The contentious future of ID bridging in digital advertising

In the shifting landscape of digital advertising, privacy concerns and the phasing out of third-party cookies have driven advertisers to explore new ways to track users and ensure effective ad targeting. Google’s decision to remove cookies for 1% of its users in early 2024 marks the beginning of the industry's transition to a cookieless world, forcing advertisers and platforms to explore solutions that balance user privacy and accurate targeting.

One such solution is ID bridging, a method designed to fill the gap left by cookies by using persistent identifiers such as hashed emails or mobile device IDs. This approach aims to provide more reliable and privacy-compliant tracking than traditional third-party cookies. While ID bridging offers a potential path forward, its use is sparking debate, particularly regarding its scalability, accuracy, and potential for fraud.

Demand-side platforms (DSPs) argue that ID bridging falls under the category of sophisticated invalid traffic, citing concerns about inserting IDs that do not belong to the native cookie. This ambiguity has raised alarms over its impact on campaign performance, with potential misattribution complicating marketing efforts.

ID bridging versus traditional cookies

Traditional third-party cookies relied on browser-specific, short-lived identifiers, offering a fragmented and probabilistic view of users across devices. This method faced significant accuracy challenges in cross-device targeting, personalisation, and attribution. Additionally, more industry critics are arguing that cookies are becoming obsolete due to privacy regulations and browser restrictions.

By contrast, ID bridging leverages persistent identifiers like hashed emails or mobile device IDs, providing deterministic tracking with improved accuracy across devices. This allows for more effective targeting, retargeting, and personalisation, while also ensuring compliance with privacy regulations. With its enhanced cross-channel consistency and long-term tracking capabilities, ID bridging is seen by many as a more reliable and future-proof option for advertisers.

An example of ID bridging in action is how a digital publisher can connect their audience in a cookieless environment like Apple’s Safari to a platform such as Google Chrome, offering consistency in targeting across environments that traditionally would not sync user data.

Unpacking the complexity

However, despite these advantages, ID bridging introduces complexity, particularly through the dual methods of deterministic and probabilistic matching. Deterministic matching focuses on accuracy by using consistent identifiers like hashed email addresses, which remain the same across devices. This approach, while effective, depends on users logging into multiple platforms, which may not always happen. Probabilistic matching, on the other hand, uses algorithms to interpret signals such as device type, IP address, and browsing habits. While this broadens the net for audience identification, it sacrifices some accuracy, relying on assumptions to match profiles to the same user.

Even though ID bridging offers potential benefits, concerns over its transparency and the potential for fraud are significant. Brands and agencies are encouraged to implement several measures to mitigate risks associated with spoofed or misleading IDs in the bidstream. Such measures include building direct relationships through supply path optimisation (SPO). 

Jon Hewson, managing director of Audience Store, explains, “Building direct relationships through supply path optimisation (SPO) is also crucial. This involves working with trusted publishers and authorised sellers using resources like ads.txt and sellers.json. Transitioning towards first-party data and contextual targeting is essential.”

However, not all industry leaders are optimistic about ID bridging. Niall Hogan, general manager of JAPAC at GumGum, for example, is not a proponent of the method. He describes it as a “desperate attempt” to circumvent privacy policies aimed at protecting consumer data. Hogan suggests that agencies and advertisers should demand more from their suppliers, taking preemptive measures such as using inclusion lists in OpenRTB bidding to limit potential fraud. He also advocates for a shift towards contextual advertising: “Rather than relying heavily on ID bridging, brands and agencies should consider shifting towards contextual advertising.”

Contextual advertising offers a more precise and privacy-compliant solution by delivering ads in relevant environments without relying on potentially flawed identifiers. According to Hogan, embracing contextual strategies ensures campaigns remain effective and better aligned with current privacy standards.

What's next for ID bridging?

The future of ID bridging in advertising remains uncertain. In May 2024, the IAB Tech Lab initiated a collaborative workstream involving over 80 participants from 40 companies. By September, they released a recommendation for new OpenRTB specifications that introduce fields to disclose bridged IDs and their origins. This added transparency allows advertisers to evaluate the quality and integrity of bridged IDs. Partners delivering low-quality or spoofed IDs can be identified, enabling advertisers to reallocate budgets to more reliable sources of data. Implementing frameworks based on the transparency and quality of IDs further streamlines the process, reducing the risks associated with misleading identifiers.

Anoop Ramachandran, chief technology officer of Preciso, sees these updates as a positive step forward: “The new ID origin fields offer greater confidence in cross-device targeting by clarifying whether the bridged IDs are deterministic or probabilistic. This understanding helps advertisers better assess the accuracy of cross-device matches and optimise their strategies accordingly.”

Despite the push for transparency, the experts' consensus remains advertisers should approach ID bridging cautiously, given the ongoing debate about its accuracy and effectiveness. While the updates from IAB Tech Lab may provide a more transparent framework for evaluating bridged IDs, the key question remains: Can these IDs reliably scale and achieve precise targeting? The debate continues.