Cloud spending has a tendency to grow faster than expected. What starts as a manageable monthly bill can quickly become a major line item, especially when teams provision resources without clear cost governance. FinOps — short for Financial Operations — is the practice of bringing financial accountability to cloud spending through collaboration between engineering, finance, and business teams.
What Is FinOps?
FinOps is not simply about cutting costs. It is a cultural practice and set of processes that help organisations make informed, data-driven decisions about cloud spending. The FinOps Foundation defines three phases:
- Inform — gain visibility into where cloud money is going, who is spending it, and how it maps to business value. This requires tagging, cost allocation, and reporting.
- Optimise — identify and act on opportunities to reduce waste, right-size resources, and leverage commitment-based discounts.
- Operate — embed cost awareness into engineering workflows, establish budgets and forecasts, and continuously improve your cost management practices.
Common Sources of Cloud Waste
Before optimising, you need to understand where waste typically occurs:
- Oversized instances — virtual machines and database instances provisioned with more CPU and memory than the workload requires. This is the most common source of waste across all cloud providers.
- Idle resources — development environments left running outside of business hours, unattached storage volumes, unused elastic IP addresses, and orphaned load balancers.
- Unoptimised storage — data stored on high-performance tiers that could be moved to cheaper storage classes, or old snapshots and backups that are no longer needed.
- Missing commitments — running stable, predictable workloads on on-demand pricing instead of using Reserved Instances, Savings Plans (AWS), or Reserved VM Instances (Azure).
- Data transfer costs — unexpected charges from cross-region or cross-availability-zone data transfer, especially in multi-cloud architectures.
Practical Cost Reduction Strategies
1. Implement Tagging and Cost Allocation
You cannot optimise what you cannot measure. Establish a consistent tagging strategy across all cloud resources with tags for team, project, environment, and cost centre. Use cloud-native tools (AWS Cost Explorer, Azure Cost Management) to create dashboards and reports broken down by these tags.
2. Right-Size Your Resources
Analyse actual CPU, memory, and I/O utilisation over a meaningful period (at least two weeks, ideally a month). Cloud providers offer built-in recommendations (AWS Compute Optimizer, Azure Advisor) that can identify instances running below 40% utilisation. Downsizing these can yield immediate savings of 30-50% on compute costs.
3. Schedule Non-Production Environments
Development, testing, and staging environments often do not need to run 24/7. Implementing automated start/stop schedules using Lambda functions, Azure Automation, or Terraform can reduce non-production compute costs by 65-70%.
4. Leverage Commitment-Based Discounts
For workloads with predictable, steady-state usage, commitment-based pricing offers substantial savings — typically 30-60% compared to on-demand. Start with one-year commitments for your baseline compute and storage, then consider three-year terms for truly stable workloads.
5. Use Spot and Preemptible Instances
Fault-tolerant workloads like batch processing, CI/CD pipelines, and development environments can run on spot instances (AWS) or spot VMs (Azure) at discounts of up to 90%. The trade-off is that these instances can be interrupted with short notice.
Building a FinOps Culture
Tools and techniques alone are not enough. Sustainable cost management requires organisational change:
- Assign cost ownership — make engineering teams responsible for the cloud costs of the services they build and operate. Provide them with dashboards that show their spending in real time.
- Set budgets and alerts — configure budget alerts at the team, project, and account level. Catch unexpected spending increases before they become significant.
- Include cost in architecture reviews — evaluate the cost implications of architectural decisions alongside performance, security, and reliability.
- Review regularly — hold monthly FinOps reviews where engineering and finance teams jointly examine spending trends, anomalies, and optimisation opportunities.
How ICTLAB Can Help
ICTLAB offers cloud cost optimisation services for Belgian organisations. We conduct thorough cloud cost assessments, implement tagging and governance frameworks, identify savings opportunities, and help your teams build the processes and skills needed for sustainable FinOps practice.