A VC« Whencraft? | Main | On A Bus Somewhere In The Rockies »
If Not Pre-Rolls, Then What?
There's been a ton of conversation on this blog and elsewhere about how to monetize web video.
I am not a fan of pre-rolls. I think they will cause many of the viewers to move on before ever getting to the video they wanted to watch. Post-rolls are also problematic because many web videos are not watched through until the end. Mid-stream or mid-rolls are more interesting to me and will probably emerge as an important monetization system over time. But the technology to do streaming ads/mid-rolls isn't widely available yet.
So what to do if you've got a hot web video property and want to monetize it?
I really like what Wallstrip did today with Lifelock.
This harkens back to 1950s TV (the show itself harkens back to 1970s TV). Lifelock got a blipvert (the 3 second pre-roll ala YouTube) and then branding throughout the show. And then a post-roll which could have and should have been way more informative than it was. After seeing Lifelock all over the show for four minutes, I was wondering what they do. Lifelock provides indentity theft protection. I could have used 15 or 30 seconds on that at the end with no problem.
But this is an approach that makes a lot of sense to me. It's not automated. It doesn't scale. It can't be targeted. And so it has it's limitations. But for where we are right now in the world of web video, it's a pretty clever model.
If you want to advertise on Wallstrip like Lifelock did, contact Howard. I am sure he'll be happy to hear from you.
Comments (21) | Posted February 15, 2007 in Venture Capital and Technology
Comments
not withstanding what wallstrip did today (which I think is extremely cool) i think the answer to a killer ad system for video (ccc or professional) is a targeted ad insertion for instream. targeting will come from a combination of speech recognition, tags, metadata, page links, surrounding web page data, and clever voodoo magic.
the ratio of ad unit:video stream being 1:1 feels wrong not to mention the disruption of pre rolls.
Posted by: bijan | Feb 15, 2007 1:44:10 PM
"It's not automated. It doesn't scale. It can't be targeted."
You sure about that? Very similar things, maybe not as deep but close could be automated, scaled and targeted.
Posted by: PRoales | Feb 15, 2007 1:47:48 PM
Thanks - lot's had to be put together to show the model.
The host was phenomenol as well.
The postroll ads were just not done in time but lots of viral stuff coming for our sponsors - so yes call me.
The footer had the link and the special offer, definately not perfect but a good start for sure.
Lifelock is a killer product for viral internet sales
Posted by: howard Lindzon | Feb 15, 2007 2:03:52 PM
I think you're right about mid-rolls, though surely they're also vulnerable to fast-forwarding, but my question with the sponsorship model would be the same question I have with traditional sponsorship. Is the sponsorship allied with the content or is it just a case of slapping the logo on the lectern and hoping that people engage with that? From a marketing perspective I'd have to wonder whether the viewer would "wonder what the sponsor does" if there was a sponsor on every web video they viewed.
Posted by: John Dodds | Feb 15, 2007 2:10:35 PM
The video is awesome!
Two related comments:
- The fact that LifeLock bar is displayed on the bottom is good, but it is not fundamentally different from the banner ad approach. The amount of information conveyed this way is by far less than with the pre-roll.
- If they could make the sketch revolve around the fun topic AND the company's product or service then this whole thing becomes great. I.E. back to commercials that are fun to watch.
Alex
Posted by: Alex Iskold | Feb 15, 2007 2:37:25 PM
Thanks man! We put alot of time and effort into the development and execution. The one other comment I wanted to add, is that this is an evergreen piece of advertising for LifeLock - a week from now, a month from now, a year from now - they will still be getting a valuable return on their investment.
Posted by: jeff Marks | Feb 15, 2007 3:21:01 PM
The Wallstrip crew, from Howard to Adam/Jeff to Lindsay is doing an awesome job of keeping quality first, expanding audience and figuring out how to monetize without offending the audience that they have built.
It's an extremely difficult balance to strike.
I pity the cold and timid souls at nyRAG who mocked Fred, Howard and everyone else who has invested in Wallstrip (one way or another)...they are doomed to never know victory nor defeat.
Posted by: Andy Swan | Feb 15, 2007 4:39:27 PM
It's an interesting notion that the advertiser could actually 'add value' to the overall experience of the content. With the current structure and approval cycles of most big brand clients, it wouldn’t be doable (and I would be skeptical on the value side) but who knows how things will change - if they want to reach the market badly enough...
Posted by: Leigh | Feb 15, 2007 5:04:27 PM
Fred...
Our company, Castfire, will insert advertising, promos, and branding into any podcast or videocast (mp3, mp4, flv - for online playback) into any position. The content producer can template each media segment, allowing for complete control of the final episode. In addition, it works real time across the entire library, allowing the content producer to take advantage of their evergreen content.
I think the days of only being able to do post-roll or pre-roll are numbered...
Posted by: Brian Walsh | Feb 15, 2007 5:32:22 PM
This is definitely a step in the right direction. The days of delivering "ads" with video are numbered in my opinion. Ads don't have the transparency to last in the internet world. Product placement will be the key. Finding the right balance of affectors will be the challenge. Examples? Movies sell a lot of music. Jennifer Aniston sold a lot of hair cuts. Much like I said about the future of music, the money will be in who can monetize/mobilize the meta data.
Posted by: NICCAI | Feb 15, 2007 5:47:27 PM
"Mid-stream"? "mid-rolls?" "instream?"
You mean plain old invasive ad spots, right? Same as the ones that have interrupted Uncle Miltie and the rest for the last 60 or so years?
I'm as much of tech jargon junkie as anyone, but come on -- can't we just admit that the old model was never broke and didn't need fixing after all?
IP video is an awesomely exciting thing, but the revolution is about increased quantity and diveristy of content and search algorithms. Disruptive, annoying, invasive -- interrupting -- ad spots work as a revenue model specifically because they are disruptive, annoying (read: memorable) invasive.
Undoubtedly some smart cookie soon will create an IP video ad insertion system that prohibits scanning thru the commercial interruption (e.g. bars the streaming of the remaing cointent unless the ad is viewed)...
Posted by: Grand Egress | Feb 15, 2007 7:57:47 PM
Fred,
I read your blog in bloglines, and for the past few times that you have posted Wallstrip videos, the video that appears in the feed is different to what I see when I come to AVC. The feed video is a short (but longer than 3 second) low-res video with the text "are you afraid of the dark". There is no Wallstrip that follows.
Does anyone know whats going on here?
Am I missing something?
Posted by: josh | Feb 16, 2007 8:48:30 AM
We are just getting this coplaint from this site. we are looking into it Josh -
Posted by: howard lindzon | Feb 16, 2007 9:13:15 AM
I have been very impressed with the ad units that VideoEgg has created. They focus on opt-in advertising and upsell the advertiser to the user using a ticker along the bottom of the screen or an image at the end of the clip (sounds like revver, but I think that there are subtle differences that make it better). The ability of a video ad network to monetize this solution depends on their ability to target ads properly, understanding the consumer's purchase intent and brand preferences. However, if they can successfully do that, they should be able to command a premium CPM for each ad view. Pre-rolls, mid-rolls and post-rolls are all intrusive and I believe that they will damage audience size unless the content is highly sought after (i.e. NOT bite-size clips). The problem with product placement is that it is not a scalable business: there is no standard measurement format, the nature of product placement can vary greatly from program to program (how deeply it's integrated, how often it's on screen, etc.) and there is a lot of friction in making a large ad buy through product placement (no network).
Posted by: Ephraim | Feb 16, 2007 10:22:27 AM
It's a sponsorship model as old as the hills - and still very much alive in traditional video media (that is, television.) The difference of course is the scale of the buy.
When a show or an event costs millions to produce and it is watched by millions of people, a sales team can sit with with an advertiser and build packages with titling rights ("blipverts"), in-show branding, and attendant media for sponsor ads throughout the show, as well as co-branded ads airing on the network outside of the sponsored property. Add things like contests, prizing, and online integration to grab viewers and keep them engaged. Take a look at the Super Bowl, American Idol, The Real World, or any of a hundred other shows to see the model in action. It takes a lot of time, coordination, negotiation, and work to get great sponsorships to work but when they do they are seamless, entertaining, and wildly valuable to both advertisers and media companies alike.
I love that Wallstrip went this way with LifeLock (I did check out the LifeLock site, btw, and am actually considering trying it out...) but with all the time and hands-on effort it takes to get these right I've got to wonder how this model scales across tens of thousands of web video properties.
Posted by: Greg Clayman | Feb 16, 2007 12:02:32 PM
Lifelock is an incredible simple and usefool prevention tactic.
You should definately try it out and yes I am an investor, but 50,000 paying customers in 12 months are proof that there is a need for this.
Amazing group of dedicated people to this nightmare. It's not so much just the prevention it is their help dealing if there is a problem. read the testimonials.
use wallstrip as a code when signing up :)
Posted by: howard Lindzon | Feb 16, 2007 12:11:36 PM
I want to second the comment about VideoEgg. I think more than this kind of sponsorship, they have a model where there are multiple opportunities to opt-in giving the advertiser and the viewer choices. I also think, just as companies insert the yellow line on a football field or a logo on a tennis court, virtually and for a TV audience, the same thing can happen here. The LifeLock podium is a pretty stable element that could be branded differently at different times and for different audiences. I assume this is what Castfire does.
Posted by: Michael Hoffm | Feb 16, 2007 1:06:06 PM
Seems like the scalability, targeting and automation are not far off (if not available already). There are already several companies working on targeted TV advertising (like www.visibleworld.com, for one). Couldn't the sponsorship on the podium become like the advertising behind the batters when you watch a baseball game on TV?
Posted by: Jeremy | Feb 16, 2007 5:08:10 PM
The embedded sponsor model doesn't really work once you get a little ways down the long tail.
Posted by: Erik Schwartz | Feb 17, 2007 11:44:16 AM
As you say Fred, this is cool for where the industry is now, but we need scalable models if anyone is going to make any real money. I think you are on the right track here too - I haven't heard of any ideas other than midstream ad insertions that sound like winners. I also think that the way forward will be to offer choices between paid for no-ads video and free to watch ad-supported video. That will start to make the value of our time clear.
Posted by: Nic Brisbourne | Feb 18, 2007 5:01:24 PM
i've seen the solution while watching a torrent, no less:
http://mattishness.blogspot.com/2007/02/internet-video-ad-model-solved.html
Posted by: mathew | Mar 1, 2007 2:01:22 AM
Main RSS Feed of All Posts
VC and Technology Posts Only
