Presented by

Want to appear here? Talk with us

This 2026 guide teaches how Cloud Unit Economics turns raw cloud spend into metrics that matter: cost per transaction, user, or AI token.

Learn how unit costs, smart allocations, and Cloud Efficiency Rate (CER) help teams:

The FOCUS 1.3 specification just launched with three major fixes that solve real headaches for anyone managing cloud costs across multiple providers. Here's what changed and why it matters:

Your contract commitments now live in a separate dataset instead of being buried in cost and usage rows. You can see start dates, end dates, remaining units, and descriptions all in one query.

FOCUS 1.3 adds new columns that show you exactly how your cloud provider split the costs across workloads.

Stale or incomplete data can wreck your month-end close process. Providers must now add timestamps showing when data was last updated and flag whether it's final or still being revised.

The spec now distinguishes between the Service Provider (who sells you the service) and the Host Provider (where the service actually runs).

Two existing columns - Provider and Publisher - are being deprecated in 1.3 and will be fully removed in version 1.4. They're being replaced by the more specific Service Provider and Host Provider columns.

The next release will add a dedicated invoice dataset, expand contract commitment details, and improve consistency across providers.

Read More

Tag Automated Backups in RDS & Aurora You can now apply tags to automated backups separately from parent instances.

Custom Container Stop Signals for ECS Fargate Fargate now honors custom stop signals defined in container images rather than forcing a standard SIGTERM.

Read All AWS Updates

Serverless Workspaces in Azure Databricks Now in Public Preview, Serverless Workspaces offer a fully managed SaaS experience.

Read All Azure Updates

BigQuery Autonomous Embeddings Now in preview, this feature automates embedding generation and upkeep as table data changes.

Spanner & AlloyDB Optimization Spanner’s Query Insights added new columns (Client IP, User Agent) for deeper cost attribution, while AlloyDB introduced Query Plan Management

VM Extension Manager Manage guest agents (like Ops Agent) across fleets via policy.

Read All GCP Updates

Your voice defines this community. 2025 is wrapping up. It is time to look back.

We are hosting a special End of Year Community Call.

This is an open conversation. We want to decide our future together.

18th of December - 6PM Spain Time / 12PM EST Time

Open to everyone. Register for the call below.

Manual cost optimization is a losing battle. If you are still chasing engineers to tag resources or manually checking dashboards for waste, you are creating a bottleneck. The future of FinOps is AUTOMATION.

In this episode, we dive deep into how to build a "Self-Driving" FinOps practice by integrating cost controls directly into your development lifecycle (SDLC).

We discuss with Ernesto (Glassity) and Dustin (KPMG) how to replace manual gatekeeping with automated guardrails in your CI/CD pipelines and Terraform code.

Two weeks in a row we got a new FinOps open source tool! A new open source tool called ecos just launched to help teams make sense of their AWS cloud costs.

ecos takes those AWS Cost and Usage Reports and turns them into something you can actually use. It cleans up the data, organizes it, and makes it fast to search through.

The tool uses what's called a medallion architecture. That's a fancy way of saying it processes your data in stages. Raw data comes in from AWS. Then it gets cleaned up. Then it gets organized for analysis. Finally, it's ready to plug into your dashboards or reports.

The tool is free and open source, which means you can start using it today without a sales call or contract negotiation.

Read More

Azure Hybrid Benefit can save your company thousands each month on cloud costs. But one wrong click can trigger a $50,000 audit bill instead.

The problem is simple. Companies think their on-premises Windows Server and SQL Server licenses automatically work in Azure. They don't. Not without the right paperwork, active Software Assurance, and careful tracking. When Microsoft auditors show up six months after your migration, they ask for proof. If you can't produce it, you pay. And it's not a small bill.

Here's what you actually need to use Azure Hybrid Benefit legally. You need active Software Assurance that hasn't expired even by one day. You need original purchase orders and license keys. You need to prove which physical server cores your licenses covered. You need documentation showing you shut down the on-premises servers within 180 days.

One financial services company learned this the hard way. They moved 156 virtual machines to Azure and enabled Azure Hybrid Benefit across all of them. Initial savings looked great at $42,000 per month.

Six months later, Microsoft sent an audit request. The problems piled up fast. They couldn't prove they owned 28 SQL Server licenses. Software Assurance had expired 18 months earlier on 12 licenses. Some licenses were OEM versions tied to Dell hardware that can't transfer to the cloud.

Worse, they were using 196 virtual CPU cores in Azure but only owned licenses for 64 physical cores. That's a shortage of 132 cores. They also kept on-premises servers running as backup for 8 months after migration, violating the 180-day grace period.

The bill came to $78,000 after negotiation. Add consultant fees, staff time, and ongoing higher Azure costs, and the first year total hit $364,000. The root cause wasn't technical. It was organizational. Nobody checked if they could legally use Azure Hybrid Benefit before turning it on.

Read More

A FinOps engineer posted a question that many in our field might face someday: what happens when you actually finish the job?

This person started in 2018 when their company's cloud setup was a mess. No organization. No controls. Just chaos everywhere.

Seven years later, their team of four has built something impressive. They have full AWS coverage with savings plans and reservations in place. They created hundreds of optimizations for developers. They wrote detailed documentation for everything. They built playbooks that work every time. They got all resources tagged and managed through infrastructure as code. They even helped other teams cut costs in their CI/CD pipelines.

When new cloud projects started, they just copied their proven methods and made them work on other platforms. Everything succeeded.

Now they feel stuck. Leadership is happy. Stakeholders are satisfied. The systems run themselves. And this engineer feels like there's nothing left to build.

One interesting take suggested that FinOps teams shouldn't be permanent structures. Once you build the systems and processes, you should shift everything left and make your team unnecessary by embedding FinOps into every team's daily work.

Read More

Participate by answering our questions. There will be surprise gifts to winners every month!

FinOps Forecasting Top Writer

François has been making a wonderful article series about forecasting, budgeting (one of the most difficult topics in FinOps) featured in the newsletter. Follow him to learn every week!

Connect with the Professional of the Week

Azure FinOps Master Early Access
(Starting February 2026)

Alfonso is creating a masterpiece.

The biggest educational we’ve done till date.

Enrollment for super early users closes the 20th.

Learn all you need to optimize resources in Azure.

199€ Super Early Bird – Ends December 20th

299€ Early Bird: Opens XXX - Ends XXX

499€ Launch Offer: Opens XXX - Ends XXX

699€ Starting Offe: Opens XXX - Ends XXX

999€ Final Price

The perfect Christmas Gift for Fellow FinOps Practitioners