Bye-bye standalone clean rooms, hello data collaboration platforms

Think back to before AI dominated adtech headlines, and clean rooms were all the hype. They promised to solve issues such as siloed data, eroding identifiable signals, and increasingly strict privacy regulations by facilitating the secure and private sharing of data between partners.

However, engineer-intensive data transformation requirements, lack of scaled universal identifiers, time-consuming importing and exporting processes, and high costs prevented clean rooms from being the be-all and end-all of collaborative data matching.

Fast forward to today and clean rooms have a new lease of life, not as standalone products, but as a feature built into a diverse and growing array of data collaboration platforms. Do clean rooms have a future on their own, or is it their fate to become fully absorbed?

Clean rooms surface overlapping data, but what can you do with it?

Let’s start by looking at clean rooms’ value. Two or more collaboration partners being able to understand the cross-section between their audiences opens opportunities for co-marketing, informing strategies like contextual targeting, refining campaign messaging to best reach target consumers, measurement and attribution, and so on.

However, where standalone clean rooms often fall short is scale. A brand and a publisher could each have fairly large audiences, but once narrowed down to authenticated users, they may only be left with 10,000 hashed emails or mobile IDs. Then, if they performed an overlap analysis and found 10% shared users, that’s just 1,000 people to target.

Simply speaking, there’s not much most marketers can do with 1,000 people. At this point, they may as well hit the streets and start knocking on doors.

Those who have been in the marketing world for a few years might be feeling deja vu for the CDP craze—when the industry was buzzing with the possibilities of collecting and activating first-party data—only for marketers to realise they didn’t have enough data to actually power scaled programmatic and marketing strategies

Like some CDPs, standalone clean rooms fall short in the ‘do something’ piece. After putting in the work to prepare, share and permit its data, what insight can a marketer get from it? And what can they do with that insight? How do they make it actionable? How do they derive value from those actions?

Those 1,000 customers might not be viable on their own for scaled prospecting, but plug them into a machine learning algorithm, perform identity resolution, or see how they intersect with second and third-party data and they could be extrapolated to a wider audience of net new consumers and IDs.

This is why it makes sense for clean rooms or clean-room-like functionality to be built into larger data collaboration platforms. Here, clean rooms are one option among many methods for two or more partners to deliver their marketing and monetisation objectives, with all the pipes to take that data further downstream already built. Clean rooms are far more effective at being a part of the data puzzle, rather than being treated as the whole board.

Now, despite the headline, there is still a place for standalone clean rooms in the market. Truly giant brands and media owners with millions of customer records—plugged into universal identifiers and with robust in-house data platforms of their own—may not need more than a point solution.

While clean rooms might be appropriate for the Disneys and Amazons of the world, the vast, vast majority of marketing happens beyond such juggernauts, and the solutions that cater to this majority are building toward data collaboration with a far broader scope.

Where are clean rooms going? Just look at the market

The strongest evidence that clean rooms are being absorbed into larger data collaboration platforms is in the movements we’ve seen in the market over the past year. Companies providing identity solutions and data marketplaces are either building their own clean room functionality or buying up clean room providers, as seen in LiveRamp’s $200 million acquisition of Habu.

This is a sensible move. Marketers want to be able to interact with the data enrichment identity resolution, scaling, and downstream activation that such platforms provide but don’t want to move their data to do so, which is exactly what query clean rooms like Habu enable (i.e. the non-portability of data).

One place where there hasn’t been any movement is within the walled gardens operated by the likes of Disney and Meta. Their clean rooms are not standalone, but they also don’t reach beyond their own realms, and are designed exclusively to commercialise their own first-party assets. A brand can bring its data in to find matches with the customer data exclusive to these walled gardens, but doing so beyond the gates of the walled garden has proved futile.

Amazon—perhaps due to its sheer scale—has been the exception, straddling the line between a walled garden and a collaborative platform; a hedged garden, if you like.

Activation and personalisation are still ultimately confined to Amazon’s programmatic ecosystem (with promises to build direct pipes with third-parties in the future), but its push for interoperability with identity solutions and the launch of modelling capabilities indicates a more open attitude that may be built upon further.

Have data collaboration platforms got everything figured out? Given the complexities of data sharing and activation, it would be naïve to say yes. But what they have is a north star goal for broad connectivity and interoperability across platforms and solutions, and it’s clear that there is widespread movement towards it.

Lana Warner is senior director of Partnerships & Strategic Solutions at Lotame.