Figure
In the early days of FinOps, the playbook was simple: find the unattached disks, delete the idle snapshots, and buy some reserved Instances. It felt like magic. But as the low-hanging fruit disappears and the cloud bills continue to evolve, many organizations find themselves hitting a plateau.
Data from the recent State of FinOps1 report confirms what many of us have learned in the trenches:
FinOps is maturing, and the retail therapy phase of simply buying more tools is officially over.
Based on my experience navigating this transition across various scales of business, here is what actually moves the needle when the easy wins are gone.
FinOps is a Multi-Technology Practice
For too long, FinOps was put in a box labeled AWS/Azure/GCP spend. That view is now obsolete. The most successful organizations realize that FinOps is a multi-technology practice.
Value leakage doesn’t just happen in your cloud console, it happens in your SaaS sprawl, your licensing agreements, and your edge computing. If your FinOps practice isn’t looking at the integration between cloud costs and shared services (like Snowflake, Datadog, or GitHub), you aren’t doing FinOps, you’re just doing cloud cost management. True maturity means applying FinOps principles to every dollar spent on technology across the stack.
Lean Headcount, Deep Collaboration
There is a common misconception that a FinOps Team needs to be a massive department. In reality, the most effective FinOps functions are lean.
The goal isn’t to build a massive bureaucracy, but it’s to build a bridge. Success is found in deeper collaboration across three pillars:
- Engineering: Moving from “you spent too much” to “how can we architect for unit-economic efficiency?”
- Finance: Moving from “why is the bill variable?” to “how does this spend correlate to revenue?”
- Product: Building a culture where cost is a first-class non-functional requirement.
The FinOps team should act as a specialized platform function that empowers others, rather than a police force that slows everyone down.
Tooling is Not a Strategy
If your FinOps practice is broken, buying a new SaaS tool to visualize the mess is just organizational retail therapy. It feels good for a week, it gives you a few new dashboards, but it doesn’t fix the underlying culture or process issues.
Tools are force multipliers, but they multiply what you already have. If you have no governance, a tool will just show you how much money you’re losing in higher resolution. You cannot buy your way out of a fragmented architectural strategy or a lack of accountability. Fix the process with what you have first, then you can automate it with aditional tools.
Moving the Goalposts
In my experience, the moment a practice truly matures is when it stops talking about the total bill and starts talking about unit economics.
Reporting a 1M monthly bill to the C-suite is meaningless without context. If your bill went up by 20% but your customer transactions grew by 50%, you are actually winning. Sustaining momentum after the easy wins requires moving toward cost per transaction or cost per user. This forces the conversation to go beyond the surface and into true business value.
AI is the New Multi-Technology Challenge
The latest data shows a massive spike in AI-related spending. This is the ultimate test for a FinOps practice. You cannot manage AI costs (GPUs, tokens, and vector DBs) with a standard cost mindset.
I’ve found that the best way to handle this is to treat AI as a self-funding R&D project. Use the savings from your infrastructure optimizations to directly fund AI experimentation. This keeps the organization agile without ballooning the budget, proving that FinOps is a growth engine, not just a savings department.
Fixing the Data, Not the Dashboard
One reason many tools fail is that every vendor sends data in a different language. Before buying more software, I suggest looking at normalizing your data into a common schema. Then your team can stop spending all time fixing broken data, and start focusing on giving leadership advice needed to make better business decisions. When your SaaS, Cloud, and Marketplace spend all speak the same language, you can manage a massive enterprise footprint without needing a massive headcount or new tools.
Sustaining Momentum Beyond the Surface Level
It is easy to get the CEO’s attention when you save 100k in a month by right-sizing. It is much harder to keep that momentum when the next 100k requires re-architecting a legacy application or changing how the organization handles procurement.
To move beyond surface-level gains, you must go deeper in two directions:
- Technical Depth: This means looking at serverless optimization, container rightsizing, and deep architectural efficiency.
- Strategic Vendor Management: Mature FinOps involves sophisticated, long-term strategic relationship management. Negotiate custom terms that align with your growth patterns rather than just clicking “accept” on a standard enterprise agreement.
The Bottom Line
There is always room for improvement, but the optimization of today and beyond looks different than it did five years ago. It’s less about cleaning up messes and more about intentional engineering and strategic alignment.
Stop looking for the magic button in a software UI. Start building the cross-functional relationships and technical depth required to make cost efficiency a part of your company’s DNA. That is where the real value lives.