Architecture Cost Model
A calculation, made during design, of what an architecture will cost to run at expected and at peak volume.
Architecture decisions are spending decisions, and the cheapest time to discover that a design costs three times the budget is before it is built.
A usable model estimates per component: compute (instances × hours, or requests × price), storage (volume × price, growing over time), data transfer — which is the one most often omitted and most often decisive — managed service fees, observability, and third-party API calls. Then it computes a unit cost at expected volume, so it can be compared against revenue per unit.
Two habits make it worth the effort. Model at least two options, so the number is comparative rather than absolute. And model at 1×, 10× and 100× volume, because the cheapest design at launch volume is frequently not the cheapest at scale, and the crossover point is a fact worth knowing in advance.