Launched Self-Service BI for Google Ads Support
TL;DR Transformed Google’s underused internal Ads support data warehouse into an indispensable tool — increased utilization by 4X with 90% lower tail latency and high CSAT.
Challenge: Google’s internal Ads customer-engagement data warehouse — the tooling support teams leaned on to understand advertiser issues — had drifted into disuse: slow, inflexible, and painful to onboard new data. As Lead PM I owned the turnaround: redesign the stack around ease of use, with a developer’s story, so non-engineers could do their own data warehousing.
Approach: I defined the mission as making support data easily accessible and useful, along three pillars — Create (build with no SQL), Organize (onboard new data and stand up a report in minutes), and Share (easy-to-maintain tools others can keep running).
Impact / Key Learnings: A full redesign delivering a self-service, “do-it-yourself” model — 4X the incumbent’s utilization, 90% reduced latency at the 95th percentile, self-service BI scaled to >12K MAU/DAU, and 8 consecutive quarters of positive CSAT. I made something 80% as beautiful as Tableau but 10X more powerful.
