The average home has 10-12 devices, not including linear TV -- which has made measuring and monitoring what people watch infinitely more difficult compared to a time when people primarily sat in front of live TV. Currently, only 20% of TV watched by teens is live.
Now both comScore and Nielsen are introducing new strategies to better measure viewers, according to company reps on a panel Thursday during the 4As Transformation Conference in Miami.
In the new measurement era, the idea of “TV households” is rapidly becoming obsolete. ComScore now calls homes "devices" because of the rising flow of content streams pouring in and out of dwelling units from a plethora of gadgets. With its recent merger with Rentrak, however, the company believes it is well equipped to track and document all viewing, said comScore CEO Serge Matta.
And all the data provides a more detailed picture of who is watching, said Matta. “Instead of targeting 18- to-34-year-olds, we can now target 18- to-34-year-olds who are going to vote for Hillary and drive an Audi Q4," he says.
Also with a new partnership with Adobe, comScore will "ingest data from an analytics cross platform," says Matta. The results will roll out in phases with the first coming in April that will plot advanced demographics. "What car you bought, what you buy at the grocery store, how you vote in national election," says Matta. “We are putting all the pieces together."
comScore will also introduce more enhanced metrics in the second quarter, and in the third quarter, the company will report daily cross-platform information gathered on a census basis for TV and digital, and overlay this information with basic and advanced insights.
The results from early testing should be reassuring to traditional media. "Everyone feared that linear TV was declining," says Matta. "That is absolutely not happening. What is happening with fragmentation is actually viewing is increasing with other devices. It's on phones, over-the-top devices."
The next frontier is tackling ad blocking, says Matta. "It will happen probably early this year or the next. But first building blocks have to happen. [We need to figure out] what do people watch on every single device and then we can get into the ads."
Meanwhile, Nielsen is refining its measurement system to better reflect today's viewing environment.
All TV clients now have access to time-shifted and data measurement, while most media buyers will have similar access shortly.
Nielsen is also incorporating panel and census insights and is working on measurement of connected devices. More recently, Nielsen has started enhancing its core TV rating, providing daily program ratings through a full week of DVR watching.
In August, Nielsen will begin reporting ratings from day 7 to 35 on a weekly basis. "This gives the industry the flexibility to transact however it wants," says Nielsen's Lynda Clarizio.
More and more I fear that we are drifting into "cross platform" audience measurements that will be device uasge, not actual viewing based. The tendency is to gloss over this important distinction by referring to "viewing" and targeted "audiences", when for the key platform which accounts for 80-85% of the activity---that's old fashioned linear TV, folks---- we don't know who is or is not watching.
Just to be clear-- we at comScore don't call homes "devices." Rather, we recognize that your archtypal US household, with mom, dad and 2. kids, may well have 4 TVs, three of which are Smart TVs; 2 Rokus and an Apple TV; 4 smart phones, 3 computers, two tablets, and a gaming console. And that all 17 of these devices might be used to watch this week's episode of Gotham. So in order to provide a comprhensive picture of today's dynamic, robust and complex viewing environment, you can't jsut stick a people meter on a TV set and punch out early. You hzve to understand the dynamic of media consumption across devices and platforms, and to follow the viewing wherever it goes.
A note to Ed, one of my favorite Mediapost column commenters: traditional measurement systems are taxed beyond reasonable limits by today's fragmented and granular environment. But I'm optimistic that new technologies and techniques will help us to understand who is out there in front of the set, better than we have been able to understand before. This will probably require breakthroughs in primary person-based data collection, coupled with some very clever math.
It's an exciting time to be a media researcher!
Sorry. Just posting again so I can check he "notify me by email" button..
I agree that "TV households" is an antiquated universe. But in order to understand differences in viewing among devices and people, we still need to be able to break out the total universe into segments: TV only, pay tv, broadband-only,cellular-only, device-specfic, etc. And by device. And these designations are just the means to an end: content consumption. Which is neither the device itself nor the click that starts the video asset. It is the viewer--and in order to understand the viewING, we have to follow the viewER.
Hey Josh. I just want to clarify that Matta did say that comScore calls households devices. Whether that is official or not, it was what he said during his panel. And your further explanation mimics his comments about the reasons why.
I think that for the time being, a hybrid measurement approach will work best- metered data, census, panels, communities, etc. At the end of the day marketers need to measure their media efforts from the ideation stage through to actual broadcast and audience reaction. TV viewing has fragmented from a delivery perspective, and the service providers and device players have shifted to accomodate for that shift. Advertisers are still trying to make that shift, and need data and insight that shows the total picture of content consumption and how it differs between audience segments that can be teased out based on viewing pattenrs, not based on overall device usage patterns per household. As far as ad blocking being solved this year, I don't see it.
Randy, what advertisers and TV programmers need is viewing, not just device activity. The former---viewing---is neither a given, nor a constant when a device is in use by time of day, by demographic, by mindset and, especially, by viewing context--program content, location of viewing, the amount of ad clutter in a break, etc.