Skip to content
Ideas Data Strategy

Adobe Audience Manager without third-party cookies

Thomas Gilbey

Ad tech is readying itself in the lead up to 2022 when all major browsers will complete their phase out of third-party cookies. For clients using Adobe’s Audience Manager, one of the stalwarts in the  DMP space, the following outlines how identity will change in the platform, the value of the DMP going forward, and what brands can be doing to reduce their reliance on third-party cookies in their DMPs.

How does Adobe Audience Manager currently use third-party cookies?

Adobe Audience Manager (AAM) uses its third-party ‘Demdex Cookie’ as the primary (although not the only)  profile identifier in the DMP. The Demdex Cookie (like all third-party cookies) is defined by its ability to be set and read by domains other than the site a user is visiting. From the DMP perspective, this creates several benefits including:

  1. Where permitted, DMPs can collect audiences who have been exposed to ads without visiting a brand’s site, providing log files for multi-touch attribution and frequency management across channels.

  2. Third-party cookies provide the DMP with an online identifier to match to first party data such as site visitors or offline customer profiles. This allows audiences managed in the DMP to be shared with integrated platforms such as commonly used DSPs and ad servers for targeting or suppression.

  3. Third-party cookies are the key identifier in most  Audience Marketplaces, including Adobe’s. This provides marketers with the ability to leverage third-party audiences  for targeting, modelling, and combining  with first party audiences directly in the DMP. 

How will Adobe identify users without a third-party cookie?

AAM doesn’t solely use third-party cookies in the DMP. It can utilise other forms of device and people-based identifiers.

Without the third-party cookie, Adobe is pursuing  two primary mechanisms that can be used to create targetable audiences in AAM:

  1. Device identifiers, including Apple’s IDFA and Google’s GAID, which can be used for in-app targeting. However, with Apple’s impending rollout of iOS 14 and subsequent changes to IDFA collection, this solution may not be suitable for all purposes. 

  2. People-based identifiers such as hashed emails, which are used today to  match audiences to major platforms like Facebook and Google Marketing Platform. Curiously, at the time of writing, hashed mobile numbers can’t be stored in AAM. As a result, AAM  will likely have lower match rates when compared to environments that support matching on mobile numbers.

In the future, where will AAM be able to push audiences?

In terms of people-based destinations there are currently three platforms where AAM can push audiences - Google Marketing Platform, Facebook and Linkedin (Twitter is also on the roadmap and expected as a fast follow).Over time, we expect other destinations such as Amazon, Pinterest, and Snapchat to follow suit as AAM’s focus on people-based identifiers is only as powerful as its ability to leverage those identifiers in marketers’ preferred platforms. When it comes to device-based identifiers, although the future of mobile device IDs are uncertain, Adobe can push certain device IDs to DSPs for in-app targeting.

Where will the DMP provide value going forward?

Going forward, the main currency in digital will be customer identifiers (such as email addresses and mobile numbers). A DMP’s value will be determined by its ability to link these IDs to the outside ecosystem.

  • Efficient first party data collection and distribution. DMP’s with a range of integrations can facilitate a one-stop solution for collecting first party data and distributing it to different platforms. Server to server integrations in Adobe allow for both real time and batch uploads of data. To complete these data uploads manually with the same frequency would be a resource intensive process.

  • A security layer. While customer data uploads can be done manually with  most buying platforms, these uploads can expose brands to data breaches if files accidentally fall in the wrong hands. AAM can continue to help brands mitigate this risk by leveraging existing integrations with platforms and providing a security layer though varying degrees of read/write access to team members and agency counterparts, for example. 

  • Profile device graphs. AAM will continue to create profile device graphs which stitch a mobile device ID and a first-party cookie to the customer ID in Audience Manager. This will allow the option to target and collect data on a customer across different devices.

  • Data Privacy Manager. Advertisers can manage consent by placing restrictions on the activation of data based on the rules which were set at the data collection stage.  This enables multiple teams and users to have access to the data whilst making sure it is activated only where it should be.

What can brands do to prepare for a DMP without third-party cookies?

Increase the Collection of Offline Customer Data

If emails, mobile numbers and addresses become the currency of DMPs, then it will be increasingly important to enable the collection of this data to continue to target customers online. Providing users with a value exchange  for sharing their data will become instrumental in facilitating this. 

Improve Customer Data Permissions

Increase the permissions available on customer data. People based identifiers will be the main identifiers to sync to external platforms. Therefore, it’s important to have clear consent and opt-out mechanisms in place in order to equitably use customer data in marketing.

Request Further Functionality from their DMP partners

To stay competitive AAM needs to increase the number of personal identifiers supported in the platform, including mobile numbers, names and addresses. There also needs to be more people based destinations available.

Test Onboarding Partners

Onboarders such as Liveramp and Zeotap have the potential to enhance customer data profiles and connect them to buying platforms. Match tests with onboarding partners can help establish the  incremental value these partners provide to a brand’s tech stack.

Second-Party Data Partnerships with Publishers and Brands

Partnerships centered on data sharing agreements with publishers or other brands, can expand datasets without requiring a third-party cookie.

Collect Mobile Device IDs from In-App Users.

Despite Apple’s move to make IDFAs opt-in only,  there is currently no timeline to phase out Google’s MAID. While Google is likely to modify mobile identification over the coming years as well, collecting mobile device ids in a privacy compliant manner whilst available can bolster a brand’s targeting capabilities.

Be Patient for More Information.

Remember there is more to come. Adobe states they are “involved in several industry organizations and are preparing to invest additional technical resources into efforts led by key partners like Google Sandbox teams, IAB Tech Lab, and other related parties.”

In closing, without the third-party cookie, Adobe Audience Manager is adapting to fit the new digital identity model by introducing people-based identifiers and people-based destinations. However, in order to remain a viable solution in a post-cookie world, Adobe still needs to invest in important functionality  such as a broader range of people based destinations and support for more customer identifiers. Time will tell if they are able to maintain an edge in the DMP space once third-party cookies disappear for good.