Usage-Based Pricing

AWS charges you nothing when your servers sit idle and a lot when they are humming, and that single alignment helped make it one of the fastest-growing businesses in history. When Amazon launched it in 2006, the pitch was radical: pay only for what you use, no upfront commitment, no minimum, no multi-year contract. A startup could ship a product without buying a server. When a customer grew, AWS earned more; when they shrank, it charged less. Twilio metered messages, Stripe took a slice of every transaction, Snowflake billed by the credit. The bill became a function of activity, not a flat seat fee.

For founders and operators, usage-based pricing lowers the barrier to start and scales naturally with customer success, but it hands you a real cost: a subscription CFO knows next month's revenue within a few percent, and a usage CFO knows it lands somewhere in a wide range. In a downturn, customers cut workloads and your revenue falls fast while subscription rivals coast on locked contracts. It only works when one specific thing is true about the unit you are metering, and there is a hybrid structure that captures the upside without the volatility. What that condition is, and the model that fixes the downside, is what the app holds back.