Azure Idle Resource Cost Optimizer

Overview

This flow is a weekly Azure cost-optimization sweep built entirely on Azure Resource Manager (ARM). Every Monday at 06:00 UTC it enumerates compute and network resources across the subscription, evaluates each for underuse, estimates the monthly savings of acting, writes every finding to a Dataverse backlog table, and reports a prioritized rightsizing list by email plus a Teams summary.

Why it matters: Cloud waste accumulates quietly. A weekly, quantified list of specific actions turns a vague "reduce spend" goal into a concrete, owner-actionable backlog.

As-built note: your-org has no Azure Resource Manager *connector*, so ARM is called via the built-in HTTP action with ActiveDirectoryOAuth (the FlowLibs ARM pattern — audience = ARM base URL + /). This is the correct connector-first outcome: ARM has no Power Platform connector, so HTTP with AAD OAuth is the sanctioned path. Three underuse categories ship in this build — low-CPU VMs, unattached managed disks, and idle/unassociated public IPs. Empty App Service plans and Azure Advisor enrichment are documented as extensions in the Customization Guide.

Use Case

A platform / FinOps team wants a recurring, evidence-based rightsizing report — *what* is underused, *by how much*, and the *saving* — rather than digging through Cost Management manually. Findings land in Dataverse so they can be triaged, assigned, and tracked as a backlog, and each weekly run is tied together by a correlationId for auditability.

Flow Architecture

Environment Variables

Connectors & Connections

Connector	API name	Actions used
Microsoft Dataverse	shared_commondataserviceforapps	CreateRecord
Office 365 Outlook	shared_office365	SendEmailV2
Microsoft Teams	shared_teams	PostMessageToConversation

Note — All connections are referenced as solution connection references; the flow is portable between environments as long as a connection is mapped at import time.

Customization Guide

Almost every realistic variant of this flow can be implemented by changing environment variable values. A few cases require small edits inside the flow definition — those are called out explicitly below.

Add App Service plans

Add a List_App_Service_Plans HTTP GET (Microsoft.Web/serverfarms?api-version=2023-01-01) and a loop flagging plans whose properties.numberOfSites == 0.

Azure Advisor

Pull Microsoft.Advisor/recommendations (category Cost) for higher-quality, Microsoft-generated findings and merge them into the report.

Auto-remediate

Behind an Approval, delete unattached disks older than N days or deallocate low-CPU VMs.

Chargeback

Attribute findings to owning teams via resource tags and route per-team reports.

Better pricing

The disk and public-IP savings use simple rate constants ($0.05/GB-mo, $3.65/mo); swap in Retail Prices API calls (as the VM branch already does) for exact figures.

Production secrets

Replace flowlibs_AzureClientSecret with an Azure Key Vault reference.

Key Expressions

The flow is intentionally light on Power Fx / WDL gymnastics — the heaviest expressions are the branch-name concatenation and the approval outcome check. They are listed below in the order they appear in the flow.

@coalesce(last(first(first(body('Get_VM_CPU_Metrics')?['value'])?['timeseries'])?['data'])?['average'],100)

@less(float(outputs('Compose_VM_Avg_Cpu')), float(variables('varLowCpu')))

@mul(coalesce(first(body('Get_VM_Price')?['Items'])?['retailPrice'],0),730)

@equals(toLower(coalesce(items('For_Each_Disk')?['properties']?['diskState'],'')),'unattached')

@empty(items('For_Each_Public_IP')?['properties']?['ipConfiguration'])

@{concat(variables('varArmBase'),'/')}

formatDateTime(addDays(utcNow(), mul(int(variables('varLookback')),-1)),'yyyy-MM-ddTHH:mm:ssZ') … / … utcNow('yyyy-MM-ddTHH:mm:ssZ')

Customize & download