In partnership with
AI
Mastering FinOps for AI
No, we didn’t make a course for AI. Yet 😉
While AI brings exciting possibilities, it also creates new challenges for teams managing cloud costs. Some challenges are familiar, like learning new terms and working with different teams.
The good news is that many parts of managing AI costs are similar to regular cloud services:
Basic math of price times usage still applies
Costs show up in regular cloud bills
You can use tags to track spending
Discounts work similarly
But there are important differences too:
Prices can change dramatically
New services appear often
Measuring usage is more complex
Special hardware like GPUs can be hard to get
Teams are still learning how to use AI efficiently
While AI brings new things to learn, many existing cost management practices still work well. Take it step by step, stay curious, and don't worry - you've got this.
In Partnership with Finout
Cut Cloud Costs by Up to 30%—Turn Alerts Into Action with Finout
You’ve been there before
Tired of Slack threads & endless dashboards that get you nowhere.
Getting engineers to take ownership of cloud spend through alerts is tough.
In large engineering orgs, cloud cost accountability often gets lost. Finout changes that—with automation that turns cost alerts into action.
Here’s how:
Alerts where engineers live—Slack, Teams, no extra clicks
Custom alerts for any team, service, or resource
Context-rich insights via MegaBill + Virtual Tags
Auto-generated tickets with tracked resolution & impact on Jira and even ServiceNow
Join Us at FinOps After Dark – London Edition!
Heading to FinOps Community Day London? We’d love to connect with you at FinOps After Dark—a relaxed, invite-only event for FinOps pros.
FinOps Framework
FinOps Framework 2025 Update
The FinOps Framework has updated its guidelines for 2025 to better match how companies handle their technology costs. The main change is adding "Scopes" to help manage different types of spending.
The framework now looks beyond just cloud services. It includes other tech costs like software subscriptions, data centers, and AI spending. Companies can now create their own custom scopes based on what matters most to their business.
Each type of spending (or "scope") needs different attention:
Cloud services might focus on saving money
Software subscriptions might care more about tracking usage
AI costs might need special monitoring
The framework made these key changes:
Updated its main definition to include all technology costs, not just cloud
Changed some basic principles to reflect broader tech management
Renamed some sections to remove "cloud-only" focus
The changes came from real feedback from thousands of people who use these methods every day. This makes the framework more useful for today's complex technology needs.
AI
Managing the cost of AI: Leveraging the FinOps Framework
Companies are looking at AI costs from two angles: how to keep AI expenses in check, and how to use AI to reduce other costs. Here’s how to make it right:
Planning ahead - Count the expected usage of AI services, which are often charged by tokens (small pieces of text).
Making smart choices - Pick the right AI models and save money by storing common responses instead of asking the AI repeatedly.
Setting rules - Create clear guidelines about who can use AI services and when they can use them.
Watching the numbers - Keep track of how much AI is being used and what it costs, just like watching your electricity bill.
Finding value - Connect AI costs to business results, like seeing if a chatbot actually helps increase sales.
The good news is that many standard cost control methods work well for AI. Companies don't need to create completely new systems - they can adapt their current approaches to include AI expenses.
📺️ VIDEO
How to leverage FinOps Dashboards with Yuriy (AWS CID)
Explore how you can leverage the Cost Intelligent Dashboards from AWS to help your FinOps practice from Yuriy, TAM of the AWS CID Team.
Azure
Microsoft Cost Updates - March 2025
Azure Kubernetes Service (AKS) now shows you exactly where your money goes. You can see which parts of your system are not being used fully, helping you avoid waste.
AWS connector service is ending on March 31st, 2025. If you use this tool to track AWS costs in Microsoft's system, you'll need to switch to a different method soon.
Azure OpenAI users can now swap their reservations directly in the Azure portal. This gives more options to manage pre-paid AI computing resources.
Events
Webinar: Top Mistakes to Avoid in your AWS Commitments
Many companies are leaving significant savings on the table due to common, yet costly, mistakes. In this interactive session, we'll dive deep into the pitfalls you need to avoid to optimize your AWS spending.
Date: March 27th, 2025
Time: 6:30 PM [Europe/Madrid Time]
Agenda
6:30 PM - 6:45 PM: Community talk by Oliver Assad
6:45 PM - 7:15 PM: Panel with Aran Khanna "Top Mistakes to Avoid in your AWS Commitments"
7:15PM - 7:30 PM: Q&A
Featuring
Aran Khanna, CEO of Archera: With years of experience helping companies optimize their cloud spending, Aran will share his expert insights and practical advice.
Oliver Assad, FinOps Author & Consultant: Oliver will deliver a 15-minute community talk, bringing a fresh perspective and real-world experience to the discussion.
Nilofar Bhurawala: Head of FinOps at Smarsh, she’ll provide a ownder
🎖️ Mention of Honour
How to Make an AI Cost & Usage Tracker
I love when people get to code in FinOps. James Barney showed us how to track and manage costs when using AI services in companies. Here’s the digest:
Three ways to track costs:
Basic: Count how many times you use the AI
Medium: Make a good guess about token usage
Advanced: Track exact token counts
The article suggests creating a special database to keep track of:
Who is using the AI
How many tokens they use
Which AI models they're using
How much it costs
Special note about bulk buying AI power (called PTUs):
Can save money for heavy users
Needs careful planning when multiple teams share it
Math formulas to split costs fairly
🆕 Announcements
Microsoft March 2025 Updates
AWS GenAI Cost Optimization Guidelines Update
🏆 Quick Wins Resources
Evergreen FinOps Resources to apply Today!
Professional Spotlight
James Barney
Bringing the Value for AI
I’ve seen quite a lot of contributions where James showed up. Loved his latest contribution to AI updates within the Foundation. If do FinOps as code, you’re on my team.
That’s all for this week. See you next Sunday!
Before You Go …
Whenever You Are Ready, Here’s How We Can Collaborate
Master FinOps with us: Learn about Tagging & Naming Conventions with our Mastering FinOps Course, taught by FinOps Author Alfonso San Miguel.
Sponsor this newsletter: Promote your company in this newsletter and reach the Cloud audience that wants to maximize business value and reduce costs.
Collaborate with SmartClouds: Our brand expands to more than just newsletters. Podcasts, Posts, Webinars, Events, and any collaboration related to FinOps are available.
Rate Today's NewsletterFeedback = Better Newsletter for You |
P.S. 1: Know someone who’ll enjoy this newsletter? Please forward it. You’ll make them Stay Ahead in FinOps. Was this email forwarded to you? Subscribe and get all editions!
P.S. 2: I’ll do my best to provide you with the best content. However, some of it may slip. If you think something should have been included, please REPLY to this