Snowflake: the aggregate is hiding the cohort

This runs through cohort behaviour rather than through the blended metric. Each enterprise cohort renews on its own anniversary, so a deteriorating recent cohort is invisible in the aggregate until it becomes a large enough share of the renewal base. The cohort data is in the supplemental rather than the release. Net revenue retention for accounts acquired in the first enterprise wave has fallen for three consecutive quarters while the blended figure was flat. The position: short on Snowflake, entry locked at publication and the exit dated rather than open-ended. The target is stated in the call block above and the horizon with it. Sized to the catalyst rather than to conviction, with the invalidation written into the kill-switch card instead of left implicit. What breaks this is a re-acceleration in the weak cohort. Renewal behaviour has inflected before on a single product release, and one good release would invalidate the cohort argument entirely.