Understanding PAIR: a first-party data solution
As the end of third-party cookies approaches, our industry has an opportunity to implement an improved privacy-first approach, and a range of ID-based and non-ID-based solutions are already emerging. PAIR, or Publisher Advertiser Identity Reconciliation, is one of those relatively nascent solutions emerging within the addressability landscape.
PAIR was first adopted by Google’s DV360 DSP and leverages clean rooms to match first-party data between media owners and marketers. Recently, Google announced that it is working with the IAB Tech Lab to open source the protocol to make it more widely available to the ad industry.
PAIR enables personalised, privacy-centric advertising across all channels—from web to streaming TV—where a deterministic identifier is available from the media owner.
It’s aimed at reaching known audiences, or those for which first-party data is available, and provides marketers with a way to engage high-intent audiences, like past purchasers. That means consumers would only see ads from brands they already have a relationship with or have shared information with.
Since PAIR is a deterministic, authenticated ID solution based on first-party data from media owners and marketers, its application will only increase in this post-cookie world.
How does the PAIR protocol work?
The design quite literally allows a media owner and marketer to pair audiences together. Imagine a list of authenticated subscribers that a media owner has, and a list of customers that a marketer has. Those two audience lists can be encrypted and paired together using a clean room.
For example, say I’m logged into a streaming platform while watching a great new movie. As a subscriber, my account will contain data like my email address or phone number, because I agreed to share it with the platform. Let’s also say I’m a repeat customer of my favorite shoe brand, and so I’m on that brand’s email list.
With PAIR, the shoe brand can serve me an ad for its new sneakers while I’m watching the movie on this streaming platform without releasing any of my personally identifiable information (PII) and without relying on cookies or device IDs.
PAIR doesn’t use any of these first-party identifiers or hashed IDs to identify me as, for example, ID 123. Instead, you have a big data web—as many as hundreds, thousands, or millions of customers—that says this paired ID matches the target audience for this campaign.
Let’s look at the workflow in more detail:
- A media owner first establishes a relationship with a marketer they want to pair with.
- To connect their audience data, media owners and marketers can use a clean room. A clean room is a third-party platform that can take in raw PII and output secure encrypted data—in this case to the media owner, marketer, and DSP. Clean rooms ensure any data shared stays secure, and only encrypted data gets shared with the DSP for reconciliation.
- The clean room operator would create three different encryption keys: a marketer key, a publisher key, and a shared key between the two parties. These keys are unique for every marketer and media owner relationship and ensure there is no way for any one party to reverse-encrypt or identify users. And since each party uniquely encrypts the data for their specific relationship, that data is not useable outside the context of this pairing.
- After encryption and reconciliation, the marketer and media owner both receive the percentage match rate of their respective first-party audience lists. No user-level matches or PII are sent back or shared with the SSP or DSP.
- Media owners pass their encrypted list to their SSP and can include it in bid requests to the DSP.
- Simultaneously, marketers pass their encrypted list to their DSP so it can match the uploaded first-party marketer data to a media owner’s first-party data in bid requests.
- When an impression opportunity arises, the SSP runs an ad auction, and the winning ad is displayed to the consumer.
What is PAIR’s future potential?
PAIR enables marketers to get the value of deterministic matching and personalisation in a privacy-safe manner. It works across devices, which creates a large potential for addressability in streaming TV. We expect traction will grow quickly here.
It’s unknown if new, more restrictive privacy regulation or platform changes in the long term will eliminate the ability to use IDs based on data pooling, like other authenticated-ID-based solutions available now. With PAIR, marketers and media owners maintain ownership of their data, and the encrypted data is not usable outside of that relationship. Data pooling is prohibited.
Therefore, you could argue that PAIR is potentially future-proofed for more restrictive privacy regulations that may come about in different regions. Plus, scale will only increase as the IAB Tech Lab evolves PAIR into an open industry standard.
Overall, PAIR aims to preserve all the value of deterministic and personalised advertising through a privacy-friendly option.
Learn more about choosing the right addressability solutions for your business.
