How I Help Companies Finally Save Thousands of Euros on Their Cloud Bill

Cloud computing offers flexibility, scalability, and speed. However, after migrating to the cloud, many organizations notice something unexpected. Instead of decreasing, their cloud bill begins to grow month after month. From my experience working with cloud infrastructure, this usually happens because companies move their systems to the cloud but continue thinking like they are still operating a traditional data center.

The reality is simple. There is no instant trick that will suddenly cut your cloud bill by half. Sustainable savings come from visibility, discipline, and smarter engineering decisions. This is where FinOps becomes essential. I often describe FinOps as a guide that helps organizations understand where their cloud money goes and how every technical decision affects the final bill.

1. Stop treating the cloud like a traditional data center

One of the most common mistakes I see during cloud migrations is the attempt to recreate on-premise infrastructure exactly as it was.

For example, a server that previously required:

• 16 CPUs
• 128 GB of memory

is often deployed in the cloud with the same configuration. But when I analyze the actual workload, I often discover that the application can run smoothly with:

• 8 CPUs
• 64 GB of memory

In other cases, the application may require more memory but less CPU power.

Cloud platforms allow infrastructure to scale easily. Instead of starting with oversized machines, I usually recommend beginning with smaller instances and adjusting resources based on real usage data.

I often explain it using a simple comparison. Many people dream of driving a Porsche. But if your daily travel is a short city commute, a compact car will serve the same purpose at a much lower cost. The cloud works the same way. Choose the resource that fits the need, not the one that simply looks powerful.

2. Gain full visibility into cloud spending

The first step in any FinOps initiative is visibility. Before optimizing anything, I always ask a simple question. Can we see the complete cloud spending?

In many organizations, the answer is surprisingly no. FinOps teams often lack the permissions required to view all subscriptions, accounts, or environments. Without full visibility, cost analysis becomes incomplete and sometimes misleading.

Even a read-only role at the billing level can reveal valuable insights. With the right access, it becomes easier to detect:

• unusual cost spikes
• idle or unused resources
• unexpected service consumption

Tools like Opslyft help centralize cost data so both engineering and finance teams can understand how cloud resources are being used. Once spending becomes visible, meaningful optimization can begin.

3. Stop chasing savings targets

Many companies measure FinOps success by asking teams to reduce cloud costs by a certain amount each year.

In the early stages, this approach can deliver impressive results. During the first year of optimization, organizations may reduce costs by:

• 20 to 25 percent

During the second year, improvements might reach around:

• 15 percent

However, once the infrastructure becomes well optimized, further reductions become harder to achieve. In later years, even a 3 percent improvement can be considered significant.

This is why I encourage organizations to shift their focus from pure cost reduction to FinOps maturity. Instead of only tracking euros saved, teams should monitor indicators such as:

• reservation or savings plan coverage
• percentage of unused resources
• automation of cost governance
• compliance with shutdown policies

These indicators provide a clearer view of operational efficiency.

4. Understand every service that generates cost

Another lesson I frequently share with teams is that cloud costs rarely come from a single service.

When organizations analyze their spending, they often focus only on compute resources such as virtual machines. However, the surrounding services also contribute significantly to the final bill.

For example, a single virtual machine environment may include:

• Storage disks
• Backup systems
• Replication services
• Monitoring tools
• Reserved capacity

Each component adds to the overall cost.

When reviewing cloud environments, I encourage teams to explore the entire service ecosystem rather than focusing on just one resource. Hidden costs often appear in supporting services that remain unnoticed during basic cost analysis.

5. Make FinOps a shared responsibility

Cloud cost optimization cannot succeed if it is handled by a single team.

From my experience, FinOps works best when multiple stakeholders collaborate. These usually include:

• Engineering teams
• Architects
• Finance departments
• Product owners
• Operations teams

For example, shutting down non-production environments during nights and weekends can significantly reduce costs. But implementing this policy requires agreement across several teams.

When everyone understands the benefits and participates in the decision, adoption becomes much easier. In many situations, simple discussions between teams reveal optimization opportunities that dashboards alone cannot identify.

6. Code decisions directly affect cloud costs

Many cloud expenses begin long before the first deployment. They start during the development phase.

Every architectural decision made by developers influences how much infrastructure an application will require.

For instance, two database queries might produce the same result. However, one query may consume far more CPU and memory resources than the other. At a small scale, the difference may seem minor. At a large scale, the impact can significantly increase infrastructure costs.

For this reason, I always recommend testing applications in environments that closely resemble production. Monitoring tools can help identify the queries or functions that consume the most resources.

Sometimes a single inefficient query can force the platform to run larger and more expensive infrastructure.

7. Adopt FinOps by design

One of the most effective strategies I use is integrating cost awareness directly into the deployment process.

This approach is often called FinOps Driven Deployment.

Instead of optimizing resources after deployment, organizations establish rules that guide infrastructure creation from the beginning.

For example:

• Non-production virtual machines automatically stop at night and on weekends
• Backup policies differ between production and development environments
• Predefined limits exist for resource sizes in certain environments

When such rules are automated, cost control becomes part of the engineering workflow.

Engineers can still request exceptions when needed, but the default behavior remains cost-efficient.

8. Build tools that support the entire organization

As FinOps practices mature, organizations benefit from shared tools and internal resources.

These may include:

• Cost analysis queries
• Automation scripts
• Documentation and optimization guidelines
• Internal playbooks for engineers

At AXA Group Operations, teams developed a shared FinOps toolkit that allowed different business units to standardize their cost management practices.

Platforms such as Opslyft support these initiatives by offering centralized dashboards and automation capabilities.

A simple rule I follow is this: if a manual task repeats more than three times, it should probably be automated.

Automation improves reliability and frees engineers to focus on strategic improvements rather than repetitive tasks.

9. Communicate in business language

One challenge I often observe is the communication gap between technical teams and financial stakeholders.

Cloud engineers frequently use terms like:

• Autoscaling
• Rightsizing
• Instance families
• Savings plans

While these concepts are important, they may not be meaningful to financial leaders.

When presenting cost insights to executives, I prefer focusing on outcomes such as:

• Staying within budget
• Improving cost forecasts
• Understanding cost per application
• Increasing operational efficiency

Clear dashboards and simple metrics help bridge the gap between engineering and finance.

Some organizations even introduce scoring systems that rate teams based on their resource efficiency. This approach encourages responsible cloud usage without overwhelming stakeholders with technical details.

Conclusion

From my experience working with cloud infrastructure, controlling cloud costs is not simply a technical exercise. It requires a cultural shift that combines visibility, collaboration, and disciplined engineering practices. When organizations begin to understand how their applications consume resources, they gain the ability to make smarter and more cost-effective decisions.

The companies that succeed with FinOps focus on long-term efficiency rather than quick cost cuts. They integrate cost awareness into development, deployment, and daily operations while using tools like Opslyft to maintain visibility and automation. In the end, the goal is not just to reduce the cloud bill but to ensure every euro spent in the cloud delivers real value to the business.