Priya Anand

Priya joined Mobidea in March 2019 as a junior analyst on the affiliate-side dashboard team. The push-format business was scaling fast — Mobidea was buying push inventory from a dozen networks and reselling segmented audiences to performance advertisers — and the data was a mess. Conversion windows were inconsistent across networks. Publisher IDs were aggregated differently per source. The attribution stack had been built incrementally over four years and no one person understood it. Priya spent her first eighteen months rebuilding the conversion-latency model from scratch, then got promoted twice in nine months when the rebuilt model surfaced $1.4M of misattributed revenue across one quarter.

What she learned over five years isn't visible in Mobidea's published case studies. It's in the gap between the panel CTR (which a network can show in real-time) and the seven-day-stabilised CR (which most advertisers never wait to see). It's in the bell-curve of publisher quality where the top 8% of sub-sources deliver 60% of human conversions and the bottom 20% are bots that have learned to defeat every fraud filter on the market. It's in the statistical-power math behind "we ran an A/B test" — a phrase she'd come to read as "we got a number we liked."

She left Mobidea in October 2024 after a conversation about a quarterly report she'd been asked to soften. The report said a $200K push campaign had failed because the audience-fatigue curve had collapsed in week three. The asked-for edit was to remove the fatigue chart and attribute the failure to creative quality. Priya pushed the chart back in, the report shipped unedited, and she resigned two weeks later. The writing started as cleaned-up internal memos, became a private Notion she shared with three former colleagues, and is now this site. She still consults. She still runs tests for friends' campaigns when the data question is interesting enough.