Data Clean Rooms Explained: What You Need to Know About Privacy-First Collaboration

If you ask any advertiser about the most disruptive factor in recent years, they’ll probably hesitate between two contenders: privacy and AI. While AI is poised to have a transformative impact far beyond advertising in the future, one thing is certain: No organization today can address use cases involving consumer data without prioritizing privacy.

Before we dive into the world of data clean rooms, let’s take a quick trip back in time to set the stage.

Governments take action for consumer privacy

The rise of the internet allowed organizations to start collecting consumer data at unprecedented scale — more efficiently than ever before, but often with little regard for how that data was collected in the first place.

It wasn’t until 2016 that a government took decisive action to address growing consumer privacy concerns. Europe led the charge by introducing the General Data Protection Regulation (GDPR), its first comprehensive privacy law.

In the United States, California followed suit in 2018 with the California Consumer Privacy Act (CCPA), strengthening it in 2020 to further protect consumers — and impose stricter rules on businesses.

The snowball effect is real: More states, including Colorado, Connecticut, Florida, Montana, Oregon and Utah, have recently implemented their own privacy regulations, with others poised to join the trend.

Globally, the movement is unstoppable. Data protection and privacy laws are now present in 71% of countries, according to the United Nations Conference on Trade and Development, and new legislation continues to roll out and evolve.

Tech companies respond with privacy-protection initiatives

In addition to government regulations, large tech companies have been rolling out their own privacy-focused initiatives in recent years. Among the most controversial — and widely discussed — efforts is the ongoing shift around third-party cookies, an element that has been the backbone of the advertising industry since the 1990s.

After years of announcements and delays, Google has opted to maintain third-party cookies in Chrome for now — no immediate cookie-apocalypse — but with tighter restrictions on access and usage expected in the future, including controls such as user consent.

According to eMarketer, up to 87% of web traffic could soon be freed from third-party cookies once Google’s consent-based solution rolls out and Microsoft eliminates third-party cookies in its Edge browser. Meanwhile, major browsers such as Apple Safari and Mozilla Firefox have already made third-party cookies inaccessible.

But it’s not just about cookies. Both Google and Apple continue to roll out consumer privacy initiatives. For instance, Apple’s App Tracking Transparency (ATT), introduced in 2021, requires apps to obtain explicit user consent before collecting device identifiers for advertising purposes.

Ultimately, these changes are transforming the entire ecosystem of advertising "currencies."

Data clean rooms: Where it all began

Around the time the first consumer privacy regulations were being introduced, another major shift was shaking up the advertising industry: Google announced it would stop sending log-level data back to advertisers.

Here’s the issue: These logs are essential for advertisers to analyze campaign performance. Without access to this data, organizations are left flying blind, unable to optimize their strategies or budgets effectively.

To address this challenge, Google introduced Ads Data Hub — a solution designed to allow advertisers to continue running analytics and reporting on their campaigns. The catch? Advertisers could no longer directly see or extract the log-level data. Instead, the platform provided a privacy-preserving environment for data analysis.

The term “next generation insights and reporting” was used at the time to describe this new approach — which would eventually become data clean room technology.

What is a data clean room?

It’s hard to say exactly what causes a specific technology category to take off, but one sign it’s gaining traction is when it gets its own widely recognized acronym. Enter data clean rooms, or DCRs.

The concept behind data clean rooms stems from the same challenge Google addressed with Ads Data Hub: enabling data collaboration between two parties without exposing the underlying data.

First-party data is among the most valuable assets an organization owns, hence the sensitivity around making it accessible. Still, there are critical scenarios where analyzing data sets owned by different parties is essential.

Data clean rooms enable a secure, controlled environment that allows multiple organizations — or even business units within a single organization — to collaborate on sensitive or regulated data without compromising privacy.

A key component of this configured protection is the use of privacy-enhancing technologies (PETs). These include methods such as differential privacy, aggregation and projection policies, and synthetic data generation.

Who are data clean rooms for, and what are the common use cases?

As discussed earlier, data clean rooms initially gained traction in the advertising industry, particularly for measuring ad campaign performance without requiring the publisher to give direct access to granular data.

Over time, the scope of collaboration expanded, involving various stakeholders with different roles in advertising initiatives:

• Brands: They focus on acquiring new customers and driving revenue through paid advertising.

• Publishers and media networks: They aim to monetize their data and ad inventory.

• Agencies: They support advertisers and publishers with campaign execution and strategy.

• Tech vendors and data providers: They sell data, identity solutions and services such as integrations within the ad ecosystem.

Through the partnerships formed among these stakeholders, typical advertising collaboration use cases now include:

• Data enrichment and identity: Partners can enhance first-party data and increase addressability.

• Strategic planning: Advertisers can decide where to spend ad budgets, and identify the most relevant audiences.

• Campaign activation: Consumers can be reached through direct or partner-supported channels.

• Measurement and optimization: Organizations can understand channel impact on conversions and refine their media spend.

For example, Booking.com partnered with Snap using Snowflake Data Clean Rooms to measure campaign performance more effectively. This collaboration increased confidence in their results from under 20% to an impressive 99%.

However, the potential of data clean rooms goes far beyond advertising and can reach other industries, as in the following examples:

• Healthcare: The industry is accelerating drug research and development by enabling secure data analysis between labs and healthcare facilities without exposing sensitive information.

• Financial services: Organizations are accelerating fraud detection and improving credit scoring models while safeguarding customer data.

Advertising is just the starting point for proving the value of this technology. As industries continue to recognize the benefits of secure, privacy-preserving data collaboration, we can expect broader industry adoption in the coming years.

How do data clean rooms compare to other technologies?

A common misconception is that data clean rooms are the same as data-sharing technologies. Secure data-sharing solutions allow data owners to share their data sets with specific controls in place. The objective of data sharing is to provide access to the granular underlying data — a direct contrast to the purpose of data clean rooms, which are designed to prevent such access while enabling data analysis.

Another technology category often compared to data clean rooms is the customer data platform (CDP). While both rely on first-party data to deliver value, the similarities end there. CDPs focus on making a brand’s first-party data accessible for marketers and advertisers to orchestrate personalized customer experiences. However, CDPs lack the tools and measures needed to facilitate secure collaboration with external data owners.

How does a data clean room work?

Once a collaboration agreement is established between two or more parties, a data owner — referred to as the “data provider” — sets up a clean room environment. The data provider determines what data is accessible within the clean room and specifies the activities permitted on those data sets, such as audience overlap analysis or lookalike modeling.

Each party involved in the collaboration retains full control over their data sets at all times. They can decide to grant or revoke data access as needed, ensuring their data remains governed and under their ownership.

After the data sets are made accessible within the clean room, a matching process between them is required. Some data clean room technologies enforce the use of a specific identifier as the matching key, while others are agnostic, allowing the collaborators to agree on the matching criteria of their choice. Successful collaboration relies on an exact match of values for a designated data point (for example, a specific field) between the data sets.

Collaboration in a clean room often concludes once the desired insights are obtained. However, in some scenarios, the clean room can enable activation of the resulting data set to a permitted channel.

Data clean rooms alone aren’t enough

While data clean rooms facilitate secure data collaboration, it’s essential to remember that privacy isn’t achieved by deploying a single piece of technology. True privacy requires a comprehensive strategy that starts with the consumer.

If an organization wants to collaborate on data with other parties, obtaining consumer consent is nonnegotiable. To secure consent, organizations must prioritize transparency and ensure there’s a clear value exchange. Consumers today are increasingly aware of the value of their data and are far less likely to share it without understanding what they’re getting in return.

Even with the advanced privacy and security technologies provided by data clean rooms, organizations need to establish robust data governance practices. These practices should govern every activity involving data access and usage to ensure compliance and maintain trust.

Snowflake Data Clean Rooms for data collaboration

The Snowflake AI Data Cloud has been adopted by thousands of organizations to securely store and process their first-party data, including sensitive and regulated data sets. With its trusted infrastructure and unified governance model, Snowflake offers comprehensive compliance, security and privacy controls that are uniformly enforced.

Snowflake Data Clean Rooms is a Snowflake Native App deployed on top of the AI Data Cloud, providing a privacy-preserving and trusted environment for data collaboration. Designed to support both technical and business teams, it can simplify secure collaboration without compromising data privacy.

Organizations choose Snowflake Data Clean Rooms not only because of its seamless integration within the Snowflake ecosystem but also for its unique advantages:

• Neutrality: Avoid conflicts of interest with an agnostic and neutral solution. Unlike alternative solutions, Snowflake doesn’t sell data, identity solutions or media, ensuring there are no conflicts of interest — making it a truly neutral option.

• Trustworthiness: Accelerate collaboration by leveraging the same technology trusted by leading publishers and industry experts for their clean room initiatives.

• Cross-region and cross-cloud interoperability: Collaborate with partners using the cloud infrastructure of your choice — whether AWS, Microsoft Azure or Google Cloud.

Attend our virtual event, Accelerate Media and Entertainment, to learn more about the future of data collaboration.