The CTO-CFO relationship has had to evolve at the speed of technology. In the cloud, these two counterparts—finance and engineering—must work in lock-step for a successful FinOps implementation.
By aligning engineering goals (e.g., performance and scalability) with financial goals (e.g., reduced unit costs, efficient utilization of resources), FinOps teams can achieve optimal outcomes related to cloud cost and rate optimization.
The FinOps Foundation, recognized for setting FinOps standards and best practices, recently released its annual 2024 State of FinOps Report. It offers a comprehensive look at the current trends in cloud financial management. This blog highlights several takeaways from the report.
TL;DR
- In the backdrop of a tighter economic environment, results from the State of FinOps 2024 data underscore a need to reduce waste and confirm the importance of automation, which has increased in priority. (compared to previous years)
- FinOps practitioners focus more on managing committed-based discounts, such as AWS Reserved Instances or Google Cloud Committed Use Discounts, which also grew in priority — second to reducing waste and unused resources.
- This change in priority for managing commitment-based discounts suggests that optimizing the price you pay for cloud (aka, rate optimization) has become more important as a complement to usage optimization for managing cloud costs.
- Managing AI costs with FinOps and using AI for FinOps itself are in the experimental phase, and AI-related costs may not have a noticeable impact yet.
- FinOps automation platforms that consistently deliver saving outcomes will continue to be invaluable for organizations as they further their FinOps adoption journey.
Reducing Waste and Managing Discount Commitments: FinOps Priorities for 2024
Considering the economic climate, it’s no surprise to see “reducing cloud waste” emerge as the top priority for FinOps teams in 2024. Cloud compute services continue to be more heavily optimized, because those services often comprise the most significant portion of cloud spend. However, compared to last year, optimization for containers has risen, suggesting more adoption of Kubernetes and microservices-based architecture (from 5th to 4th place)
Managing commitment-based discounts, including AWS Reserved Instances and Savings Plans, Google Cloud Committed Use Discounts, and negotiated discounts, increased to 43% (from just 7% in previous years) and is now the second highest priority.
Image Source: The 2024 State of FinOps Report by FinOps Foundation
These priorities suggest a changing perspective within FinOps practices, from primarily optimizing usage (i.e., workload optimization) to optimizing price (i.e., rate optimization). Traditionally, FinOps teams have concentrated on reducing waste by minimizing usage via re-architecting solutions or making engineering changes. While effective, these strategies can be resource-intensive and time-consuming for engineers, often requiring months or even years to implement. Not to mention ongoing maintenance as engineering environments change and new services in new regions are added.
In contrast, rate optimization, or optimizing the price you pay for cloud services (i.e., using commitment-based discounts), offers a more immediate solution within the financial domain without necessitating changes in engineering practices. Focusing on rate optimization ensures cost savings for the organization and allows engineers to focus on innovating.
The challenge, however, is the inelastic nature of commitments. Once an organization commits, it becomes locked into 1- or 3-year terms, with payments due regardless of subsequent changes in usage. If usage drops, organizations can end up with unutilized commitments, resulting in wasted cloud spend. To avoid this risk, organizations will typically cover only a portion of their usage with commitments, which leads to suboptimal savings outcomes.
Despite its rise in importance among practitoners, managing commitment-based discounts remains one of the less-mature FinOps capabilities, behind measuring unit costs, forecasting, and resource utilization. Data from our Effective Savings Rate (ESR) Benchmarks and Insights 2024 report supports this finding. In fact, more than half of the organizations surveyed do not utilize discount instruments, such as AWS Savings Plans or Reserved Instances, for compute (EC2, Lambda, and Fargate). This gap underscores the potential for organizations to improve cloud cost management through the efficient use of commitment-based discounts.
Effective Savings Rate (ESR) is an objective FinOps metric that measures the ROI of your cloud cost optimizations, whether rate or usage optimizations. ProsperOps uses ESR to define the discount you receive off the on-demand rate for associated cloud services. Learn more about ESR here.
According to the 2024 State of FinOps report, managing commitment-based discounts tends to be more important for larger and more mature organizations, implied by their cloud spend. As usage and cloud costs increase, so does the opportunity cost of not leveraging commitment-based discounts to reduce the price paid for cloud services.
Image Source: The 2024 State of FinOps Report by FinOps Foundation
A similar trend was apparent in our ESR Benchmark and Insights Report. Larger organizations (those spending $10M per month, or more) tend to achieve better savings outcomes from managing commitment-based discounts. This is likely due to their access to more resources and expertise.
The Impact of Automation on FinOps: A Closer Look
“Enabling automation” increased in priority for FinOps teams, ahead of “Empowering engineers to take action.” This trend indicates a strategic shift towards leveraging technology to improve optimization outcomes while minimizing manual effort — or, eliminating actions engineers are sometimes required to take. Regardless, the goal is the same. To scale efficiently, teams must learn to do more with less.
Image Source: The 2024 State of FinOps Report by FinOps Foundation
Despite the growing importance for automation in FinOps, teams are still early to adopt automated cloud financial management tools. Currently, automation is primarily used to gather data, identify anomalies, and recommend actions. Decision-making or execution of actions is still largely dependent on human input. Despite using automated tools within FinOps workflows, teams often rely on people to control and complete the work.
One barrier to embracing automation within FinOps is the reluctance to trust automated systems with financial decisions. This sentiment is stronger among larger spenders and regulated industries, where there’s more to lose. As innovation in FinOps continues, adopting fully automated cloud financial management solutions will become more commonplace.
Image Source: The 2024 State of FinOps Report by FinOps Foundation
Empower Engineers to Act with Automation
Relative priority for “empowering engineers to take action” is now ranked sixth, compared to being the top priority last year and in 2022. From our perspective, empowering engineers is more about enabling them to focus on high-value work and not empowering them to push buttons or execute tasks.
Engineers have competing priorities — not always related to optimizing cloud costs. Engineers burdened with rate optimization and managing commitment-based discounts cannot innovate or build new products. Automation, on the other hand, gives engineers time back to focus on more meaningful work.
Many Organizations Are Only Experimenting with AI/ML
Less than one-third (31%) of organizations indicate that costs from AI/ML technologies impact their FinOps practices. Within organizations spending more (generally) on cloud costs, this increases to 45%, possibly due to larger orgs experimenting more with AI, and AI-related services accounting for more of their variable spend.
Using AI/ML technologies for FinOps is a growing topic of interest, but still in the early innings. Certainly, generative AI can extract insights and summarize cost data, but developing, training, and fine-tuning a robust AI model to use data to analyze and execute tasks effectively requires significant time and resources.
Our discussion with Eric Lam (Google’s Head of Cloud FinOps) explores the role of generative AI in FinOps.
Final Thoughts
FinOps teams relying on reporting and visibility tools to understand cloud spending are only solving part of the problem. To optimize and operate, you need automation.
Book a demo or request a Free Compute Savings Analysis to learn more about how ProsperOps automates away the complexities of managing commitments while minimizing lock-in risk with zero manual effort.
