.png)
Key takeaways
1
Commitments lower cloud costs: Predictable usage can qualify for lower rates than on-demand pricing.
2
Unused commitments create risk: Changing workloads can leave teams paying for unused capacity.
3
We automate commitments: Usage.ai identifies opportunities and manages eligible commitments for you.
4
Flex Commitments add protection: Eligible unused commitments can receive cashback protection under program terms.
5
You pay for realized savings: Fees are tied to the savings Usage.ai actually delivers.
We help teams optimize cloud commitments across AWS, Azure, and GCP. We analyze billing and usage patterns, identify commitment opportunities, and manage eligible commitments on your behalf. With Flex Commitments, we also provide cashback protection for eligible financial downside under the applicable program terms.
Here is how the model works, why commitment risk matters, and where Usage.ai fits into the process.
Savings Plans work differently. Instead of committing to a particular EC2 configuration, you commit to a consistent amount of compute usage measured in dollars per hour. AWS offers one-year and three-year terms. Compute Savings Plans provide up to 66% savings, while EC2 Instance Savings Plans can provide up to 72% savings for eligible usage. See the AWS Savings Plans documentation.
GCP and Azure have their own commitment programs, with different terms, services, pricing, and eligibility rules. The basic idea is similar: commit to a predictable level of usage and receive a lower effective rate.
The tradeoff is utilization. With AWS commitments, you can continue paying according to the commitment terms even when your actual usage falls. That makes forecasting and coverage decisions important.
For more detail, see our AWS Reserved Instances guide and AWS Savings Plans guide.
Here is how the model works, why commitment risk matters, and where Usage.ai fits into the process.
How cloud commitments work
An Amazon EC2 Reserved Instance is a pricing discount, not a physical server. AWS offers Standard and Convertible RIs with one-year or three-year terms and All Upfront, Partial Upfront, or No Upfront payment options. Standard RIs can provide up to 72% savings compared with On-Demand pricing, depending on the instance configuration and payment option. See the AWS EC2 Reserved Instances pricing documentation.Savings Plans work differently. Instead of committing to a particular EC2 configuration, you commit to a consistent amount of compute usage measured in dollars per hour. AWS offers one-year and three-year terms. Compute Savings Plans provide up to 66% savings, while EC2 Instance Savings Plans can provide up to 72% savings for eligible usage. See the AWS Savings Plans documentation.
GCP and Azure have their own commitment programs, with different terms, services, pricing, and eligibility rules. The basic idea is similar: commit to a predictable level of usage and receive a lower effective rate.
The tradeoff is utilization. With AWS commitments, you can continue paying according to the commitment terms even when your actual usage falls. That makes forecasting and coverage decisions important.
For more detail, see our AWS Reserved Instances guide and AWS Savings Plans guide.
Warning:
A commitment is not automatically a saving. If you commit to usage that disappears, the discount may not offset the cost of unused commitment. How Flex Commitments work
This is where our model differs from simply buying commitments yourself.We built Usage.ai around a straightforward problem: cloud usage changes, but long-term commitments do not always change with it.
Our Flex Commitment workflow has five steps:
- Analyze your usage. We connect to your cloud billing and usage data and establish a baseline from actual workload behavior.
- Identify opportunities. We look for eligible usage where a commitment can reduce your effective cloud cost.
- Approve or automate purchases. Depending on your setup, you can review recommendations before we act or enable automation for eligible purchases.
- Manage commitments continuously. We monitor the environment as usage changes instead of treating the original purchase as a one-time decision.
- Protect eligible downside. For commitments that meet the Flex Commitment Program requirements, we provide cashback protection for eligible financial downside under the applicable program terms.
The result is a commitment strategy designed around actual usage rather than a single three-year forecast.
Why commitment risk matters
The biggest challenge with commitments is not finding a discount. It is deciding how much usage you can confidently commit to.Imagine a company with a stable $70,000 monthly EC2 baseline. It commits against that entire amount, but later its usage falls to $56,000.
The following is an illustrative calculation, not a customer result.
Illustrative baseline
- Original baseline: $70,000/month
- Later usage: $56,000/month
- Difference: $14,000/month
- Calculation: $70,000 – $56,000 = $14,000
This is why commitment coverage needs to be treated as an ongoing optimization problem.
You want enough coverage to capture meaningful savings, but not so much that a change in infrastructure creates unnecessary financial exposure.
How we decide what to commit
We start with the usage baseline rather than simply taking the highest recent spend.A practical commitment process looks like this:
Step 1: Remove obvious waste Rightsize resources, eliminate idle workloads, and account for known infrastructure changes before establishing the baseline.
Step 2: Separate stable and changing usage
Stable workloads are better candidates for long-term commitments. Usage affected by migrations, seasonal demand, rightsizing, or architectural changes needs more caution.
Step 3: Match the commitment type
AWS offers different commitment products with different levels of flexibility. For example, a Compute Savings Plan can apply across eligible compute usage, while an EC2 Instance Savings Plan is narrower.
Step 4: Choose coverage
Do not assume 100% coverage is always the right answer. Coverage should reflect both the stability of your workload and the amount of financial flexibility you need.
Step 5: Keep monitoring
Commitment management does not end after purchase. Expirations, changing usage, new workloads, and infrastructure changes can all affect the optimal coverage level.
AWS itself recommends monitoring Savings Plans and making purchases over time as usage changes. See the AWS Well-Architected cost optimization guidance.
Choosing the right approach
Different workloads call for different commitment strategies.| Situation | Practical approach |
|---|---|
| Stable workload and strong forecast confidence | Consider longer-term provider commitments |
| Compute changes across instance families or regions | Consider an AWS Compute Savings Plan |
| Usage is changing quickly | Use a conservative baseline |
| Multiple clouds or accounts increase management effort | Consider automated commitment management |
The important point is not to maximize commitment coverage at any cost. It is to maximize useful coverage while keeping the strategy aligned with how your infrastructure actually changes.
For organizations managing many accounts and services, that process becomes difficult to maintain manually.
You also need to monitor utilization, track expiration dates, identify new opportunities, and adjust coverage when infrastructure changes.
AWS provides native tools for this. For example, Savings Plans inventory provides information about plans, accounts, commitment amounts, and term dates. When a commitment expires, eligible usage can return to On-Demand pricing unless another commitment is available.
We automate much of this ongoing process.
Instead of asking your FinOps team to repeatedly check multiple cloud consoles, we continuously evaluate the environment for eligible opportunities and help keep commitment coverage aligned with actual usage.
That matters because the optimal commitment strategy can change even when the underlying business has not.
A migration can move workloads between services. Rightsizing can reduce baseline consumption. A new application can increase it. Seasonal demand can temporarily distort the numbers.
Commitment management needs to account for those changes.
Flex Commitments introduce a separate protection mechanism for eligible commitments.
We provide cashback for eligible financial downside when a qualifying Flex Commitment costs more than equivalent On-Demand usage, subject to the current Flex Commitment Program requirements and terms.
This protection applies to eligible Flex Commitments. It should not be interpreted as automatic protection for every RI, Savings Plan, or other commitment that a customer already owns.
That distinction matters.
Your existing cloud commitments continue to operate according to the provider’s terms. Flex Commitments are separately managed under our program and have their own eligibility requirements.
You can review the Flex Commitment eligibility requirements before deciding whether the model fits your environment.
We charge a percentage of realized savings rather than a separate platform fee. If we do not generate savings, you do not pay a savings-based fee.
Current pricing, terms, and eligibility can change, so the Usage.ai pricing page is the source of truth for the current commercial model.
The same principle applies to the protection program. Cashback applies only where the applicable Flex Commitment Program requirements are met.
That keeps the product model connected to the outcome you care about: reducing cloud spend while managing the downside associated with long-term commitments.
For organizations managing many accounts and services, that process becomes difficult to maintain manually.
How we manage the commitment lifecycle
Buying a commitment is only one part of the job.You also need to monitor utilization, track expiration dates, identify new opportunities, and adjust coverage when infrastructure changes.
AWS provides native tools for this. For example, Savings Plans inventory provides information about plans, accounts, commitment amounts, and term dates. When a commitment expires, eligible usage can return to On-Demand pricing unless another commitment is available.
We automate much of this ongoing process.
Instead of asking your FinOps team to repeatedly check multiple cloud consoles, we continuously evaluate the environment for eligible opportunities and help keep commitment coverage aligned with actual usage.
That matters because the optimal commitment strategy can change even when the underlying business has not.
A migration can move workloads between services. Rightsizing can reduce baseline consumption. A new application can increase it. Seasonal demand can temporarily distort the numbers.
Commitment management needs to account for those changes.
What makes Flex Commitments different
Traditional cloud commitments put the forecasting risk largely on the customer. You decide how much to commit, purchase the commitment, and absorb the financial impact if the workload changes.Flex Commitments introduce a separate protection mechanism for eligible commitments.
We provide cashback for eligible financial downside when a qualifying Flex Commitment costs more than equivalent On-Demand usage, subject to the current Flex Commitment Program requirements and terms.
This protection applies to eligible Flex Commitments. It should not be interpreted as automatic protection for every RI, Savings Plan, or other commitment that a customer already owns.
That distinction matters.
Your existing cloud commitments continue to operate according to the provider’s terms. Flex Commitments are separately managed under our program and have their own eligibility requirements.
You can review the Flex Commitment eligibility requirements before deciding whether the model fits your environment.
How pricing works
Our pricing is tied to the savings we generate.We charge a percentage of realized savings rather than a separate platform fee. If we do not generate savings, you do not pay a savings-based fee.
Current pricing, terms, and eligibility can change, so the Usage.ai pricing page is the source of truth for the current commercial model.
The same principle applies to the protection program. Cashback applies only where the applicable Flex Commitment Program requirements are met.
That keeps the product model connected to the outcome you care about: reducing cloud spend while managing the downside associated with long-term commitments.
Frequently asked questions
Does Usage.ai replace existing RIs and Savings Plans?
No. Existing customer-owned commitments continue under their original cloud-provider terms. We can identify additional eligible opportunities rather than requiring you to replace every commitment you already have.
Does Usage.ai work beyond AWS?
Yes. We support commitment optimization across AWS, Azure, and GCP. The specific commitment products, services, terms, and eligibility rules vary by provider.
What happens if my usage falls after a Flex Commitment is purchased?
With eligible Flex Commitments, we provide cashback protection to help cover the difference when a commitment costs more than equivalent On-Demand usage, subject to current program eligibility and terms. This protection does not automatically apply to commitments purchased independently by the customer.
How does Usage.ai charge?
We charge a percentage of realized savings. Our current pricing information states that there is no separate platform fee or contract minimum. See the current Usage.ai pricing for applicable terms.
How long does setup take?
Usage.ai currently states that setup takes about 15 minutes. The initial connection uses read-only access for evaluating billing and usage data, with additional scoped permissions required when commitment purchases are enabled.
Share
Facebook
X
LinkedIn
Reddit
