Is 'Sharing' The New Viral -- Or A New Virus?

Have you heard the latest video advertising pitch?

"Buying impressions is dead, buying engagements is dying... the future is in buying video sharing!  You only pay when your videos are watched by self-interested consumers and they can share it with their friends!"  

Yes, "shared videos" are being positioned as the new "going viral."  And in a never-ending quest to find media that is completely controllable, yet entirely consumed by self-interested consumers (read: the holy grail) buyers are beginning to ante up.

Unfortunately, the only viral in this cleverly packaged snake oil is the virus you wake up with in the morning when you fully understand what you purchased.

Let me clarify.

First, video-sharing companies primarily syndicate content through display inventory.  For all the social, sharing and viral components of the pitch, they are all irrelevant relative to the number of views that will come from display.  So, what's positioned as video sharing is really cost-per-click display advertising. 



Second, video-sharing companies are event-based ad networks.  They may charge on YouTube views, user engagements, video completions or whatever metric they can create, but they are arbitraging the cost of the media they buy with the amount they are getting paid.  So essentially, what we're really talking about is paying for performance, which should be priced comparably with performance metrics rather than branded CPMs.


Lastly, and perhaps most ironically, video-sharing companies actually generate a very limited amount of video sharing at all. The total views generated via sharing often equal less than 5% of total views generated during the life of the campaign.

On the positive side, if a buyer's goal is aligned with the reality of a display-based content syndication world, then these models can actually be very successful.  For example, if you want to reach 1M views on YouTube, hit a large number of unique users quickly or behaviorally target a long-form video, display-based content syndication is for you

This, of course, cannot be separated from the issues of price and performance.  On pricing, we are really talking about cost-per-click, not guaranteed views, and this pricing ecosystem is relatively mature.  On performance, you must measure your syndication vendor on a post-click metric, the same way you measure search, or you will end up with garbage clicks (or, in this case, garbage stats).

On the negative side, , you are paying to get your video in front of consumers who most likely are simply bored rather than engaged.  Engagement cannot be measured by simply getting someone to start a video in a display ad.

In summary, if your goal is to drive authentic social sharing of your video content, the best advice I have is to focus on creating content worth being shared.  Anyone who claims to have solved this across the board for advertisers is simply sharing a sales pitch.  

3 comments about "Is 'Sharing' The New Viral -- Or A New Virus?".
Check to receive email when comments are posted.
  1. The digital Hobo from, November 9, 2010 at 4:54 p.m.

    Good to see an industry insider tell it like it is.
    Nice piece, Tod.

  2. Stephen Shearin from ionBurst Media, November 10, 2010 at 11:09 a.m.

    Great topic.
    A lot of semantics here. Sharing. Paid sharing is certainly different from viral sharing, but if it assists in the viral jump start of a campaign, there is certainly no problem with it.
    Gauging the public psyche is art and science. The Kardashians can drop a pencil and cause a measurable stir. Incredibly talented artists can produce amazing works of art and get 8k views.
    'Viral' is an apt, yet unfortunate moniker for this form of advertising. Connecting it to 'Virus' is too easy and, with all due respect, not accurate. Paid media to 'share' a hopeful campaign won't infect your hard drive. Viral should not be compromised with the term Virus in spite of their inherent relationship in the real world.
    The tragedy, as with all art, is that there are companies (famous US auto maker comes to mind) who can afford to put out marginal work and ram it down the Internetz throat and call it viral when it just isn't. It's blunt force trauma. You know viral when you see it. And to that end, I have to agree with the article's sentiment. Just seeing it doesn't make it viral without some genius behind it.
    RG/A gets it. W&K, though they spent 300mm all in to make Old Spice happen, managed to make it happen. Great creative, properly activated, goes viral.
    It's all about the creative. Great creative requires minimal paid media. But if great creative falls in a forest and no one sees it....
    So paid sharing has it's place, but doth not viral make.

    In sum, as with so much in the world, it's not either/or. It's simply another magic bullet in a VC funded world filled with magic bullets. If some sharing can help your campaign grow legs, I say smoke 'em if you got 'em.

  3. Scott Button from Unruly Media, November 22, 2010 at 5:08 a.m.

    Great post, Tod. Contra your 5% figure, Unruly's ratio of earned views to paid views is 8-to-1 across a thousand campaigns or so - that's 700% uplift. It sounds crazy, but of course the metric is skewed by outliers such as T-Mobile's 'Dance', 'Welcome Back' and Evian's 'Roller Babies'. With more modest content, we're usually shooting for a 1-to-5 ratio, earning 20% free attention from sharing behavior activated by the bought media. That may still sound unspectacular, but you have to remember that those views are coming from people who've been urged to watch the film by a friend or colleague, so default engagement and favorability levels are way higher than display media environments.

    You're totally right that a click-initiated view, charged on stream initiation is more-or-less a CPC-model. This is poorly understood, though it doesn't make the inventory collapse to classical display. As you say, the quality of clicks / views / stream-initiations can vary hugely, so any reputable media vendor ought to be agreeing post-click KPIs with the buyer and should be judged on them. Some of the events you mention such as completed views and organic shares are precisely such post-click events. (If vendors are following the IAB's definition of a video view, a stream initiation is strictly-speaking post-click, too, as it should be counted not on the click but after the video has buffered; subtle, but important, because low quality media can actually open up a material gap between the click-event and the buffered video beginning to play out).

    Like you, I think there are several reasons that media vendors ought not to be selling completed views or organic shares as the billable event: it puts too much additional risk onto the media owner and doesn't even solve the quality problem, because you then need furthur KPIs to control for the quality of the completed view or the share. Most of the vendors selling on a completed view model are simply acquiring media from virtual-currency networks, so the video may get watched to the end, but only because the viewer's being rewarded.

    For relatively strong content, Unruly focuses on dwell-time and rate of sharing as post-initiation KPIs. (Though a lot of buyers are still hung up on post-initiation CTR). Where content is less strong, it's still possible to set ambitious KPIs that are beyond-the-view, but we're only comfortable in such scenarios if we can augment the campaign with additional, site-specific or platform-specific content programs, integrations and tie-ins that are optimized to spark social media reactions. This is then quite a long way from banging videos into display ads and arbitraging on the exchanges.

Next story loading loading..