion were all strong, so maybe there’s a relationship. (Ya think?) It wasn’t a great video when we look at relative audience retention, but on the positive side, relative retention increased throughout the video, indicating that interest continued to grow to the end. And it converted very well, regardless of whether viewers completed the video or not. That was the success.
That 1:48 video ran as an ad, and that's a long ad! Now let’s look at pure content - not an ad. Same category, broader interest, wider audience. This video was more than 3-times as long, 7:13. In this case there was no advertising, so all views were organic, which naturally delivers better retention numbers. But we’ll compare organic to organic as I lay this out.
At :15 we had 88% retention, but no one cuts content videos that short. (For comparison, the ad video had 56% organic retention at this point.) At 1 minute we had 78% retention (ad, 33%), 66% at 2 minutes (21% for the ad), 59% at 3 minutes, and so on to the 7:13 end where we still had 41% of the viewers. The average view duration is currently running 4:12, of a 7:13 video, or 58%.
This is worth repeating: 41% retention after 7 minutes.
Going in, no one wanted a seven-minute video. But the content was good so we tried it out and it worked. This isn’t bragging, a sales pitch or self-congratulations, it’s discovery and proof.
So, what is the "Optimal Length for Video Content?"
It’s not your choice. It’s the viewer's. But don’t put arbitrary limits on what your audience is allowed to view or learn about your brand or product. Let them decide how much they’ll consume. That is the optimal length for video content. And no one knows what that is in advance. What we do know is that if we provide more they will view more, and those that view the most also buy the most.
UPDATE: If you don't believe me, check the results Vidyard has been getting in this article by Lance Johnson.