Verizon, or more specifically it’s newly formed digital media division Oath, has an intriguing new plan. It wants to put together a wireless data powerhouse in its quest to challenge Facebook and Google in digital advertising.
Specifically, the company wants to rent reams of consumer data from major competitors like Sprint, T-Mobile, Vodafone and Telefonica that can be used for ad targeting.
Call it the Hulu of wireless data.
Besides the unusual competitive politics involved in such an undertaking (who’s in charge? How are we getting paid), it’s worth asking, what would such an arrangement look like when it comes to execution? Just how hard is this going to be from a technical perspective? Business Insider talked to some experts about what Oath will be facing in putting such an endeavor together.
Here are some of their big questions:
How hard will it be to mix different data from different wireless companies?
Austin Wright, senior vice president of strategy at the data-centric marketing firm Ansira, said that just stitching together the multiple data sets together will be a Herculean challenge, since each provider “may be supplying data at different levels.” For example, one carrier may supply more granular numbers on its customers, while another may provide summarized data, or data on broad segments of consumers.
“Let’s say Verizon’s data is tracked on zip code, and T-Mobile is based on households,” said Jeff Smith, chief marketing officer at LiveRamp, which helps marketers move data between different systems. “That’s not that simple. It’s not just a few more columns in the spreadsheet.”
Plus, some carriers may estimate portions of their consumer data, Wright said, while another may focus on more raw numbers. And some data may come from households with multiple accounts, such as family plans.
“All of which further complicates matters,” Wright said.
How with the data integration actually work?
Smith said that Oath and its would-be partners have to figure out whether they will mix and match the data themselves, or do so virtually using a third party tech company. Doing it yourself can be messy, he explained.
You either need a third party that has built software designed to reconcile reams of data, or you “need some data scientists,” he said. Smith pointed to comScore’s acquisition of Rentrak last year as a model of how big companies with big data sets were able to mesh them together quickly with the help of third parties.
Smith estimated that working with an outside vendor could take Verizon and company six to nine months to make its mixed data sets useful, while the DIY option could take more than a year.
Beside the data blending, what about security?
Steven Wolfe Pereira, chief marketing officer at Neustar, a tech company specializing in data management, noted the recent string of cyber attacks as making security concerns even more paramount.
That means that Oath and friends will not only have to make sure that nobody’s real names or addresses are shared among partners or with the outside world, but also have to take steps to ensure that whatever consumers have or have not permitted individual carriers to track is respected even as advertisers look to run big campaigns that employ data from multiple carriers.
“For example, there will have to be a way to make sure that advertisers can control how frequently they show people messages and don’t creep people out,” said Pereira.
Another security issue of importance is making sure that carriers don’t learn too much about their competitors’ customers. “You don’t want Sprint to start prospecting T-Mobile’s customers,” said Smith.
All that being said, if Oath actually pulls this off, will marketers be interested?
“Yes!,” Wright said. “Most advertisers are continually serving up most of their ad budget to a handful of sources …They want other options.”
“Email addresses, web cookies, these are fractional identifiers for marketers,” said Pereira. “But the phone number is one of the most persistent identifiers, and it can be tied to a bill in your home. You can theoretically track people’s online and offline behavior.”
That has powerful applications for ad targeting, he said. “When it’s anonymized, and done the right way, it’s a really smart play.”