- Amortized cost
- A view of the bill where an upfront reservation is spread over its term, day by day. Essential to compare months with each other.
- Azure Hybrid Benefit
- Reusing Windows Server and SQL Server licenses you already own (with Software Assurance) on Azure, so you no longer pay for the license included in the VM price. Learn more →
- Budget alert
- An alert sent when a scope's spend reaches a threshold (for example 80% and 100% of budget). It warns, but stops nothing. Learn more →
- Build vs buy
- Comparing the cost of a managed service with that of an in-house tool, maintenance included. Sometimes a small tool replaces an expensive service.
- CCoE (Cloud Center of Excellence)
- A cross-functional team that sets cloud standards and best practices. FinOps often sits within it.
- Chargeback
- Billing cloud consumption back to each budget or business unit. More empowering than showback: when a team pays, it optimizes. Learn more →
- Cleaning
- Deleting unused resources: unattached disks, old snapshots, orphaned public IPs, forgotten environments.
- CMDB
- The configuration database listing applications, their owners, cost centers and criticality. Linked to the cloud through a tag, it avoids multiplying tags. Learn more →
- Commitment coverage
- The share of eligible consumption covered by a Reserved Instance or a savings plan. A common target is at least 80% of steady compute. Learn more →
- Commitment tier
- A daily commitment level for Log Analytics or Sentinel (100 GB/day, 200 GB/day…), cheaper per GB than pay-as-you-go when volume is stable. Learn more →
- Commitment utilization
- The share of a reservation or savings plan actually consumed. Below 95%, you are paying for nothing. Learn more →
- Cost allocation
- Tying every euro of the cloud bill to an application, a team or a cost center. It is the basis of showback and chargeback. Learn more →
- Cost anomaly
- An unusual, rapid rise in spend, often caused by a configuration change (verbose logs, a loop, a migration). Detecting it in a day rather than a month changes everything. Learn more →
- Crawl, Walk, Run
- The three maturity levels of the FinOps Framework. The aim isn't “Run” everywhere, but the right level where the stakes justify it. Learn more →
- EDP (Enterprise Discount Program)
- An AWS contract in which a company commits to a multi-year spend amount in exchange for an overall discount.
- Effective Savings Rate (ESR)
- The saving actually achieved on compute through commitments, relative to what the same consumption would have cost on demand. Learn more →
- Egress
- Data leaving a cloud or a region, billed per GB. An often-overlooked cost in multi-region or hybrid architectures.
- FinOps
- Cloud Financial Operations: a discipline bringing finance, IT and engineering together to manage cloud spend and maximize the value of every euro. Learn more →
- FinOps by design
- Building FinOps into standards and building blocks, so every new application lands in the cloud already optimized. Learn more →
- FinOps dashboard
- A dashboard showing costs by application and owner, the forecast, savings and recommendations. Refreshed daily, it is the shared reference. Learn more →
- FinOps Foundation
- A non-profit organization, part of the Linux Foundation, that publishes the FinOps Framework, the FOCUS specification and the annual State of FinOps survey.
- FinOps Framework
- The discipline's reference model: principles, personas, phases, domains and capabilities. Learn more →
- FinOps KPI
- An indicator tracked at the SteerCo: tagging, commitment coverage, rightsizing, Stop/Start, cost per VM, forecast accuracy… Learn more →
- FinOps recommendation
- A quantified optimization action on a specific resource (downsize a VM, stop non-prod…), tracked until it is applied and measured on the bill. Learn more →
- FinOps scoring
- A FinOps maturity score out of 100, calculated per application from a weighted grid of best practices per service. Learn more →
- FinOps SteerCo
- A monthly committee tracking consumption, the forecast, savings, KPIs and the decisions to make. Learn more →
- FOCUS
- FinOps Open Cost and Usage Specification: a common format for cloud and SaaS billing data, to compare and aggregate several providers. Learn more →
- Forecast
- A projection of upcoming spend, based on current run rate, announced projects and planned savings. Recalculated daily, it triggers overrun alerts. Learn more →
- GreenOps
- Applying the FinOps approach to the cloud's environmental footprint: measuring emissions and reducing them through sobriety. Learn more →
- Idle
- A resource that is running but barely used (for example average CPU around 2%). It costs as much as an active one.
- List price
- The public price of a service, before negotiated discounts and commitments. Savings estimates are often calculated at list price.
- MACC (Microsoft Azure Consumption Commitment)
- A multi-year contractual Azure consumption commitment, in exchange for commercial terms. To be tracked closely so nothing is left unused.
- On-demand (pay-as-you-go)
- Paying for usage, with no commitment, at list price. The most flexible and the most expensive option for a steady workload.
- Owner
- The person responsible for an application and its costs. Without an identified owner, spend is never optimized.
- P95
- The 95th percentile of a metric: the value below which 95% of measurements fall. A VM is sized on CPU and RAM P95, not on the average.
- Rate optimization
- Paying less for the same consumption: reservations, savings plans, Azure Hybrid Benefit, negotiated discounts. Learn more →
- Reserved Instance
- A 1- or 3-year commitment to a VM size in a region, in exchange for a large discount (around 50% on a production VM). Learn more →
- Rightsizing
- Bringing a resource down to the size it really needs, based on its usage metrics (CPU and RAM P95 over 30 days).
- Savings plan
- A commitment to an hourly compute amount for 1 or 3 years, more flexible than a Reserved Instance as it applies across sizes and regions. Learn more →
- Scopes 1, 2 and 3
- Categories of CO₂ emissions: direct (1), electricity-related (2) and value chain, including server manufacturing (3). For the cloud, scope 3 dominates. Learn more →
- Showback
- Making each team's costs visible, without billing them back. The first step before chargeback. Learn more →
- Spot
- Spare cloud capacity sold at a large discount, but which can be reclaimed at any time. Suited to interruptible workloads and non-prod.
- Start on demand
- Starting a VM when the user asks for it, combined with an automatic Stop every evening. Around 90% savings on non-prod. Learn more →
- Stop/Start
- Stopping VMs when they aren't used. On a fixed schedule, savings cap at 50-60% and don't stack with a reservation. Learn more →
- Storage tiering
- Placing data on the storage tier that matches its access frequency (Hot, Cool, Cold, Archive), automatically with lifecycle policies.
- Tagging
- Labeling cloud resources (application, environment…) to allocate costs. As few mandatory tags as possible: sometimes one is enough. Learn more →
- Unattached disk
- A disk no longer attached to any VM, but still billed. A cleaning classic.
- Unit economics
- Cloud cost relative to a business unit: per customer, per transaction, per order. It tells whether spend grows faster than the business. Learn more →
- Usage optimization
- Consuming only what is useful: rightsizing, Stop/Start, cleaning, storage tiering, log filtering.
- Waste
- The share of cloud spend that brings no value: unused, oversized or needlessly running resources. Learn more →
Guide · The basics
FinOps glossary: the essential terms, simply defined
FinOps has its own vocabulary. Here are the terms you will come across in a FinOps practice, each defined in one or two sentences, with a link to learn more.
FAQ
What is the difference between showback and chargeback?
Showback shows each team what it consumes; chargeback bills it back. The latter creates more accountability.
What is the difference between a Reserved Instance and a savings plan?
A Reserved Instance commits to a specific VM size; a savings plan commits to an hourly compute amount, which is more flexible. See our dedicated guide.
Why does FinOps use so many English terms?
Because they are the terms of the Azure and AWS consoles, of the tools and of the FinOps Foundation, which keeps conversations with teams simple.
First conversation
How much are you losing every month on a poorly optimized cloud?
SMEs: we are paid 20% of the annual savings achieved. Large enterprises: day-rate engagement. A first 30-minute conversation, no commitment.