Commentary

Cost-Per-View Pricing for Digital Video: What Would It Really Measure?

Advertiser spending for online video has come to a screeching halt. Businesses believe that ROI assessment tools are unreliable and advertisers are skeptical of any new metrics have been recently introduced. This grudge towards ad metrics may or may not be valid, but one new metric bucks the traditional trend of using cost-per-thousand impressions (CPM) to measure digital video (and TV)-take it a step further to cost-per-view (CPV).

Both Google and TubeMogul endorsed this standardized CPV metric to the IAB's Digital Video Committee, and it's garnering a notable amount of industry support. With CPV, users are essentially in the driver's seat: they can choose whether or not to watch video ads. In this model, metrics only count if the viewer engages with the ad and clicks on the "skip" within the in-banner or in-player ad; advertisers only pay for the user-initiated ads. A proof point that proponents of CPV provide is that advertisers will now have a tool that measures effectiveness and will be able to create corresponding price model - potentially improving ROI. Please note my emphasis on "potentially," as it is essential to focus on the goal of your online video campaign when assessing its success.

From Google's supporting perspective, CPV is a natural extension of Adwords and delivers a value proposition that serves the interests of direct marketing. It's not clear if this works for brand advertising because if, like in Adwords, the decision to show an ad is based both on how much you pay and by the click-rate on your ad--established brands tend to get higher click-through rates regardless of the creative. Additionally, is your ad designed to reach your most loyal customers that will engage without much effort on your part? Or is it the objective to reach new customers to start building brand awareness? Honestly, who can say they only want to reach existing and engaged customers?

Promising, but is it enough?

The underlying principles of the CPV model are promising, and there's no doubt our industry needs to improve the way online video is measured.  We should continue to seek new ways to measure effectiveness and engage users - give them more control of their online viewing experience. Before we embrace it as "the new standard" for measuring online video advertising, we should consider some questions about the CPV model. For instance, how exactly does CPV apply to viral video? Viewers may choose to watch viral videos, but not necessarily because they're interested in the brand. Maybe they were lured to click on the ad because of a sexy thumbnail, or because they received a notice that their Facebook friends viewed the video.  Of course, the reality is that most people watch viral videos simply for entertainment, which doesn't necessarily lead to the sale of a product.  It's important to understand that CPV comes at a higher cost, as much as $0.20 to $0.25 for a video view, which is the equivalent of a $200 to $250 CPM for each actual viewing.  This raises the question, is the value of CPV for all advertising on viral video worth that much?

CPV does not measure effectiveness beyond the intended interaction, nor does it measure the long-term impact of heavy usage of new formats.  In reality, we still know very little about how viewers interact with online video content and current metrics are only a proxy for measuring effectiveness and discovering impacts to ROI.  CPM measures all impressions to determine efficiency, noting that some impressions are more valuable than others, while CPV measures fewer impressions, recognizing that some will be missed and that many are not valuable. The choice of metric should be based on the objective of a campaign.

The debate between CPM and CPV ultimately comes down to how we measure human attention and human response in relation to ROI. It's apparent that brands are starting to pay more for online video.  We will have to wait and see if brands that turn to CPV are willing to pay more for the actual view and potentially pay more than CPM.  Stay tuned and grab some popcorn while you're at it -- this debate will surely continue as more business cases come to life.

2 comments about "Cost-Per-View Pricing for Digital Video: What Would It Really Measure?".
Check to receive email when comments are posted.
  1. Josh Warner from Feed Company, May 28, 2011 at 3:49 p.m.

    Erick, Thanks for this piece. As the first video seeding company in the US, Feed Company introduced CPV pricing to the industry and our clients, such as Chuck McBride at Cutwater and RayBan, in 2007. The views were guaranteed so we HAD to find the audience and do our job as marketers. Our clients understood the value of a performance-based contract versus an impression-based contract. And RayBan certainly appreciated the millions of views, engagements and mainstream media that far surpassed our MINIMUM view delivery for "Guy Catches Glasses with Face" and other videos we seeded. Now I'm not saying that all of the 200+ video campaigns Feed Company seeded for major brands had the incredible viral lift of RayBan (although campaigns for Levi's, Taco Bell, Guitar Hero, Google TV, Haagen-Dazs were also hugely successful) but ALL successfully delivered user-initiated views from an audience that choose to watch the video. That's performance-based targeting and that's why CPV pricing is here to stay.

  2. David Off from ProWeb365, March 12, 2015 at 6:08 a.m.

    Great sharing! We may have some helpful cost calculator tool for Video Ads. However, Cost per view pricing usually depends on your goals and spend. The targeted the audience and the country you are targeting. Most video advertising on the Web today is made by brands whose goal is simple: convey a message to captive audience. While "views" should not be viewed in a vacuum and are not a perfect metric, they provide additional detail to measure and price paid media on that basis. This additional level of detail would make it easier for brands of all sizes to adopt video advertising online.

Next story loading loading..