To use AWS Cost Explorer like a FinOps practitioner, you stop looking at the total and start decomposing it: group by service to see where the money goes, group by usage type to see what specifically is being charged, switch on the amortized and unblended cost views to read commitments honestly, and lean on tags and the daily granularity to attribute spend to teams and catch anomalies. Cost Explorer will not optimise anything for you, but used this way it tells you exactly where to point a rightsizing, commitment, or cleanup effort, which is the See step of our See, Cut, Lock, Run method.
This how-to sits under our complete guide to AWS cost optimization, the pillar for this cluster. Cost Explorer is the interactive front end; for repeatable, queryable analysis at scale, pair it with the AWS Cost and Usage Report, and once Cost Explorer points you at the waste, act on it with the idle EC2 cleanup process.
A bill that rose 8 percent could be healthy growth or pure waste. Cost Explorer earns its keep only when you decompose that 8 percent into the services and usage types behind it. Grouping is the whole skill.
Group by service, then by usage type
Start every investigation by grouping monthly cost by service. This answers "where does the money go" in one view and usually surfaces a top three that account for the majority of the bill, often EC2, S3, and data transfer or RDS. Then narrow to the largest service and re-group by usage type. Usage type is where the real diagnosis happens: it separates EC2 running hours from EBS storage from data transfer out, and it tells you whether your EC2 line is compute you can rightsize or network egress you need to architect away. Practitioners rarely stop at the service level, because the action is always in the usage type underneath it.
Read commitments honestly with amortized and unblended views
Cost Explorer offers several cost views and they answer different questions. Unblended cost shows what you were charged as the charges actually landed, including the up-front portion of a Reserved Instance or Savings Plan on the day you bought it, which makes a purchase month look like a spike. Amortized cost spreads those commitment charges evenly across the term, which is the honest way to read whether your commitments are actually lowering your run rate. When you are evaluating coverage and savings, use amortized; when you are reconciling against an invoice, use unblended. Mixing them up is the most common way a cost report misleads its audience.
Filter to a team with tags
The total bill is nobody's problem until it has an owner. If you have activated cost allocation tags, Cost Explorer lets you filter and group by them, so you can show a team exactly what their environment costs and how it is trending. This is the bridge from a central cloud bill to local accountability, and it is the foundation of showback and chargeback. If your tag coverage is poor, that is the first thing to fix, because no amount of clever filtering recovers spend you cannot attribute. Our guide to cost allocation tags that actually stick covers how to get coverage high enough that these views are trustworthy.
Want the analysis done for you?
Our AWS cost audit decomposes your bill the way a FinOps practitioner would, by service, usage type, account, and team, and hands you a ranked list of the dollars to recover and how. On the performance model you pay only from realized savings. No savings, no fee.
Book an AWS cost audit →Use daily granularity to catch anomalies
Monthly views smooth over the spikes that matter. Switch to daily granularity on your largest services and you will spot the day a misconfigured job started writing terabytes to S3, or a forgotten test fleet that has been running every night for three weeks. Cost Explorer also exposes the RI and Savings Plans utilization and coverage reports, which tell you whether the commitments you bought are actually being used. A commitment sitting at 60 percent utilization is leaking money, and that report is where you find it. For the deeper read on those two numbers, see RI coverage and utilization explained.
| Question | Group / view | What it reveals |
|---|---|---|
| Where does the money go? | Group by service | The top three cost drivers |
| What exactly is charged? | Group by usage type | Compute vs storage vs egress |
| Are commitments working? | Amortized cost view | True run rate after discounts |
| Who owns this spend? | Filter by cost allocation tag | Team-level accountability |
| What spiked? | Daily granularity | Anomalies and runaway jobs |
Cost Explorer views, granularity options, and the RI and Savings Plans reports reflect AWS as of May 2026. Confirm the current feature set and any API request charges in the AWS documentation, as the console evolves.
The AWS Cost Optimization Field Guide includes the saved Cost Explorer view configurations we set up for clients and the anomaly thresholds we watch. It is the downloadable companion to this how-to.
The short version
Never read the total; decompose it. Group by service to find the drivers, by usage type to diagnose them, and by tag to assign ownership. Use amortized cost to judge commitments honestly and unblended to reconcile invoices. Drop to daily granularity to catch anomalies, and watch the utilization and coverage reports for leaking commitments. Cost Explorer is the See step; turning what it shows into recovered spend is an AWS cost optimization engagement, the approach behind our SaaS on AWS case study.