{"id":13230,"date":"2022-02-14T22:23:04","date_gmt":"2022-02-14T22:23:04","guid":{"rendered":"https:\/\/mailinvest.blog\/index.php\/2022\/02\/14\/mozilla-and-meta-are-working-on-privacy-preserving-attribution\/"},"modified":"2022-02-14T22:23:04","modified_gmt":"2022-02-14T22:23:04","slug":"mozilla-and-meta-are-working-on-privacy-preserving-attribution","status":"publish","type":"post","link":"https:\/\/mailinvest.blog\/index.php\/2022\/02\/14\/mozilla-and-meta-are-working-on-privacy-preserving-attribution\/","title":{"rendered":"Mozilla and Meta are working on privacy-preserving attribution"},"content":{"rendered":"

<\/a>\n<\/p>\n

\n

Mozilla, developers of the open-source Firefox web browser and Meta, formerly known as Facebook, have revealed that they are working together on an Interoperable Privacy Attribution solution (IPA). The aim is to allow advertisers insight into campaign performance without breaching individual user privacy. They have jointly proposed to be the in-development solution to the Private Advertising Technology Community Group, part of the World Wide Web Consortium.<\/p>\n

In broad terms, the solution seems simple. It would use multi-party computation to allow analysis of user behavior without allowing websites, browsers or advertisers to learn about the behaviour of individual users (a version of what is often called a \u201cclean room\u201d). Data would be aggregated to avoid traceability to single users.<\/p>\n

\u201cIPA is designed to provide a lot of flexibility for advertising businesses in terms of how they use the system. Cross-device and cross-browser attribution options in IPA enable new and more robust attribution capabilities, while maintaining privacy,\u201d said a Mozilla blog<\/a>, which also contains links to detailed technical specifications.<\/p>\n

Read next: More about multi-party computation and \u201cclean rooms\u201d<\/a><\/em><\/strong><\/p>\n

Why we care, <\/strong>It remains a work-in-progress and even a quick dive into the specifications shows that questions remain unanswered. For one thing, the solution requires the addition of random \u201cnoise\u201d to the aggregate statistics, something unlikely to please advertisers. Of course, the aggregate data theme was part of Google\u2019s FLoC and is now part of Google\u2019s Topics API too. But it is at least refreshing to read a proposal for measuring campaign performance and attribution however the ads are delivered rather than another proposal to deliver ads through some new means of tracking users.<\/p>\n

\n
\n
\n
\n

Get the daily newsletter digital marketers rely on.<\/p>\n<\/div>\n<\/div>\n<\/div>\n

\n\n
\n

About The Author<\/h3>\n
\n
\n
\n
\n