In a data centre, spending was decided once, when servers were bought. In the cloud, every engineer spends money every time they launch, resize or forget a resource. FinOps is how organisations keep that freedom and still spend well.
The FinOps Foundation defines it as an operational framework and cultural practice which maximizes the business value of cloud and technology, enables timely data-driven decision making, and creates financial accountability through collaboration between engineering, finance, and business teams.
flowchart LR ENG["Engineering: speed"] --- F["FinOps"] FIN["Finance: predictability"] --- F BIZ["Business: value"] --- F F --> D["Better decisions: cost, speed, quality"] %% step ENG: Engineers want to ship fast and choose the right architecture without waiting for approvals. %% step FIN: Finance needs costs it can forecast, allocate and explain. %% step BIZ: Product and leadership want to know what each euro or dollar of cloud spend returns. %% step F: FinOps gives all three the same data and a shared way of deciding. %% step D: The goal is not the lowest bill but the best trade-off between cost, speed and quality.
The six principles
| Principle | In practice |
|---|---|
| Teams need to collaborate | Engineers, finance and product look at the same numbers together, often weekly |
| Business value drives technology decisions | A bill that grows with revenue can be good news; measure cost per unit of value |
| Everyone takes ownership for their technology usage | The team that runs a service sees and manages its cost |
| FinOps data should be accessible, timely, and accurate | Daily or hourly cost data, not a surprise invoice a month later |
| FinOps should be enabled centrally | A central team runs tooling, rate negotiations and best practice; teams act on it |
| Take advantage of the variable cost model of the cloud | Scale down at night, use spot, buy commitments only for steady usage |
What FinOps is not
- Not just cost cutting. Spending more on a product that is growing fast is a valid FinOps decision.
- Not only finance's job. Most savings come from engineering decisions: architecture, sizing, clean-up.
- Not a one-off project. It is a loop that runs as long as you use the cloud.
- Not only public cloud. The 2025 framework talks about technology and uses scopes such as SaaS, licences, data centre and AI spend.
A tiny example
The shop team sees in Monday's report that its test environment costs as much as production. Nobody planned it: test servers simply run all weekend. Engineering adds a schedule, finance updates the forecast, and product decides the saved money funds a load-testing tool. That conversation — data, owner, decision — is FinOps.
Traps: starting with a tool before agreeing who owns which costs; and treating "save money" as the only metric, which teaches teams to hide spend instead of explaining it.