We're missing out on a ton of great videos, and something has to be done about it. I've been thinking a lot about video curation lately. It's Nalts' fault, really—he's the one that got me started off down that line of thinking with his awesome post regarding 2011 Video Predictions. There are so many crap videos out there, we need sites that can just get us straight to the good stuff (or, you know… YouTube could just do a better job of it themselves).
Nalts wrote back in December:
"It's simply not possible to "browse" for good videos on YouTube anymore, although perhaps Google will consider some of my unsolicited New Years Resolutions for YouTube. Ultimately I'm not likely to find good content surfing the "most viewed" on YouTube (now dominated by a few niche "web stars" that appear to be "crowd sourced" by a tiny segment of apparently stoned teenage video enthusiasts). Instead, we're more likely to find it via curators like eGuiders. Why aren't we seeing more curators?"
That's why we have seen the rise of sites like Devour or Buzzfeed, or even PopScreen's aggregated data. And I use places like that every week when looking for awesome viral videos to share with you in my weekly Round Up piece. But I always wonder what gems I'm missing.
See, we like to think that YouTube is the great equalizer… that any average Joe can get a record contract just by posting videos of their awesome singing talent. We think YouTube breaks the old school business model for music and movies. But it doesn't. It just creates a new business model that is exactly the same. Meet the new boss… same as the old boss. It's not a level playing field at all… we just think it is.
We still only like the videos we're told to like… because there are an almost infinite number of videos we never hear of or see at all. The curators are the ones filtering content toward us, and unless we go digging through the raw pipeline of video uploads ourselves, we never know what we're missing.
There Are Too Many Videos To Keep Up With
Even if we wanted to crawl through the raw feed of uploaded videos… we couldn't. Seriously… I just did some quick math and proved it. It's mathematically impossible for all YouTube videos to get a fair shot. Most of you probably don't need proof, but let's just run the numbers real quick just for fun.
YouTube says there are 35 hours of video uploaded every minute. So…
35 (hours) X 60 (minutes per hour) = 2100 minutes of video uploaded every minute
The average length of a YouTube video is 2 minutes 46 seconds. We'll call that 2.75 minutes for our mathematical purposes. So…
2100 (minutes) divided by 2.75 (average minutes per video) = 763 videos uploaded every minute (give or take)
Now… how many people would it take to keep up with a stream like that? Basic logic says that if you wanted to track all videos uploaded to make sure no one ever missed a great video… then you'd need 2100 employees working round the clock… it takes one employee one minute to watch one minute of video, right?
Even if you only pay $5/hour, which would be pushing it anywhere in the U.S. these days, that's $10,500 an hour:
2100 (employees) x 5.00 (dollars) = 10,500.00
That's just one hour. Generally as an employer, you have to pay employees for all the hours they work… and since there's no "business hours" at YouTube and video can be uploaded any time, day or night… we actually have to pay these video hounds for round the clock work. Oh, and don't forget YouTube doesn't take holidays off either. So…
$10,500.00 (dollars per hour) x 24 (hours in a day) = $252,000 (dollars spent on payroll in one day)
$252,000 (daily payroll) x 365 (days in a year) = $91,980,000 (annual payroll costs)
So… clearly… nobody's monitoring all videos. It's impossible—mathematically, physically, and financially–and we've only looked at YouTube just now, which means we aren't even counting videos from Vimeo, DailyMotion, etc. The sheer volume of videos being uploaded far outweighs our ability (and interest) to keep up with them. Which means that every day… every minute… right this second… there are great videos going unnoticed and getting buried in the flood. What a shame.
What's worse, is that even if we had the time and inclination to view the raw upload feed of all videos, it's impossible. YouTube won't let you search or browse by any negative things like "lowest view count" or "least subscribed," and there's certainly no "raw pipeline" view or feed we can tap into. So there's a lot of gatekeeping going on.
Curators Are The New Gatekeepers
This, of course, is precisely why a video can linger online with hardly any views for months or years at a time and then suddenly turn into a viral rocket. And it's exactly why these curation sites that aggregate the best videos and share them with readers are going to continue playing a huge role in online video's evolution moving forward.
So next time you publish a video, only to be disappointed by the low number of views you've received, take solace in knowing that you're not alone. If I were trying to become a top video creator… I'd be making friends with the most popular video curation sites as quickly as possible.
Why Video SEO & Social Media Matter
Video creators can use social media to get the word out on their videos, which can help get some initial views. And social interaction around a video can definitely lead to new viewers.
Similarly, proper video seo tactics–keywords, titles, tags, annotations, captions, and video sitemaps–are going nowhere as a prominent means of getting exposure for your video, because topical search is really the only way to discover anything in a raw pipeline of videos like the one we're facing. Even the curators will need to use search because there's simply no other way to find new videos.
Like Nalts, I think the future of discovery is in curation–YouTube may continue to be the top video host, but more and more people will find niche curators to help guide them to the videos they're most likely to enjoy. But it's a little bit sad to me that, even in that utopian future of viral video goodness, we'll still never know what we're missing.