Google engineers have developed a technology -- a cryptographic protocol -- that allows agencies and brands to work together, sharing sensitive data to gain insights that strengthen the campaign without revealing information about individual customers being targeted.
It gives the brand marketer and the agency rep insights into the campaign while securing the privacy of the targeted consumers. The protocol aims to increase trust between the brand and the agency as well as ensuring the integrity of the data.
The protocol, Private Join and Compute, is an open-source tool that makes it easier to join numeric columns from different sets of data used to calculate the sum, the count or the average. The data in the sum, count or average remains encrypted and unreadable during the entire process.
The results from the data, not the raw numbers, is the only information the agency or brand receives. Neither party ever reveals their raw data.
Google explains how Private Join and Compute combines two fundamental cryptographic techniques to protect individual data. Each party applies a private encryption that keeps its data secure, remaining unreadable to others.
Then the agency can share the encrypted data with the brand. The agency then applies another layer of encryption, so when the data goes back to the brand, it receives an additional layer of encryption.
Brand marketers no longer need to feel that they are giving away customer secrets when it comes to sharing data with agency partners. When agencies and brands work with sensitive data to execute successful advertising campaigns, it's important for both to gain aggregated insights without agency reps or brand marketers feeling anxious.
The only data that is revealed to both the advertiser and the agency is the total number of people who have made a purchase, the total amount of the purchase, and total revenue from purchases, for example.
From this data, both companies can decide whether to expand or scrape the campaign.
Google also believes this protocol can help advance valuable research in many fields that require organizations to work together without revealing anything about individuals represented in the data. Areas like public policy, diversity and inclusion, healthcare and car safety.