Viddy...how many users do they really have?

As someone who's tracked and run Internet video startups, the massive growth of Viddy earlier this year impressed me but also set off my "it doesn't seem right" gut instinct. Now that I'm no longer heads down running www.synctv.com, I had a little time to really dig into this.

So how did Viddy go from 15,000 uniques/month* (quantcast) in March to over 1M uniques/month in April, less than a month later? I've seen this happen many times before, but not in the "create a video and upload it" space Viddy, Tout, SocialCam, etc play in. As interesting, how did Viddy crash so hard only a couple months later? Was it a fad? Did something else affect their popularity?

Video is not like photos. A picture is fast to upload and fast to "consume". Video takes a lot longer to create (at least interesting videos!) and take a lot longer to consume. As anyone who's posted a video knows, it's not easy to create an engaging video your friends/family/social network will actually watch.
So back to my question: how did Viddy grow and then shrink so fast?

I looked at a number of factors, and the most likely scenario I've discovered is this. Viddy's growth happened at the same time as their integration with Facebook's open graph API and when Facebook wasn't restricting how many layers of friends would be exposed to the video. Even better (or worse, depending on who you are), the various user tracking companies (Alexa/Quancast/Compete) seemed to count every single person who got/clicked the video as a Viddy "user". In other words, I upload a Viddy clip I made to Facebook, and all of my friends see this instantly. Then, if they click on the video, my friends' friends could (depending on your settings)  also show up as Viddy "users". As you can imagine, this leads to massive virality.

But should someone who's not even signed up or using Viddy be counted as a user? I argue that the only true users are those who either create content using Viddy's service and/or regularly use the Viddy site to view videos. Someone who get's a video they know nothing about shouldn't be counted as a Viddy user, should they? At the very least, knowing how many users are really using the site vs users who just got the video because of Facebook's API implementation, is important to measure Viddy's true growth.

Now how about that steep drop off? Well, interestingly enough, it also coincides with Facebook's tightening of its API virality. Viddy videos will now no longer penetrate several layers of "friends of friends", and voila...Viddy is back to a steady state of 60,000 uniques/month for the past 3-4 months.

So hats off to Viddy for getting investors to plow money into them based on...um...creative user numbers. They have a lot of smart people there, and now have a large rainy day fund. I'm sure they'll reinvent themselves and get those numbers up again. I just hope that next time, the growth comes from actual users of their service, not these amped up numbers.


*Yes, I know, all of these tracking services are inaccurate, inconsistent, etc. However, used as a rough traffic guide, they do the job...somewhat.