With all the new capabilities ad tech can provide, it’s easy to get overwhelmed – and the same is true for your web browser.
Though most of us don’t realize it, when we visit an advertiser’s website it may be communicating with dozens, even hundreds of technology providers the advertiser works with. This includes ad networks, exchanges, analytics dashboards and offline data warehouses. In fact there’s so much third-party traffic it’s beginning to slow down browsers and it’s leaving advertisers with a mess of fragmented user information to untangle.
That’s the concern of 20 prominent ad tech companies, including the Rubicon Project, Casale Media, Dstillery, and Neustar, (see the full list here) which have come together to try and deal with the problem. They’ve formed a non-profit, called Digitrust, that aims to change the way websites communicate with third-parties, eliminating redundancies in the process and bring third-party traffic back down to a manageable level.
“Right now in this movement towards automation that we’ve seen … is you have more and more and third parties working together to automate the personalization of the Internet experience, targeting, all forms of advertising, analytics, attribution – anything that goes towards fulfilling the publisher’s and advertiser’s objectives with consumers,” explains Jordan Mitchell, vice president, product at Rubicon and interim CEO of Digitrust.
More efficient communication with third parties would have a number of benefits, the group says, including better privacy control and faster page-load times for consumers. For advertisers, it means more secure, efficient data collection and ad tech services. For providers, it means less of a headache dealing with the mishmash of partner and publisher data.
Too many third parties in the mix
In the early days of cookie-based web tracking, a site owner would use a pixel on their site to identify a user and follow them from page to page to see what they looked at. Data collection still works that way – except that today, the site owner often with dozens of different companies that all use their own pixels to collect and share different kinds of data. And those companies work with dozens of other companies, who work with…
What’s worse, each provider uses its own unique ID to follow the user. If an ad tech company wants to share information about a user with the site’s owner, they need a third “ID synch” pixel, to connect the site owner’s user ID with their own. If the company wants to communicate with another third-party provider, that’s another synch pixel to match those two IDs.
So if four parties (including the site owner) are active on a site, the user’s browser will see 12 of these pixels, and send out 12 separate server requests. If there are eight parties, the browser will need to make 56 requests. That number gets into the hundreds pretty quickly. And although each request uses only a tiny amount of the browser’s available resources, eventually the pileup starts to affect browser performance. On the other end, publisher and third-party servers are dealing with an escalating number of requests as users adopt more and more new devices.
“Third parties can’t work together until they can speak the same language, and that’s with the IDs that they use,” says Mitchell. “If everyone is using a different ID, then they have to synch those IDs with each other. That’s the leading cause of this increase in third-party requests on webpages.”
Doing more with less traffic
From Mitchell’s perspective at Rubicon, a supply-side ad tech company that runs a high-traffic ad exchange, the reality is a lot of time and effort committed to synching user IDs with other partners. They needed to create an ID synch for “every new DSP, every DMP that we worked with,” he says. “That’s just silly for us. I could just see the problem space that this was creating.”
He says Digitrust was conceived by a group of companies involved in the U.S. IAB’s “Future of the Cookie” committee, which released a whitepaper in January. The committee identified excessive third-party requests as a major hurdle for the industry, and one that would be difficult for any single company to solve independently. It was clear, says Mitchell, that a collaborative solution was needed.
But Digitrust hasn’t yet determined what that solution is. For now, the coalition is in discussion with members, potential members, privacy groups and regulatory bodies to figure out what makes the most sense and what would see the widest adoption across the ecosystem.
What’s not on the table is reducing the number of third parties that advertisers and publishers work with. Digitrust wants to cut down on traffic while preserving the services that third-party tech companies offer. The alternative is to come up with a way to make third-party traffic more efficient: to do more with less traffic.
One possible solution is a single user ID, implemented by Digitrust and shared by all participating companies. This would make synching user IDs unnecessary, and eliminate all the the leading cause of third-party requests. A single ID could have other potential benefits besides, like providing a single-user identity across desktop and mobile devices, and making it easier for companies to respect universal privacy preferences like the Digital Advertising Alliance’s Ad Choices program.
Mitchell says that despite some reports, Digitrust is not working on a single user ID and is still considering several other options suggested in the IAB whitepaper. He also says that for now, Digitrust plans to keep its goals very narrowly focused on reducing third-party requests. Other potential applications, like cross-device user tracking, won’t be discussed until the primary objective is accomplished.
Keeping a narrow focus is one way to avoid conflicts with members and other companies in the ecosystem that are working on proprietary solutions for cross-device targeting and privacy protection. “As a non-profit with a number of platforms as members, we’re not going to do anything that’s competitive. We want to fuel more innovation in the market, rather than stifle it.”
“The thing we don’t want to do is take pokes at the big, important first parties in the system – Apple, Google, Facebook, etc.,” he adds. “The problem that we’re solving is not something that they are as interested in solving right now. They’re in a different category in the ecosystem. They’re focusing on different things.”
The coalition plans to announce its proposed solution within the next few months, he said.