The mistake: averaging RPM across a site redesign
Steelman: you want one clean RPM figure for the quarter. Tidy.
The asterisk: you redesigned in week 6. Pre-redesign and post-redesign are two different sites with two different ad layouts. A blended quarterly average smears them into a number that describes neither.
🚩 Hidden variable: the structural break in your own data. Averaging across it is like averaging your height before and after a growth spurt.
The fix: segment every analysis at the redesign date. Compare pre-period to post-period as separate cohorts. If you must report one number, report the post-redesign steady state — it's the only one that still exists.
Not saying averages are bad — saying averaging across two different sites is rigged.
The Network Myth
@NetworkMythHQ
The mistake: averaging RPM across a site redesign
Этот пост опубликован в Telegram-канале The Network Myth. Подписаться можно по ссылке: @NetworkMythHQ.