Automatic Content Recognition Yields Household-Level Data At Scale

If you think the closer you are to a television screen the better the picture is, the same principle applies to the automatic content recognition (ACR) approach taken by Inscape. Formerly known as Cognitive Media Networks before its acquisition by consumer electronics provider Vizio, Inscape is closer to the TV picture than a set-top box.

“We detect at the glass” attributes of content playing on-screen, says Inscape’s SVP of Marketing & Business Development Jodie McAfee.

Those attributes are considered a “video fingerprint” for detecting everything from movies to TV shows to games.

Inscape’s technology actually detects pieces of a screen and then reassembles them on an external server to identify a particular piece of content. The company licenses data, along with IP addresses.

There’s a reason Inscape detects at the glass. “A fairly solid percentage of the behavior on our footprint is non set-top box behavior,” McAfee says. “We see all of that behavior.”

Inscape’s reach is just over 7 million TV sets across the U.S. in what amounts to “fairly close to census-level” reporting. It ingests feeds from roughly 150 national broadcast and cable networks and about 150 local feeds from DMA’s; it’s aiming for all 210 DMA’s by mid-2018.

“We maintain a standing map report against Experian demographics that we run on a pretty regular basis so that we can give people a sense of what the footprint of Vizio owners looks like,” he adds.

Data are licensed on a use-case basis and reported every hour, although there is a three-hour gap between when something aired and when the resulting data are available. Applications range from analytics to retargeting and measurement for the buy-side and sell-side of the media business.

While there’s “no such thing as the perfect data set” because ACR technology can’t tell exactly who’s watching something within a household, “it’s household-level data at scale,” says McAfee.


Read more at Huffington Post.