What the State of FinOps 2026 Report Actually Shows About Cloud Waste
The FinOps Foundation's own 2026 survey shows waste reduction is still the top priority, but diminishing returns are already setting in for mature teams.

- The FinOps Foundation's 6th annual State of FinOps survey, covering 1,192 respondents and more than $83 billion in annual cloud spend, found workload optimization and waste reduction is the single top current priority, at 42.48 percent of respondents, but that figure drops to 24.96 percent when the same respondents are asked about priorities for the next 12 months.
- Mature practitioners report diminishing returns from classic optimization work. One respondent described reaching 97 percent optimization in their own cost optimization tooling, with the remaining 3 percent intentionally left unactioned for business reasons, not because nobody looked.
- FinOps has expanded well past cloud infrastructure itself: 90 percent of respondents now manage or plan to manage SaaS spend, 64 percent manage licensing, 57 percent manage private cloud, and 48 percent manage data center costs, each a scope most teams were not tracking with the same discipline a few years ago.
- Many organizations report being asked to self-fund AI investment through optimization savings, which means classic cloud waste reduction is now competing for the same budget headroom that a growing AI spend line is trying to claim.
What the State of FinOps 2026 report actually shows about cloud waste is a more specific and more useful story than a flat percentage figure. The FinOps Foundation’s own sixth annual survey, covering 1,192 respondents and more than 83 billion dollars in annual cloud spend, does not report that most cloud bills run some fixed percentage higher than they need to. It reports something sharper: classic waste reduction is still the single most common current priority, but the teams doing it longest are already running into diminishing returns, and the real cost-management work has moved somewhere else.
What the survey actually measured
The State of FinOps 2026 report is the sixth consecutive annual edition of a survey the FinOps Foundation has run since 2020. This year’s edition reflects a notable shift in framing. The Foundation describes FinOps as having “accelerated into a proactive, technology-wide discipline,” with scope expanding well beyond cloud and AI dominating the forward-looking agenda. The organization even updated its own mission statement this year. It moved from “Advancing the People who manage the Value of Cloud” to “Advancing the People who manage the Value of Technology,” a change the report frames as catching up to where practitioners had already moved.
That framing matters because it directly contradicts the assumption behind a flat waste percentage applied to “most” cloud bills. A survey measuring a discipline that has outgrown cloud infrastructure as its sole focus is not describing a static waste problem sitting unaddressed in everyone’s bill. It is describing a moving target. That target keeps expanding into new categories of spend as teams finish working through the old ones.
The priority that is both first and already fading
The clearest number in the report is also the one that most directly undercuts a fixed-waste-percentage claim. Workload optimization and waste reduction is the single top current priority, named by 42.48 percent of respondents. That is genuinely the most common answer today. But the same respondents, asked about priorities for the next 12 months, drop that figure to 24.96 percent. Waste reduction is not a problem these teams see as permanently sized. It is a priority they expect to need less of. The reason is specific: they are running out of the easy version of it.
The report’s own qualitative detail explains why directly. One practitioner is quoted describing having “hit the big rocks of waste” and now facing “a high volume of smaller opportunities that require more effort to capture.” Another reported reaching 97 percent optimization in their own cost optimization tooling. The remaining 3 percent was intentionally left unactioned for business reasons, not because nobody had looked for it. Both quotes describe a team that has already done the obvious work. Neither describes a backlog of easy, unaddressed 25 to 40 percent waste sitting in a bill nobody has examined.
That distinction, between waste nobody has looked for and waste a mature team has already found and deliberately decided not to chase further, is the entire difference between a useful cost-management program and a number borrowed from an unrelated source. The first is a real, actionable finding. The second is a claim nobody engaged can actually trace back to anything a FinOps team would recognize as its own data.
Where the actual spend has moved
If classic cloud waste reduction is hitting diminishing returns for mature teams, the report is specific about where the remaining cost-management challenge has actually gone. Ninety percent of respondents now manage or plan to manage SaaS spend in the coming year, an increase of 25 percentage points. Sixty-four percent manage licensing, up 15 points. Fifty-seven percent manage private cloud, up 18 points. Forty-eight percent manage data center costs, up 12 points. An emerging 28 percent are even including labor costs in what they track. None of these categories existed inside a traditional “cloud bill” at all, which is exactly why a waste percentage framed around cloud infrastructure alone misses most of where spend has actually migrated.
Supporting this expansion is growing adoption of the FinOps Open Cost and Usage Specification, known as FOCUS, an open specification the FinOps Foundation maintains specifically to normalize billing data across cloud, SaaS, data center, and other technology vendors. The specification exists because a team trying to manage spend across all of these categories at once needs a consistent data format to compare them, something no single vendor’s native billing export was ever built to provide on its own.
Why AI spend is squeezing the same budget from the other direction
The report adds a pressure that makes the diminishing-returns story more urgent rather than less. Many organizations report being asked to self-fund AI investment through optimization savings, creating direct pressure to find efficiency gains specifically so they can be redirected toward AI initiatives. This creates what the report calls a “squeeze more from existing footprint to create space for AI spend” dynamic, one that is accelerating optimization urgency even as the traditional, easy waste opportunities keep shrinking. A team working this dynamic is not optimizing for its own sake anymore. It is optimizing because a separate, growing line item in the same budget is depending on the savings showing up.
Ninety-eight percent of respondents now manage AI spend in some form, up from just 31 percent two years ago. That is a genuinely fast shift, and it means the optimization work a FinOps team does today is increasingly being measured against a new, competing claim on the same savings, not against a static waste target that stays the same size year over year. AI investment itself remains strongest in cloud, but the report notes it is also increasing in SaaS, data center, and private cloud spend, which means even the newest category of cost pressure is already spreading across the same broader scope the rest of the discipline has been expanding into.
A concrete example: the optimized bill with an unoptimized shadow
Picture a payments platform whose engineering team has spent the past two years rightsizing EC2 instances, moving idle workloads to spot capacity, and negotiating committed-use discounts on its core processing infrastructure. By any classic measure, that cloud bill is genuinely well optimized, closer to the practitioner who reached 97 percent and stopped than to a team that has never looked.
Meanwhile, the company’s support desk pays for a help desk platform. Its compliance team pays separately for a GRC tool. Its sales team runs its own CRM add-ons. Its data team has started running meaningful workloads through a handful of AI APIs billed per token. None of that spend shows up on the cloud bill the engineering team optimized so carefully. All of it is real, recurring, and growing. None of it has been through anything resembling the discipline applied to the infrastructure side. The platform is not failing to optimize. It is optimizing the one category of spend that the report’s own data says is closest to running out of easy wins, while the categories it describes as expanding fastest sit completely outside anyone’s view.
What this actually means for a payments platform
A payments platform’s instinct, on hearing “cloud cost optimization,” is usually to picture rightsizing compute instances and killing idle resources, the classic version of waste reduction the report describes as already hitting its limits for experienced teams. That instinct is not wrong. It is simply increasingly incomplete, because the categories the report shows growing fastest, SaaS, licensing, AI spend, and data center costs where applicable, are exactly the ones a payments platform’s existing cost discipline is least likely to already cover with the same rigor.
A platform that has spent real effort rightsizing its core processing infrastructure, while the SaaS tools its support, sales, and compliance teams have individually subscribed to over several years sit completely untracked, has correctly optimized the shrinking part of its spend. It has left the growing part completely untouched. That is precisely the gap the report’s own scope-expansion numbers describe. It is a gap that classic cloud-focused cost optimization work was never built to catch in the first place.
What actually works at this stage
The report’s own framing points toward where the real leverage sits now: scope expansion into SaaS, licensing, and other technology categories; governance and policy applied consistently across all of them; and consistent, unified data, which is exactly what the FOCUS specification is built to provide across vendors that would otherwise each report costs in their own incompatible format. None of this replaces the classic rightsizing and waste-elimination work. It sits alongside it, aimed at the much larger share of total technology spend that a cloud-only optimization effort was never positioned to see.
Building that broader discipline, rather than continuing to optimize the one category of spend that is already closest to its limit, is exactly the work behind Midcore’s cloud cost optimization practice. Where the underlying question is how the platform’s architecture itself should be designed to make cost visible and controllable across every category from the start, cloud strategy and architecture is where that gets built in rather than retrofitted. And because tracking spend across multiple cloud providers, SaaS tools, and vendors inevitably touches the same systems that hold sensitive data, multi-cloud security and compliance work is where cost visibility and security posture get kept consistent with each other rather than managed as two unrelated projects.
Frequently Asked Questions
Do most cloud bills actually run 25 to 40 percent higher than they need to?
The FinOps Foundation’s own current State of FinOps report does not support that specific figure. It shows workload optimization and waste reduction is the top current priority for FinOps teams, at 42.48 percent of respondents, but it also reports that mature practitioners are hitting diminishing returns, having already captured the larger, easier waste and now facing smaller, harder-to-find opportunities. A blanket 25 to 40 percent waste figure does not appear in the current report’s own data.
What does the State of FinOps 2026 report actually measure?
It is the FinOps Foundation’s 6th annual survey of the global FinOps community, covering 1,192 respondents representing more than 83 billion dollars in annual cloud spend. It tracks priorities, technology categories under management, tooling adoption, and the growing role of AI in both cost management and FinOps work itself.
Where has cloud cost management actually expanded beyond classic infrastructure?
The report found that 90 percent of respondents now manage or plan to manage SaaS spend, 64 percent manage licensing, 57 percent manage private cloud, and 48 percent manage data center costs. An emerging 28 percent are even including labor costs. This is a broad technology value discipline now, not a narrower practice focused only on rightsizing cloud infrastructure.
Sources: FinOps Foundation, State of FinOps 2026, the 6th annual survey, and What is FOCUS?, the FinOps Open Cost and Usage Specification.