Mindshare’s NA Chief Data Strategy & Analytics Officer, Brian DeCicco warns that marketers are rushing into the novelty of clean rooms ahead of identifying what exactly they’re searching for and how this technology uniquely addresses those use cases. Instead, they should start by looking at why data collaboration is essential to their strategy.
In 1904, the Welte-Mignon “reproducing piano” set the music industry ablaze. The riff on the player piano allowed famous pianists to record their performances on “piano rolls” which could be sold and played on Welte-Mignon pianos found in households across the world. This remarkable invention would usher in a golden age of recorded music that introduced pianists to a massive, engaged audience.
You’ve probably never heard of the Welte-Mignon. Despite all the hype, not a lot of people had a 500-pound reproducing piano that could play those rolls in their homes. Perhaps this wasn’t the straightest line to capitalize on the growing consumer demand for recorded live music performances in the age of 78s and the Victor Victrola – the emerging convenient, affordable format of choice at the time.
I share this story because it illustrates a common tale in technology adoption cycles: when hype precedes clarity on the need the technology aims to fulfill. It’s a tale we see often in marketing technology today. That technology trend right now is the clean room.
Spoiler alert: The clean room is not the Welte-Mignon. Clean rooms have already demonstrated value on a variety of use cases and their future looks quite promising. But marketers are rushing into the novelty of clean rooms ahead of identifying what exactly they’re solving for and how this technology uniquely addresses those use cases.
The hype is intense right now, but the fact is clean rooms are just a means to an end. Data collaboration is that end.
Even in a world of rapidly eroding digital signals, marketers and brands will still need to take advantage of the benefits of precision in targeting. For many marketers, this has meant a rush towards consented first-party data records – only to find that's easier said than done. It’s now impossible to know everyone everywhere all the time. So, to achieve personalization through precision while respecting consumer privacy, the future can not only be about data collection. It must also be about data collaboration.
Data collaboration is when at least two owners of consented, first-party data, say a brand and a media publisher, intersect their customer data in a privacy-compliant way. Typically, this means without either party transmitting any personally identifiable information (PII). This can be done with the goal of creating greater insight into a brand’s customers to inform activation or measuring attribution and incrementality at a granular level.
How do you know if your business requires a data strategy that balances data collection and collaboration? Here are four questions to ask yourself in making that determination:
- Do I have a fair value exchange? If consumers view their relationship with you as predominantly transactional (as is the case in many categories), they might not see the value in sharing their data with you. Be honest about your business and the expectations that consumers have of brands in your category.
- Is my view of consumer behavior obstructed and retrospective? For many brands, the consumer of tomorrow, raised in the platform-age, is not reflected in your data today – and that blind spot will only increase. You likely will never have a full view into their wallet or why they make the choices they do. Attempting to collect all that data directly within your walls is simply not possible.
- Are there limitations in the way I will be able to activate my data? Between changing privacy laws and big tech policies, you may be increasingly limited in how you can use PII-based customer data in the future. Ethical considerations may further limit the use cases that regulatory compliance hasn't
- Am I dependent on publishers and platforms to fill in gaps in consumer insights? Activating a personalized marketing strategy hinges on your essential publishers and platforms having good first-party data as well. The challenges you face are the same that your partners do. Increasingly, there will be a requirement to collaborate with activation partners, for example retail media platforms, and use any future path of data intersection to best engage your target audiences. In these collaborations, there may be other match keys (beyond PII-based customer data) that are more readily available, don’t require as big a bet, and grant you more flexibility as the industry changes in the future.
If you answered “yes” to all these questions, you have your answer. Your future data strategy must be designed for data collaboration. And if you answered “no,” you are a unique snowflake positioned for marketing domination in your category. (Doesn’t sound like you? Repeat the above exercise.)
With data collaboration established as a vital ingredient of any data strategy going forward, how do you facilitate collaboration? Well, that brings us back to clean rooms. A clean room is simply a technology that is designed to facilitate that intersection of data from two or more companies. The “or more” is important as the clean room v1.0 that we first got to know was the media partner clean room. However, for clean room usage to scale and drive better precision through data collaboration (especially in addressing eroding digital signals like third-party cookies), then it’s essential that we see the rise and adoption of neutral, multi-tenant clean-room solutions.
Until then, clean room exploration must be considered alongside other probabilistic modeling and matching solutions. As you think about solving across the consumer journey, data-science derived audience prospecting strategies, leveraging federated data models and composable infrastructure can create the scale and interoperability across inventory sources that clean rooms simply cannot match alone today.
In short, in a future where data collaboration mechanisms exist, paired with the advancements in AI and machine-learning to act more confidently on predictions, privacy and personalization may no longer be at odds for advertisers. The lesson to learn from the Welte-Mignon, and so many other stories like it: don’t jump to the hyped technology without doing your homework. Lean in and design a flexible, modern data strategy that appreciates this future, welcomes the diversity of signals coming from across your partner ecosystem, and exploits all the new technologies coming of age now to best facilitate that intersection of yours and theirs to activate on maximum intelligence.