EDO, a three-year-old venture based in New York and Los Angeles, is focusing on machine learning techniques to increase the capability of linking TV watching and user behavior.
EDO, which has some big fishes of the media world, watches television feeds in a big DVR ranch and finds out how ads are related to users purchasing behavior online. Its DVRs have taken 47 million “airings” of TV ad spots in the previous three and a half years. According to the venture, that record of ads is matched with open source data available of people’s online doings, such as searching for items on Wikipedia, a data set of “trillions” of user behaviors.
The data science compares the Ad airing and different behaviors’ triggering that links with that brand or product, developing a sort of “A/B testing,” checking which advertisements among the many from a brand marketer had greatly triggered the behavior.
On Thursday, EDO introduced the latest round of grant, $12 million in funds from Breyer Capital, operated by Jim Breyer, ex-Accel Partners venture financier.
CEO of EDO, Krim, and the lead tech officer, Joshua Lee, have explained the venture’s usage of Artificial Intelligence (AI) to get something out of the tones of data that they have.
“We had an auto client who was launching their first-ever subcompact SUV,” explains Krim. “That client was not able to look at their own history [of advertising] for that product; they couldn’t see based on their own experience which [cable TV] networks are most effective to generate engagement.”
“With our help, they were able to look back at a very specific set of campaigns by competitors.”
“What’s so cool is that we can contextualize data for the entire industry, by segments such as luxury versus non-luxury, and SUV versus non-SUV,” says Krim. “Clients are able to get a nuanced view by looking at their data but it’s only by looking at entire data.”
Kim further added how this will help in saving money:
“We had an automaker launching a fully redesigned version of a flagship model of theirs, and they were over a long weekend this summer running six different ‘creatives.’ We could show them that between the top- and bottom-performing spots there was a 50 percent difference in impact — the top ads were 50 percent more likely to engage a person than the lower performers.”
As a consequence, the automaker had the chance of maybe investing more money in top performers, and maybe reduce their budget allocation for the lower performers, he said. This selection would create almost about 240 more prime-time ads.
EDO is unclear about how it has collected that much data but it doesn’t include any special deals with Google or other internet corporations. The user data is grouped and anonymized so it is a statistical behavior sampling, no solo person is being studied.
CTO Lee claims that the techniques used for analyzing data are important. EDO data science team is using many machine learning techniques, from linear regression to “more advanced techniques” including Random Forests and Google’s TensorFlow, Lee claimed.
