Data clean rooms: The solution for brands' identity hurdles?

The online identity landscape is rapidly changing, confronting advertisers with challenges for activation or measurement.

In the context of third-party cookie depreciation and privacy regulations, interoperable IDs and data clean rooms (DCRs) promise to reconnect data that advertisers can use for targeting, personalisation and measurement.

Fragmentation has plagued the adtech industry for a long time, and digital identity interoperability at scale is still a long way off (1% -5% in many markets) and may never be attainable. Data clean rooms offer a viable method for brands to keep up with their audience’s behaviour and create valuable insights, but they will only be as helpful as the IDs that power them.

Gartner predicts that by 2023, 80% of advertisers spending more than $1 billion annually on media will use data clean rooms. A study by the IAB found two-thirds (64%) of companies leveraging privacy preserving technology are using DCRs.

This increases to 85% when including those that are considering the technology. Most companies use the technology for vital tasks including privacy protection and controls, match rates, and interoperability.

(L-R) Veron Dai, Jonathan Joseph, Todd Rose

Veron Dai, the regional data and consultancy director for Asia Pacific at Assembly Global, explains when used correctly, data clean rooms are a powerful tool that enables brands to develop more integrated marketing strategies. Brands using data clean rooms have access to a holistic view of their audience across multiple ad channels.

“From targeting the right audience with more relevant messages to avoiding duplicated ad exposure to the same user across multiple channels, data clean rooms not only improve campaign performance but it also grants brands the ability to design a consumer-centric journey that elevates consumers’ experience and affinity for the brand,” Dai explains to Campaign Asia-Pacific.

For example, when a CPG brand is developing a campaign for a new energy drink sold at Whole Foods, they will purchase targeted advertising from the top sports publications and receive back advertising data.

As Amazon, the owner of Whole Foods, has specific data on who purchased the drink, data clean rooms will allow Amazon to marry the ads that pushed consumers to buy the products in question so that this brand can optimise and better target future campaigns.

Jonathan Joseph, head of solutions at Ketch, explains clean rooms are secure environments for processing joint data from two parties for a specific purpose. The Amazon example shows how a consumer brand’s ads led to a purchase.

“By using clean rooms, marketers have access to a new layer of analysis that wasn’t available before. For example, which publishers deliver the most actual sales? How many ad exposers led to the highest number of conversions?” Joseph tells Campaign Asia-Pacific.

“And what is the ideal number of ads to prompt the desired outcome, segmented by geography, product or any other attribute? Clean rooms answer these questions based on actual data, not proxies. While frequency control across all environments is not possible at the user level, the granularity clean rooms provide advertisers will enable marketers to preserve the consumer experience within every environment.”

How do clean rooms work? 1. Imagine a conveyor belt. Advertisers place their first-party data packages, which may consist of user-level, transactional, and historical data on this belt. Similarly, another advertiser or publisher positions their own package of first-party data on the opposite side of the belt. 2. Once placed on the belt, the packages are transported to a metal box, which serves as the data clean room. Inside this enclosure, the data from both parties undergoes matching and cleansing procedures, including audience matching, and the implementation of various privacy measures such as encryption, hashing, pseudonymization, access restriction, and noise injection. 3. Subsequently, advertisers may display ads to individuals within their desired target audience, and obtain corresponding reports that can be examined and leveraged for other advertising-related endeavors. Image credit: Clearcode

Todd Rose, senior vice president of global business development and general manager for identity and addressability at InMobi, cautions that data clean rooms are part of the comprehensive suite of tools that brands in a privacy-first world can use. However, data clean rooms in and of themselves cannot and do not resolve identity.

He explains clean rooms allow two or more parties that already have some form of an identifier to match data in a secure environment without disclosing proprietary data directly to one another.

“In other words, clean rooms are a solution for data privacy and security, but not for identity. They are a complementary technology to universal IDs, Privacy Sandbox and contextual targeting, not a substitute,” Rose tells Campaign Asia-Pacific.

How brands are using data clean rooms

Brands previously thought of data clean rooms as tools to assess audience overlap. For instance, an airline and a hotel would establish a data clean room to determine the extent of their shared audience and decide if launching marketing programs together was a good idea.

Today, however, brands use clean rooms to plan campaigns, activate and suppress audiences, and measure outcomes.

For example, Disney uses data clean room provider Habu’s technology to not only provide measurement back to advertisers but also to activate audiences directly from the clean room bypassing a traditional onboarder and activating against their first-party graph.

Aanchal Kumar, director of growth and partnerships for APAC at Jivox, says the adtech firm has seen many brands have eagerly embraced data clean rooms so that they can do advanced analytics, attribution and audience curation.

When it comes to analytics, she explains brands can automate data flow to their data warehouse and enable scalable ad-hoc manipulation of granular data with data clean rooms.

“Attribution allows them to connect with partner data to match user IDs, amplify insights and understand the customer journey without sharing sensitive data externally. When it comes to curating audiences, brands can use granular insights to create micro-segmented audiences in demand-side platforms or customer data platforms to personalise ad experiences further,” Kumar tells Campaign Asia-Pacific.

“We have worked with brands to use data clean rooms to consolidate all their digital marketing data, which is then tied into data visualisation tools to discover insights. This is a big stride in unearthing the valuable insights that many brands have been for years striving to uncover from the data they have been amassing.”

(L-R) Aanchal Kumar, Michael Bridgeman, Michael Sweeney

Michael Bridgeman, the co-founder of DataCo, shares he has helped brands ranging from large banks, and airlines, to fast-growing consumer tech scale-ups to utilise data clean rooms to increase loyalty, increase the share of wallet and optimise their retail footprint optimisation

Bridgeman explains that by analysing customer interactions with their brands, enterprises can identify opportunities to improve customer experience and increase loyalty by association.

For example, they can identify areas for improvement in their customer service or develop more personalised marketing campaigns that are more likely to engage customers. Increasing loyalty can lead to repeat business, which can be a crucial driver of long-term growth and success.

“Brands can also analyse customer behaviour and preferences data; enterprises can identify opportunities to increase their share of wallet, or the amount of money customers spend with their brands. For example, they can identify cross-selling opportunities or new products or services that customers might be interested in. This can help to drive sales and revenue growth,” he tells Campaign Asia-Pacific.

“When brands share data on store performance and customer demographics, enterprises can optimise their retail footprint and identify opportunities for expansion or contraction. This can help them to make more informed decisions about where to open new stores and to allocate resources more effectively.”

Michael Sweeney, head of marketing at Clearcode, gives an example of a data partnership between different sponsors of a sporting franchise.

The sporting franchise and its sponsors use Aqilliz, a data clean room provider operating in Asia, to match and utilise their first-party data. However, the exciting part of this was that the sports franchise’s sponsors were gathering insights from each other about the sports fans to learn more about their interests and behaviour.

“This example shows that data clean rooms are valuable for running the key programmatic advertising processes, such as ad targeting and measurement, and for consumer insights,” Sweeney tells Campaign Asia-Pacific.

With retail media networks on the rise, retailers and CPG brands are increasingly turning to data clean rooms, says Seow Ping Tan, head of SEA at LiveRamp.

LiveRamp has collaborated on retail media networks with major retailers like Boots on their Boots Media Group, and with Carrefour on Carrefour Links.

"Retail media networks are popular because retailers are looking for ways to generate more value, and own many of the touchpoints with today’s consumers - while FMCG brands may be first-party data poor and looking to reestablish their relationships with consumers," Tan tells Campaign Asia-Pacific.

"By leveraging data clean rooms to collaborate on customer and transaction data, retailers can leverage insights on what customers are buying, how much they are buying, and how often, and share these insights with FMCB brands. Brands achieve better targeting, reach the right audiences, and reduce their advertising waste. Retailers improve yields and deliver better customer experiences."

The case against data clean rooms

While data clean rooms offer a viable means of matching audiences, they are no silver bullet.

One concern with data clean rooms today is the scale and granularity of the data that partners are comfortable delivering to the clean room in the first place, says Gavin Buxton, managing director for Asia at Magnite.

“A buyer might also encounter additional complexities if they need to match data within several clean rooms, each serving a subset of publishers,” Buxton tells Campaign Asia-Pacific.

Niall Hogan, managing director for APAC at Ogury, points out different data clean rooms cannot be interoperable. There is also a natural technical barrier when using them as they don’t have a user interface, requiring these same users to master technical languages like SQL.

“The main issue is that data clean rooms aggregate data which can be, and often are, sensitive and personal. We advocate for a technology that does not rely on personal data on the user side and does not require any specific technical skills on the marketer side,” Hogan explains to Campaign Asia-Pacific.

(L-R) Dane Buchanan, Niall Hogan, Femi Taiwo and Gavin Buxton

Femi Taiwo, head of consultancy for Europe at Assembly Global, warns brands not to use data clean rooms for demand-generating activity and activation.

Taiwo explains that data clean rooms are inherently based on cookies and IDs, with insights allowing for retargeting and audience analysis but are unsuitable for brands looking to generate demand.

“In addition, outputs of the clean rooms can be transformed into tactical workflows, but fundamentally, the clean rooms are about insights and additional analytics. They rarely can turn an insight into activation-friendly actions,” Taiwo tells Campaign Asia-Pacific.

Aside from data clean rooms, there are alternatives such as universal user IDs, Google Privacy Sandbox, and contextual targeting.

Benjamin Rehberg, head of data for APAC at Teads, notes now that many brands have already tested cookieless solutions. So the best way forward is a step-by-step approach that arrives at a portfolio of solutions that work best for the brand.

He says for brands with a focus on ROI and requiring offline shopper data, unique IDs are a crucial ingredient. In addition, context alignment can deliver outstanding concentration and memorisation for brands looking to maximise attention suitably.

(L-R) Vasily Popravko, Benjamin Rehberg and Seow Ping Tan

“There has never been more confusion about privacy regulation than today. Lawmakers worldwide are looking to the EU and the US waiting to see what flies,” Rehberg tells Campaign Asia-Pacific.

“One thing is for sure cookieless targeting is more than replacing third-party cookies. It is about a new value exchange that retains addressability in some places and guarantees privacy in others. It will be sustainable if the new deal of the internet works for consumers, brands, and publishers.”

Vasily Popravko, data director for APAC at Media.Monks point out plenty of other solutions being developed in the martech and adtech space to provide a privacy-safe alternative to the ways of targeting in the past.

“I would not call them alternatives, as most would usually complement each other. But, for example, Google Privacy Sandbox, with its technologies and solutions, is something that a brand should explore alongside data clean rooms,” Popravko explains to Campaign Asia-Pacific.

“The whole topic is complex, as there are multiple players and ecosystems with solutions that require a lot of testing and sometimes vanish before being fully released. Brands are put into a tough situation now. Therefore, planning and prioritising the solutions is important as you can only invest your time and money into a few.”