Turning nine legacy audiences into a system of roughly 40 targetable segments

• I joined the audience development role after the position had been vacant for some time. The team had nine legacy audience segments built in Universal Analytics, which was being retired in favor of GA4.
• I already recognized the nine audiences because, during my first stint at GMA New Media from 2019 to 2021, I had helped write the copy describing them in a monthly bulletin circulated internally across GMA.
• The transition was not simply technical. Publishers were also preparing for reduced reliance on third-party cookies, while Sales, product, and editorial teams needed more useful ways to understand and reach GMA's audiences.
• The monthly report required the latest figures to be copied manually into a spreadsheet before Ad Operations could begin producing the bulletin. The process was manageable but repetitive, tedious, and exposed the workflow to the operational risk that accompanies manual transcription.

I owned the migration and expansion of the audience library, from designing and validating segment logic to creating the audiences in GA4 and making them available through Google Ad Manager. I also automated the monthly reporting process and coordinated the handoff to Ad Operations for the internal audience bulletin. Sales helped surface client demand, while Kasatria reviewed the proposed audience logic before implementation.

• Reconstructed the logic behind each of the nine legacy audiences and recreated them in GA4 before beginning the expansion of the library.
• Translated broad commercial requests into audience definitions that could be implemented and explained clearly. Interest in reaching micro, small, and medium enterprise owners became Emerging Entrepreneurs, while requests involving recent graduates developed into Career Builders.
• Developed the logic for each new audience and reviewed it with Kasatria, a Google Marketing Platform Partner that GMA consulted on Analytics and Ad Manager questions.
• Created each validated audience in GA4 and allowed it time to reach a useful working size. In practice, we usually gave a new segment at least two weeks before using its size for planning or reporting.
• Developed publisher-defined age-range and gender audience models from onsite behavioral patterns, then validated the group-level results against aggregated age and gender reporting already available in Google Analytics.
• Used the audience models to address a limitation in the linked GA4 and Ad Manager workflow, where audiences containing Google's native demographic or interest dimensions could not be exported for targeting.
• Automated the monthly spreadsheet updates, reducing repetitive handling before the report moved to Ad Operations for production of the internal bulletin.

• The original nine audiences grew into a library of roughly 40 targetable segments.
• Sales gained a broader menu of audiences that could be matched with different client briefs, creating more ways to package and describe GMA's existing advertising inventory.
• Internal teams used the segments in product development, while I also used their definitions and performance when preparing reports for editorial teams.
• Publisher-defined age-range and gender audience models became some of the most frequently requested options. They were presented as models rather than first-party declared demographic data.
• Automating the monthly report made the recurring workflow faster to maintain and reduced the amount of repetitive manual handling.