‘Not the future’: European publishers remain steadfast in blocking alternative IDs to third-party cookies

Working with identifiers feels like more of the same for these publishers. It’s like giving their data away to advertisers without having any real say in it. After all, ad tech vendors keep their ID solutions all hush-hush. Until things change, these publishers aren’t budging.

“We don’t believe in bringing back the world of yesterday where publishers weren’t compensated for their reporting, journalism, and contribution to society, but only for their ability to deliver useful cookies for profiling and targeting,” said Thomas Lue Lytzen, director of sales and ad tech at one of Denmark’s biggest news publishers Ekstra Bladet.

To prevent a repeat of such situations, the publisher refuses to embrace most cookie alternatives, whether they are deterministic (email-based) or probabilistic (fingerprint-based) IDs. Instead, it is using either Xandr’s ID or a publisher-provided ID solution to keep campaign fundamentals such as frequency capping and encrypted targeting  without relying on third-party cookies. 

The reason behind this choice is simple: Lue Lytzen understands how both IDs impact Ekstra Bladet’s data. Since nothing happens without a signed contract, he isn’t worried that an ad tech vendor is going to arbitrage it or worse.

“There is room for one-to-one IDs where a publisher shares an encrypted ID with individual SSPs or other demand partners, and that ID can only be used for specific purposes and can neither be shared with others nor be used to build third-party profiles,” continued the media exec. “Technologies that understand and support that, will be the winners in our industry.”

Norwegian publisher Schibsted shares the same sentiment. 

“For a while there might be pockets of inventory where probabilistic solutions might work, but the scale and transparency just isn’t there,” said Christer Ljones, head of data at Schibsted Marketing Services, the Scandinavian media group’s advertising arm.

One outlier that has piqued Ljones’ interest is the TrustID solution: it’s being built by Europe’s four largest telcos — Vodafone Deutsche Telekom, Orange, Telefónica and Vodafone — to support a pseudonymous token built on the back of an IP address that is then monetized for advertising purposes. Ljones also has a long-running interest in cross publisher logins, which would also help facilitate subscribe bundle cooperations.

“For an ID solution to succeed it needs a constant personal identifier, and to get that and remain compliant you need an interface to the user, a brand and a value proposition. Google, Amazon, Facebook and publishers login qualifies, but the cross-interoperability is an issue for the buyside,” said Ljones.

The same goes for Schibsted. 

Recently, the Nordic publisher has been making its first-party IDs widely available to advertisers to buy its audiences. It grew tired of waiting for advertisers to place their bets on an alternative to third-party cookies, so it made the choice for them. 

In fairness, moves like this are likely to be few and far between. That’s because they only really work for those publishers that can sell enough ad inventory with their own ID solution and consent. The scale and unique data needed to do this isn’t exactly ubiquitous among publishers. Very rarely do publishers have logged-in users. 

Be that as it may, the ones that can, remain resolute in their stance. 

“We don’t believe that any third-party intermediary that a human doesn’t know what they do should have the keys to the [our] kingdom,” said the digital lead at a publisher in the U.K. at a recent event in London where conversations were held under the Chatham House Rule. “So I’ve not engaged with any of the ID solutions because we think that it’s just a race for ad tech to try and replace the money they’ve been putting in their pockets for years via third-party cookies. They want to replace those cookies with something else that could be just as bad, and won’t benefit publishers.”

Another publisher at the same event said the following: “We don’t support ID graphs and we don’t believe in feeding into any ID cooperatives — i.e. things that use probabilistic mapping and identification. Solutions like this aren’t sustainable. They just rebuild the open web as it currently stands and we don’t really like that as a marketplace. 

When it comes to selling audience data without relying on third-party cookies, the rest of the publisher market is playing it safe. They have been actively experimenting with various deterministic and probabilistic identifiers for about six to 18 months, learning and fine-tuning their strategies.

While these publishers share the same mindset as Ekstra Bladet, Schibsted, and others, and they don’t want to miss out on potential ad revenue from currently available identifiers, the reality is that the amount of ad dollars spent through ID solutions like ID5 and Unified ID 2.0 is not significant. 

This situation is expected to persist for a while, mainly due to the ongoing search for scalable alternatives in the advertising landscape. It’s a classic chicken-and-egg issue: ID owners need marketers to transact using those IDs, but there’s little interest from marketers as long as there are not enough premium publishers actively promoting those identifiers in the programmatic marketplace.

Maybe, this is also why publishers are bullish on not getting swept up in the ID stampede. They don’t have to because advertisers haven’t. If they were, then the chances are the choice to avoid them wouldn’t be so easy. 

“Cross-site IDs, be they hard identifiers such as email addresses or be they probabilistic, are not the future,” said Lue Lytzen. “People advocating for them seem to have closed their eyes to strong signals from lawmakers and regulators, and concerns from users that don’t like being tracked. As publishers we are first in line; the ecosystem leaves it to us to obtain consent, so we must be able to explain and defend such practices. That, I believe, we cannot.”