There is a specific moment that changed how I think about cloud cost. It was not a workshop or a vendor briefing. It was a number on a screen during an architecture review. A team had built a well-designed integration platform. It was elegant. It was also costing three times what it needed to, and nobody in the room could have told you that, because nobody had asked the question during the design.
The cost ran for eighteen months before a finance review triggered a conversation that should have happened in the original architecture session.
The Dan Observation: I once saw an environment costing £40,000 a month more than necessary because the team was using “Standard” defaults for everything. They were being “safe,” but in the cloud, unmanaged safety is just a tax on the business. That £40k represented an entire engineering team they couldn’t hire because the budget was being eaten by idle compute.
Every line on your Azure bill is the financial consequence of a decision made in a design meeting. Most organisations are overspending because their architecture process does not treat cost as a first-class requirement alongside availability and security.
The Architecture Cost Framework
Cost is the accumulated result of decisions across five distinct categories:
1. Specification decisions — Choosing the right SKU. The biggest win here? Switching to Linux for App Services. It eliminates the “Windows Tax” and can slash licensing costs by 50% instantly.
2. Design pattern decisions — Choosing Async over Sync. Decoupling services with a Service Bus allows you to scale independently and avoid paying for a “simultaneous peak” across your entire chain.
3. Geography decisions — Moving Dev/Test to cheaper regions like North Europe. If production needs to be in UK South for compliance, your sandbox doesn’t.
4. Commitment decisions — Stopping the “Later Never Comes” trap. If a workload has been running for six months, it’s stable. Stop paying the Pay-As-You-Go premium for flexibility you aren’t using.
5. Governance decisions — Moving from “Audit” to “Deny.”
The Dan Rule: The single most impactful governance move is switching Azure Policy to Deny for missing tags. It stops the “It wasn’t me” defense and ensures 100% cost attribution from day one. It’s the difference between being a “Cloud Accountant” and a “Cloud Architect.”
Where to Start
If you’ve inherited an unmanaged Azure environment, follow this sequence:
- Run Azure Advisor: Identify the abandoned resources.
- Review Commitment Coverage: Aim for 70% on your production baseline.
- Audit the OS: Switch App Services to Linux where possible.
- Establish the Review Cadence: If you aren’t making one architectural decision every month to reduce spend, you aren’t managing the cloud—you’re just paying for it.
Cost as Architecture Discipline
Cost-optimised architecture is not about being frugal; it is about being deliberate. Every VM tier and region should be chosen because it matches a measured requirement, not because it was the default in a Terraform module.
My Philosophy: The cloud isn’t inherently expensive; it just provides enough rope for poorly designed architecture to hang your budget. Treat cost as a performance metric, and the bill usually takes care of itself.
Ready to make cost a first-class concern in your Azure architecture? Get in touch to talk through your current architecture and where the biggest opportunities are.
Explore the full Cost-Optimised Cloud series for practitioner-depth guides on each area covered here.
