.jpg)
The FinOps conversation changed at re:Invent 2025. After years of focusing on visualizing cloud spend, a new fundamental truth is emerging:
Visibility alone doesn’t lead to efficiency.
Nearly every enterprise has dashboards, alerts, and allocation models, yet waste persists, engineering backlogs grow, and “optimization” remains reactive rather than an operational discipline. In 2025, cloud cost management remains a key challenge for 84% of organizations.
At re:Invent, we participated in many conversations with nearly everyone, from engineers to Fortune 500 leaders, expressing a similar sentiment. We don’t need another unused-volume alert. We need to better understand the architecture.
AWS confirmed the shift toward ongoing efficiency by introducing its new Cost Efficiency Metric. The Cost Efficiency Metric is a standardized score that compares your last 30 days of optimizable cloud spend against today’s potential savings. But this only addresses part of the problem: it quantifies your optimization gap but doesn’t close it.
To address this gap, organizations are emphasizing an emerging discipline: ongoing efficiency.
Ongoing efficiency reduces costs by continuously monitoring how workloads consume resources and detecting architectural waste early. It gives teams the context they need to fix issues as they occur, not months later. This creates a stable efficiency posture regardless of how environments evolve...
Here is our take on why 2026 belongs to ongoing efficiency:
For more than a decade, billing data flagged basic waste like unattached volumes, idle nodes, or simple rightsizing opportunities.
This worked when infrastructures were predictable, monolithic, and VM-centric. But modern architectures like microservices, serverless, Kubernetes, managed databases, and AI inference workloads are too dynamic and interconnected for such shallow detection. Waste today hides as normal usage patterns. Modern waste identification requires deep analysis of workload behavior, scaling logic, configuration drift, token usage, data movement, and cross-service dependencies. Billing data offers no insight.
The biggest barrier to cloud efficiency is credibility. Engineers ignore optimization recommendations when alerts lack context.
Legacy Cloud Cost Management (CCM) tools generate large volumes of non-actionable alerts. Alerts without consistent, reliable context deliver little value and are ignored, creating a barrier to ongoing efficiency. When the tools engineers are supposed to rely on miss performance realities and architectural context like dependency chains, scaling behavior, ownership, and nuance, engineers spend more time validating than acting. It becomes easy to lose trust.
Cloud Efficiency Posture Management (CEPM) tools differ from CCM tools. Rather than focusing on cost management alone, they focus on ongoing efficiency. This approach gains engineering trust by delivering Context-Rich Intelligence, the “who, what, and why” behind every insight:
With CEPM tools, you can further solidify engineering buy-in with Detection Tuning. Detection Tuning allows you to adjust thresholds, override scopes, and align logic to each environment’s architecture. The resulting alerts have fewer false positives and highly tailored insights. This leads to engineering buy-in.
Reliable detection turns “cost policing” into an Engineering-Native Workflow, bringing efficiency into how systems are designed, reviewed, and operated, and is a major step toward achieving ongoing efficiency.
Automation is one of the most desired, least trusted aspects of cloud efficiency. Enterprises want faster, safer remediation, but can’t grant open-ended write access. Automation that acts without transparency is a dealbreaker. CEPM addresses this trust gap by including and empowering engineers through engineering-native capabilities that drive action with confidence.
Automation accelerates the work, but engineers authorize the action. This model preserves speed and accuracy without sacrificing control.
CEPM is the engineering-native approach to ongoing efficiency. In addition to continuously monitoring your environment for hidden waste, CEPM tools integrate directly with familiar tools like Jira and ServiceNow.
CEPM doesn’t replace the working mechanisms teams rely on, like reports, alerts, tickets, and scripts. Instead, it elevates their capabilities to match the complexity of modern cloud environments:
2026 will be the year cloud efficiency becomes an engineering competency, not an afterthought, cost-center mandate, or reactive cleanup ritual.
With CEPM, organizations enter a new era where:
re:Invent 2025 confirmed much of what we’ve been working on here at PointFive with the need to keep pace with the major cloud providers, so we can further our customers’ pursuit of ongoing efficiency.
When AWS introduced Database Savings Plans, its major expansion of commitment options across managed databases and serverless offerings, PointFive immediately worked to further this initiative with our customers. Within days, PointFive integrated support for this new model across its optimization workflows.
Commitment mechanisms now extend far beyond compute into DocumentDB, DynamoDB, Neptune, DMS, Aurora Serverless, Valkey Serverless, and other services that power mission-critical workloads. As these models proliferate, enterprises need systems that adapt instantly, without waiting for quarterly releases or manual recalibration.
This is precisely what CEPM anticipates: a world where optimization is continuous, adaptive, and engineered into the workflow itself.
.jpg)
