Good Luck With Three-Screen Metrics

Invariably, whenever you speak to folks in the digital media industry about what’s holding back online video, measurement is cited as problem one, two and three. Often the complaints range from not being figure out basic reach — i.e. how many people watched a particular Web series — to tracking the audience of content that gets syndicated across the Web, to simply not grasping what actually constitutes a view.

I can see why some of these problems might prove quite challenging to solve. For instance, calculating reach for a viral video which gets shared and re-posted on all sorts of blogs and social networks sure seems like it would be technologically difficult. But one metrics obstacle that is constantly cited as a major deterrent for getting more big brands to spend more dollars online is the lack of cross platform measurement. And that challenge seems so fundamental to the medium’s growth that you’d think the industry would have tackled that long ago.

But you’d be wrong. Third-party audience researchers said they were going to tackle this years ago. The idea was that all media is going digital, therefore measurement will be digitally precise. Nielsen promised that it would be able to provide some sort of ratings equivalent for TV, video games, mobile phones and PCs. By 2008 the company was issuing its quarterly “Three Screens” report, which is full of nuggets such as how many people use DVRs or watch TV shows on the Internet. Meanwhile, in 2010 ComScore made similar promises, and is running pilot programs for three-screen measurement.

Nielsen’s three-screen reports are actually quite rich, it’s just that they come out once a quarter. I suspect that’s not very useful in day-to-day planning. From my best recollection, when Nielsen sold its three-screen capabilities years ago, it boasted of being able to track users across every screen far more frequently and intricately.

To be honest, I don’t have any idea how far along ComScore’s efforts are. According to a report from last year, its three-screen pilot program is in 25,000 homes globally. To push this forward, the company has joined with Arbitron as well as a coalition called The Coalition for Innovative Media Measurement, which includes media companies like NBC Universal and Discovery and brands like Conagra. But it’s hard not to have doubts, given that ComScore hardly has but a toeprint in TV audience research.

In fact, it’s hard not to have doubts about this whole endeavor, given the baseline problems of marrying disparate data sources and tracking different technologies. Put more simply, how are you ever going to track a person’s media habits on TV, PC, phone, tablet, video console and whatever else the folks in Cupertino invent next when none of them are interoperable (today) and three of the four don’t accept cookies? Is Nielsen going to convince anybody to install some kind of tracking software on four different devices? Is that even possible? Why would ComScore be any more successful? And if that isn’t likely to happen, is the online ad industry going to accept some kind of survey-driven, or worse, diary-driven methodology, even if it offered real time data? Doubtful. Sure, maybe if the cable companies, set top box companies, ISPs and wireless telecom companies all got together and shared data — and maybe I’ll end up playing for the Knicks this year.

It’s more likely that the answer is something nobody in the ad business wants to hear: it’s going to be impossible to ever track people’s media habits on every device, even if they are all digital, with very much specificity. Buyers and sellers are going to have to stick with formulas, guesstimates and mismatched metrics. And brands are going to have to decide if they can live with that. But hey, sketchy measurement hasn’t hurt TV much.

https://staging.digiday.com/?p=1220
Digiday Top Stories
Ad rendering preventing in staging

Ad position: web_bfu