What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Saving more than 20% on cloud costs is achievable when you first remove waste, then commit only to the portion of usage you can reliably forecast. AWS and Azure advertise discounts above that threshold for qualifying services and commitments, but their “up to” rates are not a promise to cut your total bill by the same amount.
What “save 20%” should mean for your cloud bill
Measure the goal against your actual baseline, not a provider’s advertised discount. Separate compute, storage, databases, networking and support charges, then compare your bill before and after optimization over equivalent periods. A discount on eligible compute does not apply automatically to every line item, and a commitment that goes unused can erase some or all of its apparent benefit.
The scale of the issue is substantial: the FinOps Foundation’s 2025 State of FinOps survey covered organizations responsible for more than $69 billion in cloud spend. That is a survey measure of participating organizations’ spend, not a forecast of what any one company can save.
Cut waste before you commit
Start with the resources you already pay for. Identify idle services and workloads that use less capacity than they have provisioned. Rightsize those resources, shut down what is not needed, and reassess the resulting bill. Only then should you estimate the steady baseline that could support a long-term pricing commitment.
Recommended Free Tools
#1 Best Overall
- Compute: Look for instances or other compute resources that are idle, oversized or running outside the hours they are needed.
- Storage and databases: Check for unused capacity and resources that remain active after their workload has ended.
- Networking and support: Keep these charges visible in the baseline; compute discounts do not establish savings on either category.
- Forecasts and utilization: Track actual usage against expected usage so commitments reflect demand rather than a temporary spike.
The FinOps Foundation treats rightsizing and commitment management as related public-cloud capabilities. That order matters: buying a discount against an inflated baseline can leave you paying for capacity you no longer need.
Choose a discount that fits the workload
The published percentages below are provider ceilings for qualifying usage, compared with the named pay-as-you-go or On-Demand price. They are not account-wide savings estimates. A deeper discount generally comes with narrower eligibility, less flexibility, interruption risk or a longer obligation.
Rank #2
| Option | Published maximum | Best fit and trade-off |
|---|---|---|
| AWS EC2 Instance Savings Plans or Standard Reserved Instances | Up to 72% versus On-Demand pricing, according to AWS documentation | Consider for a stable, forecastable EC2 baseline. These options are more targeted than Compute Savings Plans; confirm that the covered usage will remain eligible. |
| AWS Compute Savings Plans | Up to 66% versus On-Demand pricing, according to AWS documentation | Broader flexibility across instance families and services than EC2 Instance Savings Plans, in exchange for a lower published ceiling. |
| Azure Reservations | Up to 72% from pay-as-you-go prices, according to Microsoft Azure documentation | For eligible resources with predictable demand. Azure Reservations generally use one- or three-year commitments. |
| Azure Savings Plan for Compute | Up to 65% on eligible compute usage, according to Microsoft guidance | For eligible compute usage that can support a commitment. Check which usage qualifies before estimating savings. |
| AWS Spot Instances | Up to 90% discount, according to AWS documentation | For fault-tolerant, restartable or batch work that can withstand interruption. AWS may reclaim capacity when it is needed elsewhere. |
AWS Savings Plans and Azure Reservations generally require one- or three-year commitments. A longer commitment can make a published discount attractive, but only if the associated usage persists. AWS describes Spot as offering its largest discounts, up to 90%, while warning that instances can be interrupted when capacity is needed elsewhere.
Match each pricing option to the right demand
Use commitments for the stable floor
For services that run predictably, estimate the minimum usage you expect to retain after rightsizing. That dependable floor is the portion to evaluate for a Savings Plan or Reservation. Favor broader flexibility when demand may move across eligible instance families or services; consider a more targeted option only when the underlying usage is well understood.
Rank #3
Use Spot for work that can recover
Spot is a poor fit for a process that must run uninterrupted. It can suit batch jobs, fault-tolerant services and workloads that can restart or resume after capacity is reclaimed. Account for the engineering and operational work needed to handle interruption before comparing its discount with a commitment-based option.
Keep uncertain demand flexible
When future usage is unclear, avoid committing the uncertain portion merely to reach a target discount. Pay-as-you-go flexibility may cost more per unit but avoids an obligation based on a forecast that fails to materialize.
A practical sequence for lowering the bill
- Establish a baseline. Break the cloud bill into compute, storage, database, network and support charges, and choose a representative period for comparison.
- Find and remove waste. Identify idle or underused resources, rightsize them and clean up resources no longer needed.
- Recalculate expected demand. Use post-optimization usage to separate the predictable baseline from variable or uncertain workloads.
- Assign the right pricing model. Evaluate Savings Plans or Reservations against stable demand, and Spot only for work designed to tolerate interruption.
- Check the commitment before purchase. Confirm covered services, eligible usage, term and expected utilization in the provider’s current documentation and account tools.
- Review results continuously. Compare commitment coverage and utilization with actual usage, and update forecasts as workloads change.
Count net savings, not the headline discount
Calculate whether the change lowers the bill after including unused commitments, migration work, engineering effort and interruption handling. Compare like with like: the same workloads, service scope and time period, while accounting for any changes in demand. If a commitment sits unused or the workload shifts outside its eligible scope, the advertised discount is not realized as a saving.
Treat 20% as a target to test against measured spend, not an automatic outcome. The FinOps Foundation reported in 2024 that fewer than 20% of FinOps teams collaborated with sustainability teams. That finding concerns collaboration, not cost reduction; where practical, teams can consider cost and sustainability goals together without assuming one proves the other.
The Tool Desk
Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →Quick Recap
Best Value
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
