# Where On Earth Is the Online Video Arbitrage Model?

On the text web, arbitrage has become the word of the day as whole ecosystems have sprung up to optimize and monetize the link economy.  But when it comes to online video, the arbitrage model is failing badly.

Whether it’s buying a keyword for a buck, and making $1.02 on each visit, or buying a shopping link for a quarter and making a product sale for a few pennies more of profit, whole online ecosystems have sprung up to optimize and monetize pay-per-click and the search engine- and display-based arbitrage models. Video, though, has yet to find a profitable niche based on arbitrage. It’s not for lack of trying, though. The online video landscape is littered with companies that tried to buy their way to big audiences, yet failed. Take Veoh, for example. I was assured by someone on the inside that the company routinely spent big bucks to drive viewers to its videos, yet couldn’t recoup the investment with advertising or product placement. I’ve had the opportunity to look at the books of a few other failing companies in the space, and seen a similar story. Big quarterly marketing spends lead to big show views – right up until the money runs out. Turn that spigot off, and the traffic disappears too. The problem with using pay-per-click or display to drive viewership is that the economics are still out of whack. Take YouTube, for example. YouTube has a form of pay-per click advertising that you can use to drive viewers to your videos. Unfortunately, you have to spend at least a penny per click, which works out to ten dollars for every thousand views. In practice, I’ve found that you need to spend more like five cents to get any meaningful action – and that means you’ll need to make at least a$50 CPM on any overlays or pre-rolls you might run.  That’s not likely to happen very often, given the CPM (cost per thousand impressions) pricing of most video advertising.

In fact, it looks like bottom-of-the-barrel CPC (cost-per-click) marketers have already flooded the platform anyway, driving the price of viewership campaigns through the roof.  For example, head over to YouTube and search on Olympic Skier “Julia Mancuso”. I just initiated a paid campaign there to try and drive views to our new Digg Dialogg interview with the sexy star.  I’m bidding .05 – or a $50 CPM, which is more than what I’ll bring in with any pre-rolls or overlays that I’m likely to serve. Yet I’m consistently outbid for ad placement by Visa, a video featuring “The Hottest Girls of the Winter Games 2010”, and, inexplicably, a video helping you repair drywall as well as one hawking a gadget that’s guaranteed to cure bad breath. We’ve tried a wide range of other arbitrage-style viewership promotions, from Facebook widgets to SEM, but none deliver profit. And we’ve taken a pass on many, many more, priced at well over$100 per thousand views.

However, there’s a new, sneaky arbitrage model making the rounds, one that seems to deliver real results.  It involves buying up a wide range of super-cheap  300 x h250 display ads – typically for well under a dollar CPM – and then auto-playing pre-rolls and other video into that ad unit.  This actually seems to work, in many cases, because the auto-play video can be hard for users to find, and shut off, before enough of the pre-roll is served to constitute a view.

It’s annoying, but lucrative.  Video ad networks routinely deliver remnant pre-rolls for \$5 CPMs, and even after ad serving costs, revshare and the inventory payout to the serving web site, you can generate enough real money to eke out a profit.

But in the end it’s duplicitous and ineffective.  Web sites serving these auto-play video ads are delivering a terrible experience to their site visitors.  Advertisers paying for these impressions are being misled as well, as most of them are unasked for, and angrily terminated by the waylaid web surfer.  And now the IAB (Interactive Advertising Bureau) is getting involved to try to curtail – or at least pull the curtain back – on this shady practice.  Back in December the industry trade group issued an updated set of guidelines requiring web sites to disclose details on auto-play video when it happens.  And this is just the “first part of a broader auto-play initiative by the IAB’s Digital Video Committee”.

Maybe it’s wishful thinking, but I believe the IAB will move to outlaw, or seriously curtail autoplay video ads when they are viewed out of context.  Because even though I’m all for figuring out a successful video arbitrage scheme, it’s simply not right when built on underhanded, sneaky and deceitful tactics.

What do you think?  Have you figured out an honest video arbitrage system that works?  Do you believe in auto-play at all times?  Post your thoughts in the comments.

Jim Louderback is CEO of Revision3. He was previously vice president of Ziff Davis Media and Editor-in-Chief of PC Magazine and PCMag.com.