Cloud migration was supposed to reduce IT costs. Yet, over 60% of companies report that their cloud service bills are higher than anticipated, and 30% admit they waste over 30% of their cloud budget on unused resources. If your organisation spends 5 million zloty annually on AWS, Azure, or Google Cloud, there’s a high probability that at least 1.5 million PLN is going down the drain. Unoptimised instances, forgotten resources, lack of governance—these are pain points for practically every company operating in the cloud.
The solution isn’t new tools or additional dashboards. The key is FinOps—a cultural and operational practice that combines technology, business, and finance within a shared responsibility framework for cloud costs. Without appropriate team competencies, however, FinOps will remain merely a fashionable buzzword. In this article, you’ll learn exactly what FinOps is, which roles and skills are essential in an organisation, which tools support FinOps practice, and how much you can realistically save through training that prepares your team for effective cloud cost management.
Quick navigation
- What is FinOps and why is it not just a task for the CFO?
- The FinOps Framework: Inform, Optimise, Operate phases
- Why do cloud costs spiral out of control?
- What FinOps competencies does your team need?
- FinOps Roles: FinOps Practitioner, Cloud Economist, Cloud Architect
- FinOps Tools: AWS Cost Explorer, Azure Cost Management, Kubecost
- Case studies: How much can you save? Examples from the Polish market
- FinOps Certified Practitioner certification—is it worth it?
- How does EITT train teams in FinOps?
- FAQ: answers to the most common questions
What is FinOps and why is it not just a task for the CFO?
FinOps (Financial Operations) is a discipline and cultural practice for managing cloud costs that brings together financial, engineering, and business teams within a shared responsibility framework for optimising cloud expenditure. Unlike traditional IT, where budgets were static and predictable (CAPEX: purchasing servers, licences), the cloud model is based on variable operational costs (OPEX)—you pay for what you actually consume, in real time.
This is a fundamental shift that brings enormous flexibility but also risk: without conscious management, costs grow exponentially, and their sources become opaque.
The FinOps Framework: Three phases
The FinOps Foundation (part of the Linux Foundation) developed a framework based on three cyclical phases:
1. Inform
The phase of increasing cost visibility and awareness:
- What we do: Aggregate cost data from various sources (AWS, Azure, GCP, Kubernetes), assign costs to teams, projects, environments
- Key questions: How much are we spending? Who is responsible? What are the trends?
- Challenges: Multi-cloud chaos (multi-cloud makes comparing costs between providers like comparing apples and oranges), lack of resource tagging, diffused responsibility
- Metrics: Cost per service, cost per team, cost per customer, cost trend analysis
2. Optimise
The phase of identifying and implementing optimisation actions:
- What we do: Look for waste, rightsize instances, negotiate Commitment Discounts (Reserved Instances, Savings Plans), automate removal of unused resources
- Key questions: Where are we wasting money? Which resources are over-dimensioned? What can we shut down at weekends?
- Challenges: Resistance from dev/ops teams (“this slows down our work”), lack of ROI metrics for optimisation, fear of downtime
- Actions: Reserved Instances/Savings Plans (20-70% savings), rightsizing (15-30% savings), spot instances for flexible workloads (50-90% savings), lifecycle policies for storage
3. Operate
The phase of implementing FinOps culture as part of daily operations:
- What we do: Create processes, policies, automation and governance, integrate costs with architectural and business decisions
- Key questions: How do we build a culture of cost awareness? How do we make teams take responsibility for costs? How do we measure unit economics efficiency?
- Challenges: Cultural change (dev doesn’t feel responsible for costs), lack of incentives for optimisation, cost review processes before deployment
- Actions: Cost anomaly alerts, budget controls, showback/chargeback, cost governance policies, FinOps KPIs in team OKRs
FinOps is a cycle of continuous improvement: Inform → Optimise → Operate → Inform… Without team competencies, you’ll remain locked in the Inform phase—you know how much you’re spending, but you don’t know how to change it.
Why do cloud costs spiral out of control?
Before we move to solutions, let’s understand the root causes of the problem. Here are typical cost traps that companies fall into:
1. Lack of visibility and accountability
In traditional IT, purchasing a server required a request, budget, approval. In the cloud, a developer can launch 50 instances with one click in the console. If there’s no system of tagging, cost allocation, and accountability, nobody knows who’s responsible for what. Result: resources grow, and nobody owns up.
2. Over-provisioning out of habit
Teams accustomed to on-premise infrastructure apply a “better to have a buffer” mentality. In the cloud world, this means paying for 30-50% of unused resources. An instance with 32 GB RAM when the application uses 8 GB? You’re paying for wasted capacity.
3. Orphaned resources
A developer created a test environment, finished the project, forgot to shut it down. Instances, EBS volumes, Load Balancers, elastic IPs—everything is still running and generating costs. Studies show that 20-35% of cloud resources are orphaned resources.
4. Lack of rightsizing
Lift-and-shift migration: servers transferred 1:1 from on-premise to cloud, without optimisation for the new model. Result: instances with parameters from 2015, inadequate for actual application needs.
5. Non-optimal storage tiers
Data that should reside in S3 Glacier archive (cheap storage for rarely used data) sits in S3 Standard (10x more expensive). Lack of lifecycle policies moving data between tiers.
6. Multi-cloud without unified governance
AWS for compute, Azure for databases, GCP for machine learning. Each platform has its own cost models, tools, practices. Without unified governance, multi-cloud is costly chaos.
7. Lack of commitment-based discount optimisation
You’re paying on-demand rates (most expensive), despite 70% of workloads being stable and predictable. Reserved Instances or Savings Plans offer 30-70% savings, but require analysis and decisions.
All these problems share a common cause: lack of team competencies in FinOps. This is why training is crucial—it’s not an investment in a certificate on the wall, but in measurable savings.
What FinOps competencies does your team need?
Effective FinOps practice requires competencies distributed across the entire organisation. Here are competency maps by role:
1. Executive Leadership (C-Level)
What they need:
- Understanding of the cloud economic model: CAPEX vs OPEX, unit economics, cloud ROI
- Awareness of FinOps impact on P&L (profit and loss statement)
- Governance principles: who decides on cloud budgets, how do we allocate costs?
- Strategic metrics: cost per customer, gross margin, burn rate
Why it’s important: The CFO must understand that cloud cost management isn’t just “cutting costs”, but optimising value for money. The CTO must know that architectural decisions have a direct impact on P&L.
What training changes: After training, leadership can:
- Interpret the FinOps dashboard during board meetings
- Make conscious trade-off decisions: performance vs cost
- Support a culture of cost awareness (e.g., incorporating cost KPIs into team OKRs)
2. FinOps Practitioner (new role)
Who is this? A person/team responsible for coordinating FinOps practices in the organisation. Connects finance, engineering, and business. Often comes from a financial background but must understand technology.
Key skills:
- Knowledge of cloud cost models: compute, storage, networking, data transfer
- Ability to work with cost management tools (AWS Cost Explorer, Azure Cost Management, CloudHealth, Cloudability)
- Creating cost allocation models (tagging strategy, account hierarchies, chargeback/showback)
- Cost trend analysis, forecasting
- Soft skills: communication with technical and financial teams, stakeholder management
Certification: FinOps Certified Practitioner (FinOps Foundation)—the most recognised certification in the industry. Covers FinOps framework, best practices, case studies.
3. Cloud Architects / Cloud Engineers
What they need:
- Cost-aware architecture: how to design systems not only for performance/security, but also cost efficiency
- Knowledge of pricing models: on-demand, Reserved Instances, Savings Plans, Spot Instances
- Rightsizing ability: how to match instances to actual CPU/RAM consumption
- Storage optimisation: lifecycle policies, tiering strategies
- Serverless economics: when are Lambda/Azure Functions cheaper than instances?
- Kubernetes cost management: namespaces, resource limits, cluster autoscaling
What training changes: An architect after FinOps training can:
- Design architecture with cost as a design constraint (alongside performance, security, reliability)
- Estimate monthly cost of planned infrastructure BEFORE its implementation
- Propose optimisations: “If we move batch jobs to Spot Instances, we’ll save 60%“
4. DevOps / SRE Teams
What they need:
- Cost optimisation automation: scripts shutting down dev/test environments after hours, removing orphaned resources
- Cost monitoring as part of observability: cost alongside logs, metrics, traces
- CI/CD with cost awareness: cost estimation in pipeline (e.g., Infracost)
- Working with budgets and alerts: AWS Budgets, Azure Cost Alerts
What training changes: DevOps team after training:
- Builds automation: “every Friday at 6 PM we shut down dev/test environments, saving 30% of compute costs”
- Integrates cost checks with PR reviews: “This Terraform plan will increase costs by $20k USD/month—is this intended?“
5. Product Managers / Business Owners
What they need:
- Understanding unit economics: how much does it cost to serve 1 user, 1 transaction, 1 GB of data?
- Trade-offs: performance vs cost (do we need <10ms latency if it costs 3x more?)
- Cost as input to product roadmap: “Feature X will increase costs by Y—what’s the expected revenue?”
What training changes: Product Manager after training can:
- Include cost in the definition of success metrics for a feature
- Negotiate with engineering: “Can we optimise this function, because costs are 40% higher than assumed?”
FinOps roles in the organisation: who does what?
Typical FinOps team structure in a medium/large organisation (500+ employees, cloud budget >$1M/year):
1. FinOps Lead / Head of FinOps
- FinOps strategy, governance, relationships with cloud vendors (negotiating Enterprise Discount Programmes), reporting to CFO/CTO
2. Cloud Financial Analysts (2-3 people)
- Cost trend analysis, forecasting, cost allocation, preparing cost reports for business units
3. Cloud Optimisation Engineers (1-2 people in Engineering team)
- Implementation of optimisations: rightsizing, Reserved Instances recommendations, automation, working with dev teams on code reviews for costs
4. FinOps Champions in teams (distributed model)
- Each product/engineering team has a FinOps Champion—a person responsible for cost awareness in the team, budget reviews, anomaly escalations
In smaller companies (50-200 employees), roles are combined: Cloud Architect + FinOps Practitioner or CFO + FinOps Analyst.
Key principle: FinOps is not one person, it’s a culture of shared responsibility.
FinOps Tools: AWS, Azure, GCP and third-party platforms
Effective FinOps requires appropriate tools. Here’s an overview:
Native cloud tools (built into platforms)
AWS Cost Explorer + AWS Cost & Usage Reports
- Cost visualisation with granularity down to resource level
- Forecasting based on historical usage
- Reserved Instances recommendations
- Pros: Free, full AWS integration
- Cons: AWS only, interface unintuitive for non-technical users
Azure Cost Management + Billing
- Cost analysis, budgets, alerts
- Adviser recommendations (rightsizing, Reserved Instances)
- Pros: Free, integration with Azure Policy
- Cons: Azure only
Google Cloud Cost Management
- Cost breakdown, budgets, recommender (rightsizing, idle resources)
- Pros: Machine learning-driven recommendations
- Cons: GCP only
Third-party platforms (multi-cloud, advanced features)
CloudHealth (VMware)
- Multi-cloud cost management (AWS, Azure, GCP)
- Governance, policy enforcement, custom reports
- Best for: Enterprise with multiple clouds, needing advanced governance
Cloudability (Apptio)
- Cost optimisation recommendations, anomaly detection
- Business intelligence for cloud costs
- Best for: Companies with mature FinOps practice, needing deep unit economics analysis
Kubecost
- Dedicated to Kubernetes (on-premise or cloud)
- Cost breakdown per namespace, pod, service
- Best for: Organisations heavy on Kubernetes
Infracost
- CI/CD integration: cost estimation for Terraform plans
- Pull request comments: “This change will cost $2,400/month”
- Best for: Infrastructure as Code teams, GitOps workflows
Key insight: Tools won’t replace competencies. You can have the best dashboard, but without a team understanding metrics and taking action, it will remain just a pretty chart.
How much can you save? Case studies from the market
Case Study 1: International software house (Poland), 300 developers, AWS budget $80k/month
Initial situation:
- No resource tagging, impossible to assign costs to projects
- Dev/test environments running 24/7
- No Reserved Instances, everything on-demand
Actions after FinOps training:
- Implemented tagging strategy, 80% of resources tagged in 6 weeks
- Automation: dev/test environments shut down outside working hours (5 PM-8 AM + weekends)
- Rightsizing: 40% of instances were oversized, reduced by 1-2 size tiers
- Purchased Reserved Instances for 60% of stable workloads (3-year commitment)
Result:
- Savings: $28k/month ($336k/year)
- Training ROI: training cost $15k, return after 2 weeks
- Additional benefit: visibility enabled chargeback to projects, increasing cost awareness in teams
Case Study 2: Polish bank, Azure migration, budget €150k/month
Initial situation:
- Lift-and-shift from on-premise, no architecture optimisation
- Storage in most expensive tiers (Premium SSD for all workloads)
- No lifecycle policies
Actions after FinOps training:
- Storage tiering: 70% of data moved to Standard SSD or Cool/Archive tiers
- VM rightsizing (Azure Adviser analysis + own CPU/RAM utilisation analysis)
- Azure Hybrid Benefit (used existing Windows Server licences)
- Reserved Instances for core banking systems (1-year commitment)
Result:
- Savings: €52k/month (€624k/year)
- 35% cost reduction
- IT team obtained FinOps Certified Practitioner certificates
Case Study 3: E-commerce startup (Poland), GCP, budget $15k/month
Initial situation:
- Fast growth, team of 10 developers, zero cost management processes
- Costs growing 15% m/m, but nobody knew why
- Load balancers and Cloud SQL left in dev environments
Actions after FinOps training:
- Setup GCP Budgets + alerts (alert at 80% of budget)
- Removal of orphaned resources (Cloud SQL instances, Load Balancers)
- Committed Use Discounts for Compute Engine
- Preemptible VMs for batch jobs (data processing pipelines)
Result:
- Savings: $5.2k/month ($62k/year)
- For a startup with an 18-month runway, savings extended runway by 4 months
- Cost anomaly detection prevented an incident (script in infinite loop generated $8k in costs in 3 hours, alert stopped it)
Common denominator of case studies: Savings of 25-45% of cloud budget within 6-12 months. Key success factors: executive sponsorship, team training (not just FinOps, but also dev/ops/product), automation.
FinOps Certified Practitioner certification—is it worth it?
FinOps Certified Practitioner is a certification issued by the FinOps Foundation (Linux Foundation), representing the industry standard for FinOps competencies.
What the exam covers:
- FinOps framework (Inform, Optimise, Operate)
- Personas and roles in FinOps organisation
- Cloud cost models (AWS, Azure, GCP pricing)
- Optimisation practices (rightsizing, commitment discounts, storage optimisation)
- FinOps metrics and KPIs
- Case studies and real-world scenarios
Format:
- 50 multiple-choice questions
- 60 minutes
- Passing score: 75%+
- Cost: $300 (exam), preparation: self-study or training
Who should consider it:
- FinOps Practitioners/Leads—must-have
- Cloud Architects, Cloud Engineers—highly desirable
- Product Managers working with cloud-native products—worthwhile
- CFO/Finance teams in cloud-first companies—worthwhile for leadership
Certification ROI: Certified FinOps Practitioners have on average 20-30% higher salaries. For organisations: certification ensures the person knows best practices and framework, reducing onboarding time and risk of wrong cost decisions.
Preparation:
- Official FinOps Foundation course (self-paced, $500)
- EITT training (2-3 days intensive, with exam voucher)
- Practice in organisation (6-12 months) + certification
How does EITT train teams in FinOps?
At EITT, we believe that cloud technology is a business tool—and FinOps is the competency connecting IT with finance to maximise the value of cloud investments. Our FinOps training is designed for the Polish market, taking into account the specifics of local companies (banks, telecommunications, software houses) and practical case studies.
Our approach: from assessment to action plan
1. FinOps Readiness Assessment (optional pre-work) Before training, we offer an assessment:
- Analysis of current state: how do you manage cloud costs?
- Inventory of tools and processes
- Gap analysis: what do you have, what’s missing?
- Prioritisation of challenges for your organisation
2. Core training: FinOps Practitioner Training Format: 3 days (24h), hands-on workshop + lectures For whom: FinOps teams, Cloud Engineers, Architects, Financial Analysts working with cloud Programme:
- Day 1: FinOps framework, shared responsibility culture, stakeholder management
- Day 2: Cloud cost models (AWS/Azure/GCP), cost management tools, rightsizing, commitment discounts
- Day 3: Automation, governance, building FinOps practice in organisation, certification preparation
Hands-on labs:
- Analysis of real-world cloud bill, waste identification
- Rightsizing recommendations in AWS Cost Explorer
- Building cost allocation strategy (tagging, hierarchies)
- Setup cost anomaly detection + alerts
- Terraform + Infracost: cost estimation in CI/CD
3. Specialist training (follow-up)
FinOps for Kubernetes (Kubecost)
- 1 day, for teams heavy on K8s
- Cost allocation per namespace, rightsizing pods, cluster autoscaling strategies
FinOps for Leadership
- 4h executive workshop, for C-level and senior management
- Cloud economics, strategic KPIs, governance models, building cost awareness culture
Cloud Financial Analysis
- 2 days, for finance teams
- Forecasting, budgeting, chargeback/showback models, financial reporting
4. Continuous support After training, we offer:
- 3 months of access to our experts (email/Slack helpdesk)
- Monthly webinars: new AWS/Azure/GCP features, best practices updates
- Templates: cost allocation models, governance policies, dashboard templates
- Networking: access to Polish FinOps Community (organised by EITT)
Why EITT?
500+ experts with experience in cloud engineering, DevOps, FinOps 2500+ training sessions delivered for leading Polish and international companies 4.8/5 average rating from training participants Practical approach: case studies from the Polish market, real-world cloud bills, hands-on labs Certified trainers: our experts hold FinOps Certified Practitioner, AWS/Azure/GCP certifications, experience in organisations with mature FinOps practice
Return on investment guarantee: On average, companies save 25-40% of cloud costs within 6 months after training. For an organisation spending $50k/month on cloud, that’s $150-240k in annual savings. Training cost? $10-20k. ROI: 10-20x.
👉 Check our Cloud and FinOps training 👉 Contact an EITT expert
FAQ—most common questions about FinOps
How does FinOps differ from traditional IT cost management?
Traditional IT cost management is CAPEX planning (purchasing servers, licences), depreciation, annual budgets. FinOps is OPEX management in a pay-as-you-go model, where costs are variable, real-time, and require continuous optimisation. Key difference: in the cloud, costs are directly linked to engineering decisions (choice of instance, architecture), whilst in traditional IT, cost decisions were made in procurement/finance.
FinOps connects finance, engineering, and business in shared responsibility—everyone is responsible for costs, not just the IT department.
Does FinOps only apply to large companies?
No. FinOps is scalable. In a small company (10-50 people), FinOps can be part of the Cloud Architect or DevOps Lead role (5-10% of time). In a medium-sized company (100-500 people), it’s a dedicated person (FinOps Practitioner). In enterprise (1000+ people), it’s a team of 3-10 people.
Key principle: the larger the cloud budget, the greater the savings potential. If you spend $10k/month on cloud, 30% savings is $3.6k/year—worthwhile, but may not require a dedicated team. If you spend $500k/month, 30% savings is $1.8M/year—must-have.
How long does it take to build a FinOps practice in an organisation?
Typical timeline:
- Month 1-2: Assessment, training, tool setup (tagging, cost allocation, dashboards)
- Month 3-6: Quick wins (rightsizing, orphaned resources cleanup, dev/test automation)
- Month 6-12: Medium-term optimisations (Reserved Instances, storage tiering, architecture refactoring)
- Year 2+: Mature FinOps practice (unit economics, continuous optimisation, FinOps embedded in culture)
First savings visible after 1-2 months. Mature practice: 12-18 months.
What are the most common mistakes in FinOps implementation?
1. Lack of executive sponsorship FinOps requires cultural changes. Without C-level support, teams won’t take cost accountability.
2. Focus only on tools A dashboard won’t change costs. You need people with competencies, processes, culture.
3. FinOps as a cost-cutting exercise FinOps isn’t “cut costs”. It’s optimising value for money—we spend wisely, not less. Sometimes it’s worth spending more if the ROI is positive.
4. Lack of engineering team involvement If dev/ops don’t feel cost ownership, FinOps will remain in the finance silo. Incentives are needed: cost KPIs in OKRs, per-team dashboards, gamification.
5. Commitment discounts too quickly without analysis Reserved Instances are a 1-3 year commitment. Purchased too early, for poorly selected instances, they can be wasteful. First rightsizing, then commitment.
Can you do FinOps without third-party tools?
Yes, especially in the initial phase. Native cloud tools (AWS Cost Explorer, Azure Cost Management) are sufficient for single-cloud organisations, up to $100k-200k/month budget. Above that or in a multi-cloud environment, third-party platforms (CloudHealth, Cloudability) significantly accelerate work.
Key point: tools are an accelerator, not a prerequisite. Start with native tools + team training, then consider third-party.
How do you measure FinOps success?
Key metrics:
- Cost savings (absolute savings vs baseline)
- Cost avoidance (how much expenditure we avoided through optimisations)
- Unit economics (cost per customer, cost per transaction, cost per GB processed)
- Forecast accuracy (how accurate are our forecasts vs actual costs)
- Reserved Instances/Savings Plans coverage (% of workloads with commitment discounts)
- Resource utilisation (% CPU/RAM utilisation of instances)
Not just “how much did we save”, but “are we spending on the right things that deliver business value”.
How does FinOps impact DevOps and Agile?
FinOps should be a natural extension of DevOps/Agile:
- Shift-left cost awareness: Just as security went “shift-left” (DevSecOps), cost awareness should be in the design/development phase, not just at the end.
- Cost as a non-functional requirement: Alongside performance, security, reliability—cost is a design constraint.
- Cost estimation in CI/CD: PR review contains not only code review, but cost impact (via Infracost, AWS Cost Estimation API).
FinOps doesn’t slow down Agile—it optimises it. Teams deliver value quickly, but with cost consciousness.
Conclusion: FinOps isn’t a trend, it’s a necessity for every organisation operating in the cloud. Cloud costs grow exponentially, but with appropriate team competencies you can save 25-45% of your budget whilst increasing operational efficiency. Training isn’t a cost—it’s an investment that pays back in weeks, not years.
Ready to optimise cloud costs in your company? Contact EITT to receive a personalised FinOps training offer for your team.
👉 Check our Cloud, FinOps and DevOps training 👉 Schedule a consultation with an expert
Read Also
- FinOps in the Cloud - How Training Can Save Millions
- IT Training for Telecommunications - from 5G to Cloud Native
- NIS2 Training - What Your IT Team Needs to Know
Develop Your Skills
This article is related to the training FinOps: cost management and financial optimization in the cloud. Check the program and sign up to develop your skills with EITT experts.