How to read a retention table without flattening the story
A retention table is a set of named groups moving through time. Treating the last column as a score hides the weeks that actually changed the story.
Field notes
Notes from the margin of the table
Short pieces on customer behavior analytics as it shows up in extracts: seasons, thin groups, first-week silence, and the annotations that keep a meeting honest.
When a cohort split hides a seasonal pattern
Join-week cohorts are honest only if the calendar around them is named. Holiday closures and school calendars can look like churn if nobody annotates the weeks.
First-week activity and the people who go quiet
Early clicks and early visits are not the same kind of evidence. This note separates recorded first-week actions from later return in customer behavior files.
Annotating cohort reports so operators can act
A table without the week a price changed, a store closed, or a reminder stopped is a table that invites argument. Annotation is part of the report, not decoration.