The Case for Online Video Ratings, Revisited

 

Let’s say you’re an advertiser (on either the agency or client side) who’s in charge of a multimillion-dollar budget for a new car launch. The launch is an event. Your first stop on the media train will likely be television. And when you’re deciding what to buy on television, the first thing you’ll likely look at are all the forms of ratings the networks, and Nielsen, make available. The next thing you’ll probably consider is the content, making sure it’s programming you’d like to be associated with. After considering several other factors, you’ll make your decision in much the same way advertisers have for dozens of years.

Now, let’s say you’re that same advertiser looking to buy ads around video content. Essentially, you’re relegated to either run in-page video, go publisher-direct with in-stream video on a handful of millions of Web sites, or do the same with a reputable video ad sales network.

The science involved? None, really.

Impressions are bought against numbers publishers report (which may or may not be audited by a third party). The perceived value is in the number of videos streamed, regardless of the content it runs around. Click-through on companion ads may provide an added bonus, but the reason you’re likely buying in-stream video isn’t for companion ad performance.

You don’t have to look too hard to see a tremendous leap in logic and procedure when it comes to buying television ads versus buying online video ads. Though they’re two different media, the same companies are buying both and ultimately will want to compare performance and impact. But how do we do that when measurement methods are so disparate? One thing is certain: as online video rapidly matures, we’ll need at least one accurate, accountable, accepted third-party ratings system for video content — and advertising. ComScore has a product it launched in 2005 for this, but why isn’t it more widely used? Why isn’t it part of every media buyer’s or advertiser’s online video buying process?

There are many reasons online videos can and should be rated. Advertisers have to make apples-to-apples comparisons with other media, not to mention with online video itself. We must be able to (relatively) compare video on sites that make up both the head and the long tail of Web content. The same was done with broadcast versus cable television many years ago. The Internet must follow suit.

The need for ratings is further justified by the proliferation of online video “events,” such as Live 8 on AOL, March Madness on CBS SportsLine, and Super Bowl commercials just about everywhere else. Audited streams and demographics would be invaluable in planning advertising against the next iterations of each of these events.

Interestingly enough, online video content becomes more like TV programming every day. Shows are syndicated across multiple sites, with ad networks acting as an inventory aggregator. They sell ads against the series, regardless of the site it’s running on.

On the other hand, we should also be able to gauge the success of standalone, nonserialized content, such as viral videos. Technically, the “Pokemon Theme Music Video” on YouTube would outrate most video programming on the Web. I can name a dozen advertisers that might want to be around that.

Because the Web is an accountable medium, it’s also held to higher standards, often unfairly. But if the advertising economy does ever take a hit (it inevitably will), the first media likely to lose budgets will be the unaccountable ones. That’s why online video ratings should ideally feature rich measurement criteria, such as percentage viewed, number of repeat viewings, number of sends or forwards, and favorability. The technology exists to make this happen, but the methodology needs to be refined.

Current methodology allows for data to be collected by panels. Granted, that’s how Nielsen has done it for years, but that has gotten it into a quagmire with numerous interest groups (not to mention advertisers) who question the method’s legitimacy and accuracy. If we limit ourselves to panel-based data collection for online video ratings, we risk the same problems.

Is the answer to have publishers adopt coding that can be appended to all video, allowing for centralized data collection? Is the answer really a panel-based measurement scheme that’s supposed to be representative of the online audience?

Ultimately, the best answer will be a hybrid model that allows for both real-time and panel-based measurement. There are companies that are ready to challenge Nielsen in digital cable and IPTV. With so much content delivered digitally, logic suggests data can be captured in real time. The Internet industry must pioneer this if we expect to be the most accountable medium moving forward, both for brand and ROI (define) advertisers.

Meet Ian at Online Video Advertising Forum in New York City, June 16, 2006.