Library · Subscriptions · How to run this
Profile merge is not segment stale
Merge collisions ÷ merges attempted make collision %. Type both.
Pair: Profile-merge collision rate on Subscriptions. Stale: segment stale rate. Unsub: flow unsub cost.
What you type
Merge collisions — profile merges that hit an identity collision / failed clean merge.
Merges attempted — profile merges attempted in that same window.
Collision % = merge collisions ÷ merges attempted × 100. Fix identity rules when collision % is fat — do not invent a collision chip.
What not to do
Do not invent collision %. Do not mix segment stale counts into merge collisions. Do not treat suppressions as merges. Do not average every list when one owns the hole.
Related: Subscriptions, segment stale rate, flow unsub cost, predictive vs actual LTV.
Procedure
- Type merge collisions and merges attempted.
- Read collision %.
- Fix identity rules when collision % is material.
Failure modes
- Inventing collision %.
- Mixing segment stale counts into merge collisions.
- Treating suppressions as merges.
Worked example
None on this page yet. When one lands it will be a real case: what broke, the number, and what changed. Not a composite.
Glossary
- Profile-merge collision %
- Merge collisions ÷ merges attempted × 100.
Work with us — We run this job on your accounts. Services
FAQ
- Is this segment stale rate?
- Sibling list grain. [Segment stale rate](/manuals/klaviyo-segment-stale-rate) is stale profiles ÷ segment size. This page is merge collisions ÷ merges attempted.
- Is this flow unsub cost?
- Related lifecycle burn. [Flow unsub cost](/manuals/klaviyo-flow-unsub-cost) is unsub LTV burn. This page is profile-merge collision rate.
- What collision % should I invent?
- None. Type merge collisions and merges attempted from the same Klaviyo window.