.png)
How It Works
AWS Savings Plans are a pricing model where you commit to spending a specific dollar amount per hour on AWS compute for a 1-year or 3-year term. In return, AWS applies discounted rates to your eligible usage automatically. Unlike Reserved Instances, which are tied to a specific instance type and region, Savings Plans apply discounts based on your hourly spend commitment rather than a specific resource. AWS offers three types: Compute Savings Plans (the most flexible, covering EC2, Fargate, and Lambda across all regions and instance families), EC2 Instance Savings Plans (higher discounts for a specific instance family in one region), and SageMaker Savings Plans (covering SageMaker workloads). The equivalent pricing model on Azure is called Azure Savings Plans. GCP uses a different mechanism called Committed Use Discounts (CUDs).
Why It Matters for Cloud Cost
AWS compute is typically the largest line item on an AWS bill. Without a Savings Plan or Reserved Instance in place, every EC2 instance, Fargate task, and Lambda invocation runs at on-demand rates, the most expensive pricing tier. For a company spending $500,000 per year on compute, that gap between on-demand and discounted rates can represent hundreds of thousands of dollars in unnecessary spend. The challenge is that managing Savings Plans requires accurate forecasting of future usage, daily monitoring of commitment utilization, and a willingness to absorb financial risk if usage drops. Most teams lack the time or tooling to do this well, which means they either under-purchase and leave savings on the table, or over-purchase and pay for unused commitment.
Key Characteristics
- Compute Savings Plans apply discounts automatically across EC2, Fargate, and Lambda regardless of instance family, size, OS, or region, making them the most portable commitment type.
- EC2 Instance Savings Plans offer up to 72% savings versus on-demand for a specific instance family and region, delivering higher discounts in exchange for less flexibility.
- Savings Plans are purchased as a dollar-per-hour commitment, not a resource reservation, so they adapt to workload changes without requiring manual modification.
- Any usage not covered by a Savings Plan commitment automatically falls back to on-demand pricing, making right-sizing the commitment critical to maximizing savings.
How Usage AI Handles This
Usage AI’s Autopilot mode manages AWS Savings Plans commitments daily on the customer’s behalf, purchasing and adjusting coverage to match actual usage without requiring manual oversight. The Usage Flex Savings Plan covers EC2, Fargate, and Lambda with savings of 40 to 60% versus on-demand, $0 upfront, and a cashback plus credits guarantee on any underutilization.
See how Usage AI saves 30 to 50% on AWS, GCP, and Azure.
Common Questions
What is the difference between AWS Savings Plans and Reserved Instances?
AWS Savings Plans apply a discount based on your hourly spend commitment across eligible services, giving them flexibility to move across instance families and regions. Reserved Instances lock discounts to a specific instance type and region but can offer savings of up to 72% for the most specific configurations. For most teams, Savings Plans are easier to manage because they do not require changes when workloads shift.
What happens if my usage falls below my Savings Plan commitment?
You are billed for the full committed hourly amount whether or not you use it. If your actual spend is lower than the commitment, you pay for the unused portion with no discount applied to that gap. This financial risk is one of the primary reasons teams either avoid commitments entirely or under-purchase them relative to their actual usage.
Do AWS Savings Plans cover all AWS services?
No. Compute Savings Plans cover EC2, Fargate, and Lambda. SageMaker Savings Plans cover SageMaker workloads. Services such as RDS, ElastiCache, Redshift, and DynamoDB are not covered by Savings Plans and instead use Reserved Instances for commitment-based discounts.