Our new favorite media news source, Axios, is reporting that the Media Ratings Council (MRC), the folks who do accreditation for Nielsen, Arbitron and the like, is working to create a single standard for measuring video, whether that video is on TV or online. The new metric will be calculated by “measuring impressions (the number of times someone is reached) that are viewable by the MRC’s standards (viewed for at least 2 consecutive seconds), combined with how long each of those verified views lasts.”
This is a welcome development indeed, as few things harm the media industry more than the current system, where social platforms self-report billions of views that, upon closer inspection, consist of no more than a quick glance, often without the sound turned on, and then the vast majority of industry news outlets report those numbers as gospel, frequently comparing them unfavorably to TV’s Nielsen ratings, which measure average viewers per minute over the course of a 30 or 60 minute show, not average number of sound-off glances.
It is, as we’ve noted here before, like comparing apples to elephants.
A measurement system that holds everyone to the same standard would be a godsend, and we hope that advertisers hold the social platforms’ feet to the fire until they’re on board with this, because it’s definitely in their self-interest to put off implementation for as long as possible, if not forever.
Speaking of media measurement… TV[R]EV has been working on a report about smart TVs and ACR (automatic content recognition) that’s due out in a few weeks. It’s a joint contribution from myself, Dade Hayes, Dawn Chmielewski and Jason Damata. ACR data is fascinating in that it’s rapidly replacing set top box data as the primary adjunct to Nielsen.
It’s winning out because it turns out to be a lot more accurate: set top boxes and TV sets have different on and off buttons. It seems many people turn the TV off but leave the set top box on, and the unseen programs frequently wind up being counted as viewed. While set top box data companies claim to take this into account when smoothing their data, it makes that data inherently less accurate than smart TV-based ACR.
The second advantage to ACR is that it tracks whatever is on the TV, regardless of source. So that OTT viewing, set top box viewing, OTA, VOD and DVR viewing are all accounted for. Since ACR data can be processed in 24 hours or less, it further allows networks and advertisers to track behavior in something close to real time. That’s particularly valuable in smaller markets and with smaller demographic groups, where companies like Nielsen often rely on modeling to get results. ACR data can help refine the existing measurement data, providing greater accuracy along with greater immediacy.
There’s an interesting dynamic too, in terms of where the data comes from. The two largest players, Samsung and Vizio control two-thirds of the smart TV market. Samsung, the largest player, is a walled garden and as such, they do not share their data with anyone. That leaves Vizio, who supplies their data via Inscape, a wholly owned subsidiary. Other TV manufacturers—Sony, Sharp, Toshiba, Sanyo and Magnavox, among others, also supply ACR data, but they control much smaller pieces of the market, which, as we’ll explain, makes marshaling their resources far more complex.
We’ll be exploring all aspects of ACR in the report, everything from privacy concerns to how the data collection actually happens to how companies are using ACR data to fuel everything from ad placement to recommendations to attribution tracking.
If you’d like to be notified when the report is available, let us know.