Loading...
Commit to spending a fixed amount per hour and get discounts with built-in flexibility.
Quick Definition
Savings Plans are a flexible commitment-based pricing model offering discounts in exchange for committing to a consistent amount of compute spend (measured in dollars per hour) over one or three years. They cover broader usage than Reserved Instances while delivering similar savings
Savings Plans are a commitment-based discount where you promise a steady spend rate, for example $50 per hour of compute, for one or three years, and receive discounts comparable to reserved instances with far fewer strings. The commitment is to dollars, not to a specific instance type.
That flexibility is the point. Compute-type savings plans apply across instance families, sizes, regions, and even container and serverless usage, so normal architectural evolution does not strand the discount the way rigid reservations can. The remaining risk is the floor itself: commit above your true baseline and the unused commitment bills anyway.
Example. A company commits to $80 per hour against a measured baseline of $110. Over the year it migrates services between instance families twice; the plan keeps applying automatically, utilization stays near 100 percent, and the effective savings rate lands at 31 percent.
Build coverage in layers, committing conservatively and topping up quarterly as the baseline proves itself. The Cloud Cost Management Guide walks through commitment strategy, and the FinOps KPIs guide covers the utilization metrics to watch.
Savings plans for most compute, thanks to flexibility. RIs remain relevant for specific services and some database engines.
It simply bills at on-demand rates. The plan discounts usage up to the committed rate; the rest overflows normally
Below your proven 24/7 baseline, often 70 to 90 percent of it, then increase in increments as confidence grows.