It's a nice study, but it obscured the real problem. I suspect TubeMogul came up with that general conclusion first, and worked backward.
Why do I suspect that? Because online video networks and publishers have been shoving in-banner and auto-play video ads down media buyers' throats since their inception. The real elephant in the room? There's not enough quality pre-roll ad inventory to satisfy advertiser demand, and a few unscrupulous publishers are trying to bring in more money by duping advertisers with faux "video inventory." It's a selling practice that hurts everyone.
Pre-roll ads are usually the top-performing video ad type for most advertisers. There's a very simple reason for that -- the video that contains the pre-roll ad is requested by the user. In such situations, the user will sit through a :15 or :30 (hopefully, a :15) in order to get to the content they really want to see. It's typically an engaged experience, with the user's attention focused on the ad message.
In-banner, interstitial and auto-play video ads, in contrast, have no such lock on the user's attention. They play regardless of whether the user is engaged or they're just casually browsing around. An interstitial typically plays between pages of content, just as a user decides he's done with the first page and want to go to a second. In such cases, a video often comes as a complete surprise. Don't even get me started on in-banner video. Suffice it to say that it's a no-brainer as to why these formats are less effective than pre-roll.
Imagine what it must be like to be a publisher affiliated with a video network. The network will happily sell any pre-roll inventory you can't sell through your direct sales force. Meanwhile, anywhere between 50% to 70% of your banner inventory is being sold at heavily-discounted rates through ad networks and exchanges. It's only natural to want to turn some of that remnant banner inventory into video inventory. That's where in-banner video comes in. If a publisher can call it "video inventory" and sell it at a premium to its regular banner rates, its value jumps several hundred percent.
From a publisher's point of view, the notion of convincing agencies and advertisers that in-banner and interstitial inventory has strategic value is directly aligned with their goal of selling more ads. After all, if pre-roll is selling like hotcakes and demand outstrips supply, you need more inventory to sell.
Reframing the issue to make it appear as if a pre-roll-only campaign is lacking the necessary reach to be effective suits that purpose. Selling more in-banner inventory means publishers can fulfill on more agency buys and sell remnant display, instead of falling short on large-scale buys and needing to turn money back to the advertiser.
But what if you want to skip the non-user-initiated ads and buy pre-roll video only? If you're looking at a large-scale buy against a simple demographic, you might have clearance problems. Many video ad networks have significant overlap within their respective pools of available pre-roll inventory. Making a large buy with one video network may mean that other networks have less to sell. To compound the problem, publishers may sell pre-roll ads through their direct sales force, further decreasing the amount of available inventory that's visible to a video ad network. In short, reaching scale with pre-roll only is difficult because publishers don't make inventory readily available.
For many large-scale video advertisers, extending reach into the audience of a particular site may be less important than reaching their overall target at scale. In other words, advertisers may care less about how much reach they're able to get on a particular site if they can reach their target across a video ad network. Therefore, reaching 13.1% of a site's unique users may not be a problem at all. In most cases, advertisers would be more interested in reaching people through pre-roll in relevant areas across the web than in extending reach on one particular publisher's site. In reality, if advertisers wants to reach more unique users on a specific website, they'll need the publisher to create more pre-roll inventory, rather than buy less-engaging formats masquerading as video inventory.
In the end, I don't think you can blame TubeMogul for understating the effectiveness of pre-roll and inflating the value of something publishers and video networks alike are ready to sell in abundance. But I do think we should all acknowledge the real problem here: all video ad impressions are not created equal, and publishers are simply not offering enough of the best pre-roll impressions.
|ABOUT THE AUTHOR|
Tom Hespos is the chairman and president of Underscore Marketing. He blogs at hespos.com.