Every time a page loads, so do dozens of scripts that allow programmatic players to sync their identifiers with each other, slowing the load and causing audience loss.
To solve the multiple identifier challenge, DigiTrust is bringing together a nonprofit co-op of platforms and publishers to create a single ID for demand-side platforms (DSPs), supply-side platforms (SSPs), data management platforms (DMPs) and data exchanges.
In the form of a publisher first-party cookie, the single ID would dramatically reduce the audience loss that happens when two parties match their IDs. Since SSPs, DSPs and DMPs wouldn’t need to drop and sync separate cookies, a single ID would also speed up page load times.
A single ID already exists on mobile apps for Apple (IDFA) and Android (AAID) devices, but desktop and mobile web browsing started with separate IDs. Inertia kept these IDs apart even as RTB added more complexity and slowed down the user experience.
“The rise of programmatic has forced this to be an acute issue,” said Digitrust CEO Jordan Mitchell.
Ten publishers have committed to use the single ID, and another 35 will do so as long as the platforms buy in.
Several platforms also have committed – provided that the publishers deploy the tech. That presents a chicken-and-egg problem that must still be solved if DigiTrust is to fire up the ID in Q4 through Q1 2017 as planned.
One of the platforms that has committed to joining the co-op is Index Exchange.
“The inefficiency of cookie matching is detrimental to user experience and costly to platforms,” said Steve Sullivan, VP of partner success at Index and the former ad tech VP at the IAB, where he worked with Mitchell on early efforts to solve this issue.
To get the project to the broad buy-in level it needs, DigiTrust needs to have many more conversations with industry players and privacy groups to explain the benefits.
Speed would improve. During an anecdotal look at 40 premium publishers, 37 trackers loaded on each page, making 85 third-party requests and slowing down load time by 6.1 seconds, Mitchell said. Although pixel syncs don’t use a lot of data, browsers can only handle a limited amount of requests at a time, creating a choke point. A standardized ID would cut those requests by 75%..
AdExchanger Daily
Get our editors’ roundup delivered to your inbox every weekday.
Daily Roundup
Reducing data loss would yield even more significant benefits and increase revenue for publishers since advertisers could find more high-value users. A universal ID would produce a 100% audience match, Mitchell said, compared to 60-70% today.
A 100% match rate would protect audience scale. During the ID sync from the web browser to publisher DMP, roughly 65% of IDs survive. During the second hop to the SSP, the audience match declines to 42%. Once the ID syncs with the DSP, the match rate falls to 27%. Three-quarters of audience scale disappears in this scenario, Mitchell explained.
With a universal ID, advertisers would know who they are bidding on, driving up rates for media on the mobile web. That revenue bump will deliver the biggest benefit for publishers, Sullivan said.
A standardized ID would also give publishers and platforms an advantage against the two digital advertising behemoths.
“Google and Facebook don’t have this problem,” Mitchell said. “The walled gardens have a single tech stack where it’s 100% [ID match] all the way through.”
Similar to how Google and Facebook drop first-party cookies, the universal ID will work by using a publisher’s first-party cookie. First-party cookies are treated differently by desktop and mobile browsers; some browsers block third-party cookies, for example, but not first-party cookies.
The universal ID will load with an opt-in box, similar to the cookie consent pop-ups on European websites. That’s one reason why Mitchell’s seen interest from UK publishers in particular. For others, support boils down to better revenue and user experience.
But a single ID also solves problems for platforms and users, Mitchell said. “Common identifiers happen to benefit everyone.”
