Data collaboration: bringing together actionable insights

The digital marketing industry’s spring conference season just ended as many of the big tech companies hosted their annual customer/prospect/all-other-stakeholder events from mid-March to mid-June. Though AI was the topic du jour, data collaboration was the subject of several major announcements – continuing its prominence as a big theme in marketing and advertising.

Databricks unveiled its data clean room offering as an extension to its widely used Lakehouse. And Adobe announced new data collaboration capabilities in its real-time CDP. These breakthroughs continued the general industry shift toward more open data standards and integrations – a trend including Salesforce’s and Snowflake’s respective data clouds.

When it comes to data infrastructure, the vast majority of brands are in grow vs. maintain mode. A decisive 93% say they are increasing their investment in data infrastructure and technology. In the digital marketing industry, this has been most obvious with the trend of data clean rooms, first building steam in 2020 and gaining formal recognition with the Interactive Advertising Bureau (IAB) announcing data clean room standards in 2023.

But despite lots of activity, investment, and even some mergers and acquisitions, this segment is still very dynamic and in its early days. Today it is defining itself as a category of data collaboration platforms (DCPs) that facilitate privacy-compliant, permission-based data sharing for marketing and advertising use cases.

Winterberry recently published an Adobe-sponsored report entitled The Current State and Future Outlook for Data Collaboration Platforms. So I will take a closer look at this emerging technology and explore some key trends.

Data collaboration isn’t a new concept in digital advertising. Data clean rooms that facilitate secure data sharing have been a key area of both discussion and investment for the last few years. This had been largely in response to pressures introduced by the large media platforms to reduce the granularity of data for measurement and targeting, necessitating new approaches and technology for advertisers and agencies.

As the Winterberry report explains, “DCPs have emerged as the primary data collaboration solution for brands, media owners, and agencies across the advertising and marketing ecosystem.” It reveals brand adoption of DCPs has increased from 23% in 2020 to 47% in 2024.

So, what’s behind this speedy growth?

For the last 20 years, the ad server has acted as the technology that enables any brand and any publisher to work together – an easy one-to-many platform. With the demise of the third-party cookie and increasingly stringent privacy regulations, we’re entering a new many-to-many world where every brand has to create specific relationships with every publisher. The DCP provides an environment for this new form of collaboration – a place to easily and compliantly bring together a brand’s customer data with a publisher’s audience data.

Why haven’t we arrived at a standard? Because we are still only a few years into the cycle, the future we are solving for remains vague, and the existing solutions are seen as overly complex, expensive, and lacking the ability to scale. The industry continues to evolve to address these challenges.

As part of this new world, we’re seeing a convergence of adtech and martech. Because all advertising and marketing engagements will now need to be based on an identifiable list, in an identifiable customer set, brands understandably want to use the same data, audiences, and segmentation to reach customers in both paid channels (adtech) and owned channels (martech). They also want to use consistent data and technology to measure campaign performance across all channels.

As we enter this omnichannel, omnibrands era, DCPs provide an environment for marketers, advertisers, and agencies to collaborate with publishers and execute data-driven media campaigns.

The Winterberry report identifies a variety of potential models for the technology. These include standalone DCPs (start-ups), DCPs with embedded identity (identity vendors), a data warehouse with embedded DCPs (data cloud platforms), a walled garden with embedded data collaboration (media properties), and customer data platforms (CDPs) with a DCP (martech vendors).

I expect that all DCPs will require tightly integrated identity, regardless of the model. The success of DCPs will hinge on the extent of the overlap between the list of customers the brand is trying to reach and the audience of the publisher they are collaborating with. Integrated identity capabilities deliver the best possible match rates between these two data sets and will maximize the performance of the solution.

DCPs will need to be relatively simple for marketers and advertisers to operate to gain wide-scale adoption. But there are still barriers that get in the way of that.

The report cites the three biggest challenges as corporate and legal compliance, regulatory adherence, and data governance and ownership. The key to getting past these barriers will be making use of existing platforms, standards, and partnerships that are already in place for other technologies, and adapting these to DCPs, rather than starting from scratch. The industry is moving toward this with the recent announcements mentioned earlier.

In the meantime, marketers should expect to fill that gap with people and services. IAB’s data says data clean rooms are supported by teams that average eight to 12 people. As with other types of marketing technology, marketers should plan for DCPs to need human expertise to implement and operate them for the next several years while the complexities of the evolving ecosystem are worked through. Partnering with a trusted third party in the agency or systems integration space will be a must to access these scarce skills.