The machine
Layer 0: Delivery
The precondition. Nothing above this works until this is true.
The most common objection to any argument about expansion, and it is the right instinct pointed at the wrong number. Two conditions genuinely preclude expansion. A churn rate is not one of them, because churn is an outcome with three different causes behind it and they point in opposite directions.
In this layer
-
Real downward pressure exists. But when a customer says they could rebuild your product in a weekend, what dropped was their belief about what they were buying, not the product. That belief came from somewhere, and where it came from is usually you.
-
Five mandates in six months, each one correct when it was issued, and the accounts don't reset between them. The finance cost is the small one. The expensive part is that a customer who got a concession under one mandate and a refusal under the next has not experienced a policy update.
-
The model asks customers to become excellent at something that isn't their job, and asks your team to be world-class at getting them to. Two impossible asks, stacked, and then everyone acts surprised about the burnout.
-
The most defensible subscriptions are the ones where the customer is getting something new every month — not just access to something they could have gotten last month.
-
Every cost-to-serve cut gets defended with a real number, and almost none get modeled against retention. The test is embarrassingly simple and nobody applies it: did you also lower the price?