As I watched Super Bowl XLIII (that reads 43 to all you foreign speakers (j/k)) I kept thinking to myself about how awful they were. What happened? I know the economy is bad, but a good idea shouldn't cost that much, right?

One that caught my eye was the Dorito's ad, which to me was funny (not the funniest commercial I've seen - not even that day but still made me look) and the reason it made me look was because it had something special to it which I later saw was part of the "Crash the Super Bowl" promotion.

They were going to pay USD$1M to the winner of a video upload contest for a Doritos' Ad which would showcase in the, well, Super Bowl. It's a pretty good idea but is that what's left of advertising?

I believe in time the cost of bandwidth will decrease allowing forlarger videos to be easily uploaded to the Internet by users.

I agree that there is much work to be done in findingthe right business model for this, but it will continue to require innovationand making a run at it even if some of the runs fail.

Despite all the collective wishful thinking on the economics of video, it's very challenging to make the math work. Here's why:

The primary cost of video lies in distribution and storage, and when you figure in basic costs for CDN storage and delivery, assuming normal file sizes, completion rates, etc. it costs about $1 for every 1,000 streams you deliver. (In practicality, you're either going to have to commit to $1,000 or more of spend with a CDN on an annual contract, or assume that these numbers are going to be much more expensive. This number also assumes a 90-second video. If your videos are longer, just multiply the costs).

In theory, assuming you own the camera, the lights, the computer, and the editing software, you could value your own time at $0, thus reducing the overall cost of your video to the cost of distribution. 

In this Polyannic scenario, you'd need either a $1 CPM in banner ads and sponsorships for the page on which you display your masterpiece, or you'd need a $2 CPM if your going to monetize it via video ads (since coverage tends to be about 50%). The first scenario is probably more viable, since untargeted video ads via Adap.tv and others tend to have CPMs that dip below $0.30, and selling video ads to an advertiser typically requires a scale of distribution that few sites can achieve.

In the real world, all the overhead costs money, but for sake of argument, let's say that the creation of your oeuvre costs $500, and work from there.

Assuming that you can get a $2 CPM for banners on the page where the video will be hosted, and assuming our $1 cost of distribution per thousand streams, in order to just cover your costs, you'd need to have 500,000 views of your video. This is, practically speaking, impossible without SEM or a massive existing audience.

Now assume you actually want to make money. In order to make $5,000 off of your video you'd need to have 5,500,000 views of your video. Trust me when I say - it's not going to happen.

So until the cost of bandwidth virtually disappears, or ad spending against video dramatically spikes, it's going to be hard slogging here. The Hulus of the world will make money, but only because a) they have already absorbed the costs of production, distribution, brand-building, marketing, etc. and have a wildly popular product before they ever incur a dollar of online costs; b) they have massive distribution and traffic, and can afford to spend huge dollars to attract and maintain that audience; and c) they can attract premium advertisers because of a and b.

For the rest of the world, video offers a ton of potential benefits in terms of audience engagement, viral distribution, and customer conversion. However, real monetization through advertising - in other words, actually making money of the video itself - remains an opium pipedream that refuses to die, mainly because we're unwilling to abandon our desire to believe it can work.