Learn / Glossary

The FinOps & cloud cost glossary

Plain-English definitions for the terms behind cloud cost, compliance, reliability and monitoring: the same vocabulary emnode uses across the product and these lessons.

Where this glossary uses the FinOps discipline’s vocabulary, it follows the framework defined by the FinOps Foundation. The definitions here are our own, written for practitioners. Reviewed by Nathan and Michael.

FinOps fundamentals

The practice and vocabulary of cloud financial management. The discipline and its framework are defined by the FinOps Foundation; the definitions below are our own, written for practitioners.

FinOps: FinOps is an operational practice that brings finance, engineering and leadership together to manage cloud spend with shared accountability, so usage decisions are made with cost in mind. It is a continuous discipline, not a one-off cost-cutting project.
Inform, Optimize, Operate: These are the three iterative phases of the FinOps framework: Inform (visibility and allocation), Optimize (better rates and usage), and Operate (continuous governance). Teams cycle through them repeatedly rather than finishing them once.
Cost allocation: Cost allocation assigns each line of cloud spend to the team, product or environment that caused it, usually through tags and account structure. Without it, a shared bill cannot be made anyone’s responsibility.
Showback: Showback reports each team’s cloud cost back to them for visibility, without moving any money. It builds cost awareness and is often the step before chargeback.
Chargeback: Chargeback bills each team or business unit for the cloud cost they actually incurred, moving the spend onto their own budget. It turns cost from information into a real constraint.
Unit economics: Unit economics expresses cloud cost per unit of business value (cost per customer, per order, per GB delivered) so spend can be judged against what it produces rather than in absolute pounds.
Tag coverage: Tag coverage is the share of cloud spend or resources carrying the tags needed to allocate it. Low coverage leaves "untagged" spend that cannot be attributed to an owner.

Cost & optimisation

Finding and removing waste, and matching spend to what you actually run.

Rightsizing: Rightsizing matches a resource’s size to what it actually uses (for example moving an over-provisioned EC2 instance to a smaller type) cutting cost without hurting performance. Right-size your compute
Idle resources (waste): Idle or wasted spend is money going to resources that are provisioned but doing nothing: unattached EBS volumes, idle load balancers, orphaned IPs. It is usually the fastest saving to realise because deleting it carries no performance risk. Kill idle waste
Cost anomaly detection: Cost anomaly detection flags spend that deviates sharply from its expected pattern (a sudden daily spike) so a runaway cost is caught within days instead of at the month-end invoice.
Forecast: A cloud forecast projects month-end or future spend from current usage and trends, so a budget can be defended before the bill lands rather than explained after it.

Commitments & rates

Paying a lower rate for steady-state usage, and the metrics that say whether it is working.

Commitment: A commitment is an agreement to pay for a baseline of cloud usage over one or three years in exchange for a lower rate. On AWS these are Reserved Instances and Savings Plans; on Azure, Reservations and savings plans. Lock in your commitments
Reserved Instance (RI): A Reserved Instance is a one- or three-year AWS commitment to a specific instance configuration in return for a discount versus on-demand pricing.
Savings Plan (SP): A Savings Plan is a flexible AWS commitment to a steady amount of compute spend, measured in pounds per hour, applied automatically across eligible usage for a lower rate.
Coverage: Coverage is the percentage of eligible on-demand spend that your commitments actually cover. Low coverage means you are paying full on-demand rates for steady-state workloads that could be discounted.
Utilisation: Commitment utilisation is the percentage of a commitment you actually use. Low utilisation means you have over-committed and are paying for discounted capacity that sits idle.
Effective Savings Rate (ESR): Effective Savings Rate is the blended discount your commitments really deliver, total savings as a percentage of what the same usage would have cost at on-demand rates. It is the single number that says whether your rate strategy is working. Commitments in emnode

Compliance & governance

Measuring cloud security posture continuously and fixing what matters first.

Compliance posture: Compliance posture is how well your cloud configuration meets a security framework’s controls at a point in time, expressed as a score and a ranked list of findings rather than a once-a-year pass or fail. Compliance in emnode
Finding: A finding is a single detected deviation from a control (for example an unencrypted volume) carrying a severity from critical to low that determines how urgently it should be fixed.
MTTR (Mean Time To Resolve): Mean Time To Resolve is the average time between a finding being raised and remediated. It measures how quickly issues actually get fixed, not just how many exist.
Remediation SLA: A remediation SLA is the target time to fix a finding of a given severity, for example criticals within 7 days. SLA-met percentage shows whether the team is hitting those targets. AWS Security Hub controls

Reliability & monitoring

Knowing your data would recover, and that the right things are being watched.

DR posture: Disaster recovery posture is a continuously scored view of whether your data would actually recover (backup coverage, job success, retention and cross-region copies) rather than a snapshot from the last audit. DR in emnode
RPO and RTO: Recovery Point Objective (RPO) is how much data you can afford to lose, measured in time; Recovery Time Objective (RTO) is how long recovery may take. Backups are judged against both.
Backup coverage: Backup coverage is the share of in-scope resources that sit inside a governed backup plan. The gap is the set of resources with no recoverable copy when one is needed. Build in resilience
Monitoring maturity: Monitoring maturity is a weighted score of how well your estate is actually watched (alarm posture, alarm history and coverage) trended over time, not just whether monitoring is switched on. Monitoring in emnode
Alarm coverage: Alarm coverage is the percentage of resources that have alarms attached, exposing new resources that went live without anything watching them. Get your alarms right

Want the lessons behind the terms?

Browse the library