By Enterprise cloud bills are climbing in nearly every sector, yet the capital that rising spend was supposed to generate isn’t materializing. According to Azul’s 2026 CFO Cloud Cost Optimization Report — drawn from a Censuswide survey of 300 U.S. CFOs and senior finance leaders — the average organization estimates that roughly 23% of its total cloud spend is wasted. Not trimmed. Not deferred. Wasted.
The pattern isn’t new, but the stakes have shifted. For years, cloud overruns were absorbed as a cost of growth and kept largely within IT. That era is over. Two-thirds of finance leaders in Azul‘s survey say cloud cost has escalated to a board-level concern. The CFO, once insulated from infrastructure decisions, is now expected to govern them.
The pressure isn’t arriving in isolation. Fifty-six percent of CFOs in the report identify AI and automation investment as their top financial priority for 2026. The problem is that AI budgets have to come from somewhere. For most enterprises, the only viable source is the cloud spend already on the books. Cloud optimization is no longer a cost-cutting project — it’s the prerequisite for AI funding.
But here’s the catch: the tools most enterprises are using to manage cloud costs are visibility tools. They show what is being spent and where. They don’t change how efficiently applications consume the infrastructure they’re running on. According to Azul’s report, nearly half of organizations use AI-powered cloud spend analytics and native cloud provider tools — but only 16% use Java runtime optimization or JVM tuning, the lever that directly reduces compute consumption at the application layer.
That gap — between identifying waste and actually eliminating it — is the core argument Azul’s CFO and COO Peter Maloney makes in this conversation. His thesis: most enterprises are stuck in stage one FinOps. Getting to stage two, and staying competitive through the AI wave, requires going deeper than the dashboard.
The Guest: Peter Maloney, Chief Financial Officer and Chief Operating Officer at Azul
Key Takeaways
- 88% of CFOs report rising cloud spend; average estimated waste sits at 23% of total cloud budget — now a board-level governance issue, not an IT problem
- Most FinOps programs are stuck at stage one: consumption visibility and commercial negotiation. Stage two — application-layer optimization via JVM tuning — is where real cost reduction happens
- Only 16% of enterprises use Java runtime optimization or JVM tuning, despite it being the lever that directly reduces compute consumption, not just tracks it
- 43% of CFOs say AI is making cloud costs harder to measure; cloud discipline now determines how much capital is available to fund AI transformation
- Cloud cost as a percent of revenue, EBITDA impact, and unit economics are increasingly shaping company valuations in both public markets and private investment
***
In this exclusive interview with Swapnil Bhartiya at TFiR, Peter Maloney, Chief Financial Officer and Chief Operating Officer at Azul, discusses the findings of Azul’s 2026 CFO Cloud Cost Optimization Report, the two-stage FinOps maturity model, Java runtime optimization as a structural cost lever, the compounding dynamics of AI spend, and what CFOs should be doing right now to move beyond visibility into genuine application-layer efficiency.
Cloud Waste as a Hidden Tax on Profitability
Azul’s 2026 CFO Cloud Cost Optimization Report — based on 300 U.S. CFOs and senior finance leaders — establishes a striking baseline: 88% of respondents report rising cloud spend, yet average estimated waste sits at roughly 23% of total cloud budget. For Maloney, the framing that makes this actionable for finance leaders is the concept of a structural hidden tax.
Q: Why has cloud waste at this scale become a norm rather than an exception?
Peter Maloney: “You look at what most CFOs are trying to do today. They’re trying to help their companies scale profitably. And there’s a lot of things that go into profitability — it’s the gross margins, it’s your cost structure. And just like other areas within your cost structure, cloud waste is a hidden tax. And it’s not one thing that causes it and it’s not one thing that solves it. In order to go and solve identifying the hidden tax and take action, it really comes down to a group effort with what we call FinOps. And FinOps is starting to take on in many, many companies that are adopting cloud and trying to become more mature with cloud. And so CFOs are now starting to see this because that hidden tax is taping up the ability to use that capital or that investment in other things, just like AI is starting to take off and we all want to be able to invest in AI to improve our businesses. And so the tax is more relevant and working with other organizations or other parts of your business is really important to identify and sort of solve the hidden tax.”
The CFO’s New Role in Cloud Governance
Cloud economics becoming a board-level issue represents a meaningful organizational shift. The CFO is no longer a passive consumer of cloud cost reports — they’re expected to drive governance outcomes. Maloney connects this to the need for common metrics and cross-functional accountability rather than siloed activity data.
Q: How are CFOs working across functions to bring predictability and discipline to cloud economics without stifling innovation?
Peter Maloney: “I think it starts with a cultural change or a cultural common desire to have a common outcome. And a lot of it — there’s a lot of data that’s available to CFOs and other people within IT, product engineering, et cetera. But it’s using common metrics and having a common goal as to what you’re trying to accomplish within your cloud investment. As I mentioned before, only about 18% of companies that we surveyed have FinOps centers of excellence, and so it is evolving. But I think that the CFO working with IT leaders, product leaders, and engineering leaders, and having common language, common metrics — not just activity metrics, consumption, commercial metrics, but output metrics — what defines success and working together to achieve what defines success, it allows the investment to be optimized and potentially make room for other investment.”
Understanding Cloud Consumption: Cost of Goods vs. R&D
Cloud spend doesn’t live in a single line item. Maloney breaks it down into its two primary buckets — cost of goods sold for product delivery, and engineering and R&D cycles — and explains the two-stage FinOps model that corresponds to each.
Q: Do CFOs and finance leaders actually understand where their cloud is being consumed and what it’s being used for?
Peter Maloney: “When you look at a P&L view of where cloud spend is going, it’s going into two areas, primarily. One is your cost of goods sold to go supply your product or your service to your customer. And number two, in your development cycles, in engineering and R&D. And so there’s a lot of tools out there and a lot of data to measure the activity and to help you manage. And there’s actually tools with the big cloud providers as well as individual software companies that will help you to manage consumption and automate things. There’s also tools out there to help you with the commercial side of things. What’s been happening now is the adoption of those tools. Stage two is more the tools that will help you at the application layer and actually help performance improve. That’s one of the things that Azul does for our customers. And so I look at the adoption of FinOps in two stages: identification, working on consumption and commercials, and then secondly, actually working on the efficiency at the application layer.”
From Data Availability to Output-Oriented ROI
Cloud pricing complexity is frequently cited as a barrier to optimization. Maloney reframes the problem: it isn’t a data shortage, it’s a failure to connect data to business outcomes. The insight CFOs need isn’t more dashboards — it’s a clear definition of what success looks like before spending begins.
Q: What insights do CFOs actually need but often don’t have when it comes to cloud usage, and what is Azul doing to help?
Peter Maloney: “Azul definitely helps. And like I said, Azul helps at the application layer. I think it’s not really a data issue. There’s a lot of data that’s available. I think the gap is taking that data and applying it not just to consumption, but to outputs. You think about from a CFO’s perspective, it’s not just about cost management, it’s optimizing ROI or optimizing outcomes. So as you’re investing in the cloud, having a clear definition of success — what are you investing in? Are you investing in product? Are you investing in customer-related activities? And knowing what true success looks like helps you use the data that you have to achieve that success.”
Cloud Efficiency, Software Performance, and Company Valuation
Cloud cost optimization is increasingly a valuation story, not just an operations one. Maloney connects gross margin improvement, EBITDA, and the ability to scale without proportional infrastructure growth to how companies are being evaluated in both public markets and private investment contexts.
Q: Is software efficiency becoming a real driver of company valuation, and how does the calculus change when you can run leaner, faster workloads?
Peter Maloney: “I think what’s happening — and you see it in the public capital markets, and it’s definitely happening in private company valuations — the amount of cloud spend is becoming important for valuations. It’s reflected in gross margins, it’s reflected in EBITDA, and the ability to scale without adding a lot of cloud resources is really important. The other thing, and I think this is the most important power of optimizing your cloud spend, is every dollar that you free up from spending on cloud, you can use to invest in your business for either growth or let it go to the bottom line. And we all know that what’s coming is continued investment in AI. AI is going to be a game changer for people’s growth models and their cost models. And so freeing up dollars to be able to invest in AI is very, very important.”
The Hidden AI Tax and Lessons from the Cloud Era
Azul’s report finds that 56% of CFOs name AI and automation as their top financial priority for 2026, while 43% admit AI is making cloud costs harder to measure. The risk: enterprises are repeating the same unchecked adoption patterns that created cloud waste in the first place, now with token costs and AI inference spend added to the stack.
Q: Is there a hidden AI tax that organizations are walking into, and what lessons from the cloud era apply?
Peter Maloney: “I agree. I think making sure that you have your cloud layer and your cloud spend under control and running efficiently with accurate and clearly defined outcomes is important. AI can just compound the problem. I think what we’re going to find is the companies that are focused on their cloud cost optimization will have an easier time of adopting and getting adoption from their employees, their AI tools. You’ll also see that they’ll free up dollars to be able to more quickly invest in AI. And so I think they’re completely linked. I actually think just like there’s a whole industry around trying to help companies optimize their cloud costs, I think the same thing is going to happen with AI. And so you have to start with your cloud costs, make sure you understand them, make sure that you’re spending for real outcomes, and that will help you to be able to implement and get traction with AI.”
Application-Layer Optimization: Why Stage Two Is Where Real Savings Live
The distinction between stage one and stage two FinOps is the crux of Azul’s thesis. Automation and consumption visibility tools are necessary but insufficient. Companies tuning their Java runtimes and modernizing their JVM environments are the ones achieving the deepest, most durable cost reductions — because they’re changing how applications consume infrastructure, not just tracking the bill.
Q: Are enterprises looking at quick-fix band-aids or long-term architectural solutions for cloud cost?
Peter Maloney: “I’d say, as I said earlier, I think there’s a couple of stages of getting ready and preparing yourself and doing better at managing your cloud costs and optimizing. One is I put it in the consumption and commercial bucket and some automation tools. There’s a lot of automation tools out there. The cloud vendors provide them. There are spot solutions that are very good that are used, but that really doesn’t give you anything more than just visibility and automation. What you really need is go to the application layer. Companies that are focusing on tuning their Java runtimes and modernizing their applications and their Java environment are the ones that are going to get the biggest return. And by doing that, it’s more pervasive and you’re actually investing in something that takes action. If you think about a lot of the tools in phase one of FinOps that you’re using, it’s for visibility, governance, accountability, where you can really take action is something that is going to help you to make your application layer run more efficiently. And that is pervasive and material. And that is actually what Azul is very, very good at. We have many customers using our solutions to achieve that level of FinOps excellence.”
Where the JVM and Java Runtime Fit for CFOs
Azul is, at its core, a Java platform company — its flagship product, Azul Prime, is a high-performance JVM with a pauseless C4 garbage collector, a Falcon JIT compiler, and ReadyNow warmup elimination technology. For a CFO, this may sound like deep infrastructure plumbing. Maloney explains how it maps to executive outcomes.
Q: Where does the runtime layer fit into the cloud cost conversation, and is JVM optimization something a CFO should even think about?
Peter Maloney: “From a CFO’s role, and you try to learn more and more about technology and how applications run, you’re not going to have all the answers. You’re going to have to partner with your CIO and your product people, engineering people to make sure you’re making the right decisions here. I think that what we’ve been able to do as a company is do exactly that. We work with our customers — CFOs, CIOs, product people, engineers — and try to help them go from the next level of just managing consumption and commercials to actually changing how their applications operate. And so as a CFO, what I would do is I would work with the other executive partners and functional departments to go find the optimal solution to go beyond just visibility and automation, to go to something that actually changes the performance.”
The Conversation CFOs and CTOs Are Not Having — But Should Be
Maloney offers direct, prescriptive advice for the CFO preparing for a joint conversation with engineering and technology leadership about cloud cost. The framework is: accountability first, outcome clarity second, application-layer action third.
Q: What is the conversation a CFO and their CTO team are probably not having about cloud cost that they should be?
Peter Maloney: “I think they probably have started to use tools to get visibility and identify what’s happening, but I would say number one, there needs to be accountability. And we here at Azul, we literally take our cloud bill and we are able to identify down to every single project who’s responsible, and we manage it very actively — we’re looking at reporting weekly, et cetera. So it’s important to get down to full knowledge and visibility and accountability. I think secondly though, and I mentioned it earlier, is understanding what are the big outcomes that you want. Are they customer-related outcomes? Are they product-related outcomes? Are they development-related outcomes? And if you start identifying that and you know where you’re spending, you can take quick actions on the consumption side or with partners on the commercial side to negotiate actually better deals. That’s stage one. But stage two, that’s really identifying where are you consuming the cloud, what products, what applications in your business? And you can focus on those applications and actually implement something like Azul that will make those applications much more productive. And so stage one is identifying and understanding where you’re spending, what you expect your outcomes to be, and then looking at tools that can help you actually solve and take action rather than just identifying. I think, and you can see it in the results — things are happening fast here. And I think, as you said, AI is starting to drive people to have to think about cloud even more or maybe even before they wanted to. And so there’s sort of a learning curve. CFOs are always going to go to the concept of scaling profitably, scaling their businesses profitably. So when something becomes material and it can make a difference for the ability to scale, CFOs, their attention will go to it right away. And so until recently, we haven’t seen that. Now we’re seeing it and it’s happening very fast. And I think that model of stage one versus stage two is important because companies and CFOs are going to have to go from stage ones to stage two fast to remain competitive and have the right business model for the future.”
What CFOs Still Wish They Had: Predictability and AI Cost Clarity
Even at Azul — a company with mature internal FinOps practices — AI cost management remains an open problem. Maloney describes Azul’s own internal model (visibility, accountability, predictability, zero surprises) and where the AI cost layer creates new complexity that most enterprises haven’t yet solved.
Q: What insights do CFOs wish they had — or wish more people used — when it comes to cloud and AI usage?
Peter Maloney: “We’re going through a process right now ourselves. I think we’ve done a great job on cloud cost optimization and we use the model of making sure we have great visibility, we have accountability, and now we have predictability. We don’t have any surprises. And we use a number of tools from our cloud providers to do that, as well as one or two other point solutions that help us on the consumption side and the commercial side. And we obviously, being at Azul, we use Azul products. I think that with AI, as you pointed out, it’s going to add a whole other layer of complexity. And I think as AI is adopted, I think companies are going to focus on adoption first and return second. And so figuring out — once again — what are your expectations and the outputs that you want to see from your investment, not just in cloud, but AI, and then figuring out the KPIs. I don’t think most companies have done that. So I think the effort to identify success, figure how to track it, and then actually hold people accountable is going to be one of the key things that a CFO can help their businesses be successful, not only with cloud, but with AI.”
Q: Do you need to be precise about defining what success means, given how differently it can be interpreted across teams?
Peter Maloney: “That’s right. Success on investment is not always measured in just a number, a financial outcome. It can be far reaching across the whole organization, and it can have a longer term impact if it’s changing your culture or your ability to serve customers.”
