Skip to main content

Using FinomateAI with Azure

AvailableReviewed 24 August 2026

FinomateAI treats Azure as a first-class billing source. You connect a Cost Management export and a least-privilege service principal, then every product page can be filtered to Azure.

This section is the Azure operator’s book: how to use Dashboard, Cost Explorer, anomalies, recommendations, reservations, budgets and reports against Microsoft subscriptions.

FinomateAI Dashboard filtered to Azure with spend by Virtual Machines, SQL, Blob Storage and AKS
Sample Northwind Retail Azure estate for August 2026. Totals are illustrative.

What you need from Azure​

Azure artefactWhy FinomateAI needs it
Cost Management export to a storage accountDaily actual-cost or FOCUS files
Microsoft Entra app + client secretIdentity FinomateAI assumes
Storage Blob Data Reader on the export containerRead the files
Cost Management Reader on in-scope subscriptionsOptional API checks and reservation metadata
Subscription IDs in the onboarding formBilling scope

The principal is not a subscription Owner. FinomateAI does not create VMs, change SKUs or buy reservations in Azure. You do that in the Azure portal or Bicep/Terraform after you accept a recommendation.

Azure path through the product​

JobGuide
Connect the first subscriptionConnect Azure
Orient a monthly reviewAzure Dashboard
Explain a total against Cost AnalysisAzure Cost Explorer
Investigate a spikeAzure anomalies
Rightsize VMs and act on idle disksAzure recommendations
Reservation coverage and quotesAzure reservations
Subscription and team budgetsAzure budgets

Provider chip​

On every page, select Azure. Totals, charts and tables must change. If they do not, you are still looking at All-clouds or another provider.

Sample estate used in these guides​

Screenshots use Northwind Retail:

SubscriptionPurpose
checkout-platformProduction VMs, AKS, App Service, Azure SQL
data-productsBlob Storage, analytics

Region is australiaeast unless a row says otherwise. Your names, SKUs and dollars will differ.

Next step​

Connect Azure, then spend 30 minutes on the Azure chip.