Why activation rate is a poor proxy for feature adoption
A healthy first week can hide a surface nobody returned to. Here is the split we insist on in class.
Journal
Field notes on uptake, written for people who already have a warehouse.
These pieces come out of studio critiques and alumni disputes. They are not tool tutorials.
Instrumenting first-use without drowning in noise
Three events will do more work than a thirty-name taxonomy nobody trusts.
Reading Thailand-market cohort curves after a regional launch
Store day, language-pack day, and payment-rail day are three clocks, not one.
Exposure, trial, and habitual use are not a funnel cartoon
Why we refuse to draw the three stages as a pretty triangle on slide one.
When to retire a feature that never crossed the adoption floor
Missing twice is usually enough. Politics will ask for a third window anyway.