Skip to article
ecomservices.io
Subscriptions · How to run this
EconomicsCalculatorApprovalCalculatorDunningCalculatorAccount updaterCalculatorIdeasCalculatorImplementationNot on this page yetOptimisationNot on this page yetProjected cashNot on this page yet

Library · Subscriptions · How to run this

Day-2 refund is not rebill-day drop

How to run thisSubscriptions3 steps
CalculatorDay-2 refund spike · Calculator lives in the articleOpen

Day-2 refunds ÷ converts make spike %. Type both.

Pair: Day-2 refund spike on Subscriptions. Rebill: rebill-day drop. Keep: convert vs refund.

What you type

Day-2 refunds — converts that refunded by end of day 2.

Converts — paid converts in that same cohort window.

Spike % = day-2 refunds ÷ converts × 100. Fix offer or fulfillment when spike % is fat — do not invent a spike chip.

What not to do

Do not invent spike %. Do not mix day-N rebill drops into day-2 refunds. Do not treat trial starts as converts. Do not average every offer when one owns the spike.

Related: Subscriptions, rebill-day drop, convert vs refund, EPC vs refund rate.

Procedure

  1. Type day-2 refunds and converts.
  2. Read spike %.
  3. Fix offer or fulfillment when spike % is material.

Failure modes

  • Inventing spike %.
  • Mixing day-N rebill drops into day-2 refunds.
  • Treating trial starts as converts.

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

Day-2 refund spike %
Day-2 refunds ÷ converts × 100.

Work with us — We run this job on your accounts. Services

FAQ

Is this rebill-day drop?
Sibling BuyGoods clock. [Rebill-day drop](/manuals/buygoods-rebill-day-drop) is day-N due vs billed. This page is day-2 refunds ÷ converts.
Is this trial convert vs refund?
Related refund grain. [Convert vs refund](/manuals/buygoods-trial-convert-vs-refund) pairs keep vs refund. This page is the day-2 refund spike.
What spike % should I invent?
None. Type day-2 refunds and converts from the same BuyGoods window.