How FinOps Drives Value for Every Engineering Dollar
Learn from a SaaS $10B Company how to do FinOps for Engineering
Victor Garcia
September 14, 2025
Together with DoIt
Stop Cloud Waste Before It Eats Your Margins
Learn how leading FinOps teams bring clarity to chaos, eliminate waste, and bring spend back in line with business value. It’s time to fix the fundamentals and build for sustainable, profitable growth.
In this webinar, DoiT FinOps experts share proven frameworks, real-world examples, and practical steps to align cloud spend with measurable business value.
ENGINEERING
What Engineers Need to Know about FinOps
The company's main rule is simple: user experience comes first. If saving money makes the app worse for learners, they won't do it. For example, they pay more for better AI models to make sure Lily the owl sounds just right in their Video Call feature, even though cheaper options exist.
Business value is the ONE THING in FinOps. And Duolingo proves it again.
See where every dollar goes
They track costs down to the hour and tag everything. One engineer found that their Video Call feature cost 60-100 times more for text-to-speech than a similar feature called Roleplay. The fix was simple: add memory caching like Roleplay had. This cut their overall text-to-speech costs by 40%.
Focus on waste, not just spending
Instead of strict budget limits that might slow down new ideas, they look for pure waste. They moved expensive monitoring tools to cheaper ones without losing any useful data. They celebrate when experiments save money without changing how well people learn.
Fix old tech problems
Outdated systems cost more money to run. Small changes like switching to better computer instances, removing unnecessary API calls, and updating old software versions can save big amounts.
Let computers do the boring work
With only five people on their FinOps team, they can't check everything by hand.
They use scripts to automatically spot when costs jump up across their 100+ services.
They also create weekly and monthly reports that keep everyone informed without extra work.
One recent success story shows how well this works: their dashboard caught a 10x spike in AI costs tied to certain users. Finance spotted it, talked to the product team, and engineers fixed it in hours instead of weeks.
FINOPS CULTURE
Revolutionizing C-Suite with the FinOps Cycle
FinOps teams have been stuck in a one-way relationship with company leaders for too long. Borja Martínez, a FinOps expert at NTT Data, thinks it's time to change that dynamic completely.
Traditional FinOps follows three phases in a loop: Inform, Optimize, and Operate. But Borja suggests flipping this model to create a two-way street between executives and FinOps teams.
Phase 1 - Inform (But Different)
Instead of FinOps teams just reporting numbers upward, executives share their business strategy and goals downward.
Phase 2 - FinOps-ize (Beyond Simple Savings)
With strategic context, FinOps teams can do more than basic cost cutting. They integrate financial thinking into every tech decision to drive real business value. It's about making the whole company think financially, not just finding savings.
Phase 3 - Operate (FinOps Guides Strategy)
FinOps teams return to executives with data that shows strategic insights. They present which projects create the most value, where money is being spent, and how tech choices affect business results.
The cycle then repeats with even better information sharing both ways. Executives share more detailed strategy, FinOps teams provide deeper insights, and the whole company gets stronger.
FINOPS EVENTS
The Hybrid FinOps Advantage
FinOps has expanded far beyond public cloud. Managing costs across data centers, AWS, Azure, SaaS, and AI workloads separately prevents total cost visibility and missed savings.
Discover how FinOps 2.0 strategies deliver comprehensive optimization across your entire technology portfolio.
You'll Learn How To:
Achieve total cost visibility across data centers, multi-cloud, SaaS, and AI infrastructure
Optimize the complete technology stack with unified intelligence and automation
Break down silos between FinOps, ITAM, procurement, and engineering teams
Speakers
Jeremy Chaplin, Gerhard Behr & Victor Garcia
November 23rd - 6:00 PM CEST / 10AM EST
PRIVATE CLOUD
Tesco Sues Broadcom for £100M Over VM Licensing Dispute
Tesco, one of the UK's biggest supermarket chains, is taking Broadcom to court over a software fight that could affect how food gets to your local store. The grocery giant filed a lawsuit asking for at least £100 million in damages.
Back in 2021, Tesco bought what they thought were permanent licenses for VMware software. They also paid for support services that were supposed to last until 2026, with an option to extend for four more years.
But after Broadcom took over VMware, things changed. Broadcom now wants Tesco to switch to a subscription model. Tesco says this forces them to pay much higher prices for software they already own.
The stakes are higher than just money. Tesco warns that VMware software runs about 40,000 of their computer servers. These systems connect to cash registers in stores and help manage operations across the UK and Ireland.
Tesco isn't alone in this fight. Other big companies like AT&T and Siemens have also sued Broadcom over similar issues. Some European customers report price increases of up to 1,500% after the VMware changes.
Another one for Broadcom. Read More
CLOUD PROVIDERS
AWS Graviton & AMI Updates, Azure EA Updates
AWS
Amazon ElastiCache expanded Graviton3-based M7g and R7g nodes to more regions. These deliver up to 28% higher throughput and 21% better P99 latency compared to Graviton2, offering significant price-performance improvements for your caching workloads.
New AMI Usage feature in Amazon EC2 helps monitor Amazon Machine Images across accounts. Track AMI consumption, identify resource dependencies, and find unused AMIs for cleanup to optimize image-related costs.
AWS WAF now includes 500 MB of free CloudWatch Logs ingestion for every 1 million requests processed. This baseline helps manage security logging costs while maintaining compliance visibility.
AWS Clean Rooms supports configurable compute sizes for PySpark jobs. Specify instance types and cluster sizes at runtime to match workload requirements and avoid over-provisioning smaller tasks.
Microsoft Azure
Azure's latest updates include Service Principal support for partners, enabling programmatic access to customer cost data. New user-friendly features like collapse buttons in the Pricing Calculator simplify complex estimates for better cost planning.
Azure Firewall's ingestion-time transformation is now generally available. Filter or modify logs before storage in Log Analytics to significantly reduce logging volumes and costs while retaining critical security insights.
Google Cloud
🫙🫙🫙
AI
Big Developments on Anthropic’s AI Economics
Bigg changes are happening in the world of AI cost management, specifically for companies using Anthropic's AI models. These updates could change how businesses track and control their AI spending.
Anthropic's New Cost Tracking Tool
Anthropic just released a new way for companies to see exactly how much they're spending on AI. Think of it like a detailed receipt that shows you every penny spent on AI tokens. Companies can now see costs broken down by different teams, users, or projects.
The tool also tells you where you can save money. It points out two main ways to cut costs: using batch processing and improving something called prompt cache hit ratios.
You can send this cost data to popular monitoring tools like Datadog or Grafana without writing any code.
AWS Makes Caching Automatic
Amazon Web Services announced automatic caching. This works best for coding assistants that use longer prompts and responses. Most business users with shorter prompts won't see as much benefit.
These developments show how quickly the tools for managing AI costs are improving as more companies start using AI in their daily work.
📺️ VIDEO
Mastering AI Economics in 2025: FinOps Strategies to Handle AI Spending
From tokens to GPUs: the right model, prompt caching, and batching to cut costs up to 50% and avoid surprises. Real cases, KPIs, and practical governance.
Get the insights from Alon Shvo, David Gross & Vaibhav Sharma on how to control AI costs and making business thrive with this technology.
🎖️ MENTION OF HONOUR
Demystifying Azure Log Retention
Amazing deep dive by Shannon!
The Main Players:
Activity Logs track big changes like creating or deleting resources. They hang around for 90 days for free, but you can export them elsewhere to keep them longer.
Resource Logs show what happens inside your services, like who accessed your Key Vault or what queries hit your database. These don't save automatically - you have to turn them on first.
Log Analytics is where most of your monitoring data ends up. It keeps things for 30 days by default, but you can stretch that to 12 years if needed. Just know that longer storage costs more money.
Metrics aren't exactly logs, but they matter too. They track performance stuff like CPU usage and stick around for 93 days.
Entra ID logs track who signed in and what changed in your directory. Free accounts get 7 days, while premium accounts get 30 days.
Application Insights watches your apps and keeps data for 90 days, extendable to 2 years.
The Money Talk:
Here's where it gets real. Five gigabytes of logs per day adds up to 1.8 terabytes per year. At ten cents per gigabyte, that's about $1,000 annually just for storage.
The smart move is to think about what you actually need. Keep critical security logs longer, but maybe those debug logs from last month can go.
Archive storage is cheaper but comes with a catch - you can't search through archived data easily, and getting it back can be unreliable.
Professional Spotlight
Peter Crenshaw
Manufacturing FinOps Excellence
Peter is been one hell of a supporter of the summit. He always shares good FinOps content and it’ll be a pleasure to learn from him in our event.
That’s all for this week. See you next Sunday!
The FinOps Event You Can’t Miss (And it’s FREE)
Move from quick fixes to strategic planning
Strengthen your financial metrics
Use AI to optimize costs
This is the last big FinOps event of 2025, showcasing proven strategies from companies handling large cloud budgets.
October 23rd, 2025 | 4:00 PM - 8:00 PM CEST
Limited seats available
FinOps for all!