Where does NME show me Azure cost data?

Introduction

Nerdio Manager for Enterprise (NME) gives AVD administrators a clear picture of what their environment costs and where money can be saved — without requiring anyone to become an Azure billing expert. Rather than exposing raw Azure invoices, NME layers pricing data over the usage telemetry it already collects from your environment, then presents the result as actionable estimates, savings figures, and dashboards.

This article explains how NME's key cost-aware features source Azure pricing data, how they calculate and manipulate that data, and what an admin actually sees in the product.

Important

The figures NME produces are estimates and projections, not reconciled billing. They are designed to guide decisions and demonstrate value, and they are built from Microsoft's own published pricing.

1. Nerdio Advisor

Overview

Nerdio Advisor scans your environment and produces prioritized optimization recommendations, each attached to a projected cost impact. The goal is simple: show admins which changes will save the most money, so effort goes where it matters.

How costs are sourced

Advisor pulls pricing from Microsoft's publicly available Azure pricing APIs — the same open sources Microsoft publishes for its retail rates. No private billing access is required; the pricing side of the calculation comes entirely from these public feeds.

How data is calculated and manipulated

Advisor combines two inputs: the price of a resource (from Azure pricing APIs) and the observed uptime percentage of that resource in your environment. Multiplying price by observed uptime produces a projected monthly cost. Advisor then models the same resource under the recommended change and produces a second projection. The difference is the estimated impact of acting on the recommendation.

Advisor constructs a projected monthly cost by applying public pricing to observed usage (price × uptime, extrapolated across the month). This is the baseline figure everything else is measured against.

It then surfaces this through three related advisors:

  • Right-sizing — automatically analyzes actual usage data across your Azure Virtual Desktop or Windows 365 environment and recommends smaller or larger VM SKUs and host pool scaling settings.

  • Platform TCO — Aggregates these estimates into a total-cost-of-ownership view across the platform. For both AVD and Windows 365.

  • License Suitability — Assesses whether current licensing is appropriate for observed usage, flagging cases where a different licensing choice would be more cost-efficient.

What the admin sees

For each recommendation, Advisor shows a cost-before and cost-after projection. Admins can sort and prioritize recommendations by financial impact, making it straightforward to tackle the highest-value optimizations first.

2. Executive Insights

Overview

Executive Insights is a high-level dashboard built for leadership and non-technical stakeholders. It answers the question executives actually ask: "What are we getting for our AVD investment?"

How costs are sourced

Executive Insights uses the Azure Cost Management API to retrieve actual resource costs directly from Azure — not estimates derived from pricing tables. This is distinct from how Advisor and Auto-Scale Savings work.

Important

Requires customer action: Cost data is only collected for subscriptions where the customer has explicitly enabled Cost Management API read access in NME settings. If access is not granted for a subscription, Executive Insights will not show cost data for it.

NME only retrieves costs for resources it manages or is related to:

  • Resources tagged by NME (managed resources)

  • Resources that are part of NME's own infrastructure (NME-related resources)

Unrelated resources in the same subscription are excluded.

How data is calculated and manipulated

The feature aggregates actual cost and usage metrics over time, packaging them for a leadership audience. Rather than per-resource detail, it emphasizes trends, breakdowns, and summaries.

What the admin sees

  • Total spend trends over time

  • Cost breakdowns across environments

  • Savings achieved

  • Utilization summaries

3. User Cost Attribution

Overview

User Cost Attribution (UCA) is a cost report that distributes the costs of your AVD environment to users according to their usage. It supports chargeback and showback scenarios — giving departments or cost centers visibility into their share of AVD spend.

UCA builds each report by:

  • Fetching and caching all relevant data for the month into your configured storage account

  • Attributing direct VM costs to users who recorded sessions on those VMs

  • Attributing remaining "non-session costs" to all users according to the selected allocation method (unallocated, uniform, or proportional)

How costs are sourced — three information sources

UCA pulls from three distinct sources when fetching data:

#

Source

Used For

Scope

Key Notes

1

Azure Cost Management / Cost Analysis

Identifying the cost of all resources within the month

Tags and subscriptions configured in the UCA report

Historical — can be rebuilt for prior months. Tags are applied during processing, not fetching: UCA fetches full subscription costs then applies tag filters.

2

Log Analytics Workspaces (LAW)

User sessions (WVD Connections table); VM uptime (Heartbeat table)

Sessions: LAWs connected to the UCA report. VM uptime: all LAWs in subscription

Requires read permissions on each subscription's LAWs. Dependent on LAW retention (default 30 days) — check retention before reporting on older months.

3

Azure Session Host data

Identifying personal vs. pooled VMs; Host Pool membership

Subscription set in the UCA report config

Not historical by default. NME caches environment state at month-end to protect historical reports from configuration changes.

How data is calculated and manipulated

When processing the fetched data, UCA attributes costs using the following principles:

  • Personal VMs/Disks — the full cost is attributed to the assigned user.

  • Pooled VMs/Disks — costs are split between users who recorded sessions on each VM, proportional to their share of total session time on that VM. For example, if User A accounted for 50% of session time on VM 1, they receive 50% of VM 1's costs. If no sessions are found for a VM, it is treated as a non-AVD resource and categorized as 'other' costs.

  • Non-session costs — all remaining costs that cannot be directly attributed to a user are split across all users in the report according to the selected allocation method (unallocated, uniform, or proportional).

The total cost of a UCA report will always equal the total cost of the nominated subscription(s) in Azure Cost Management. If tag filters are configured, only tagged resources are included and untagged resources are excluded.

Important tagging considerations

Note

UCA bypasses Azure's historical tag context. Normally Azure Cost Management only shows tagged costs from the date tags were applied. Because UCA applies tag filters during processing (not fetching), it considers the full monthly cost of any resource that meets the tag criteria at the time of report generation — regardless of when the tags were applied. Resources must be correctly tagged at the time UCA fetches data to be included.

Additionally, not all Azure resource types support tags in Cost Management. UCA automatically includes SaaS costs (such as Nerdio licensing) within nominated subscriptions to ensure they appear in the cost report. If resources are missing from a report, verify they are correctly tagged and that the resource type supports tags in Azure Cost Management.

What the admin sees

UCA produces cost reports and dashboards within NME showing the attributed cost per user and per group. Admins can use these for internal chargeback (billing departments for their share) or showback (providing visibility without billing).

4. Auto-Scale Analytics

Overview

Auto-Scale Analytics helps admins understand how auto-scale is performing operationally. This feature does not include a financial or cost component — it is purely operational.

Tip

If you're looking for the dollar value of auto-scale, see Auto-Scale Savings below.

How data is calculated and manipulated

The feature focuses on operational telemetry — session data, scale events, and capacity utilization — to show how auto-scale behavior maps to demand.

What the admin sees

Admins see metrics on scaling activity and capacity: when scale events fired, how sessions were distributed, and how well capacity tracked demand. The purpose is to tune and validate auto-scale behavior, not quantify its cost.

5. Auto-Scale Savings

Overview

Auto-Scale Savings puts a dollar figure on the value auto-scale delivers, showing how much NME saved compared to running hosts at full capacity around the clock.

How costs are sourced

Like Advisor, Auto-Scale Savings uses Microsoft's public Azure pricing APIs for its rate data. Where possible Nerdio license costs are derived from the deployment, however this is not universally possible. Where this cannot be directly sourced, NME uses the default pricing cost for Nerdio license costs in the calculations.

How data is calculated and manipulated

NME observes the peak count of running VMs at any point in the calculation window and compares that against actual session and usage demand. It models two scenarios:

  • Peak-static cost — what the environment would have cost if VMs ran at peak capacity continuously.

  • Actual auto-scaled cost — what was actually spent with auto-scale active.

Peak-static cost − actual auto-scaled cost = savings attributed to NME auto-scale.

Note

The date window matters: Because peak VM count is a critical input, the measurement window significantly affects results. Executive Insights uses a fixed window starting from the beginning of the current calendar month, not a rolling 30-day period. A rolling window and a month-to-date window capture different peaks and produce different savings figures — this is expected and by design. The month-start window provides a consistent reference frame aligned to how costs are typically reported.

What the admin sees

Admins see a dollar-figure savings estimate that quantifies the value of auto-scale — a direct number to share with stakeholders and demonstrate platform ROI.

Limitations

NME cannot calculate figures using customers' Microsoft Reserved Instances or Microsoft Savings Plans as the product has no access to these within the existing permissions setup.

  • MS Reserved Instance Permissions

  • MS Savings Plans Permissions

Summary Comparison

Feature

Cost?

Data Source

Core Calculation

What Admins See

Nerdio Advisor

Yes

Azure public pricing APIs + observed uptime

Price × uptime; before vs. after recommendation

  • Cost projections per recommendation

  • Total cost of ownership, license suitability, and AVD/Windows 365 resizing recommendations

Executive Insights

Yes

Azure Cost Management API (customer must enable; NME resources only)

Aggregation of actual resource costs over time. ROI: [(Unoptimized Azure Spend − Optimized Azure Spend) − Nerdio SRP Licensing Costs] / Nerdio SRP Licensing Costs × 100

Spend trends, cost breakdowns, savings, utilization summaries

User Cost Attribution

Yes

AVD compute cost + session telemetry

Prorate compute cost by session duration and consumption

Per-user/group cost in reports and dashboards

Auto-Scale Insights

No

Operational telemetry

Session, scale-event, and capacity analysis

Auto-scale performance metrics (no cost)

Auto-Scale Savings

Yes

Azure public pricing APIs

Peak-static cost − actual auto-scaled cost

Dollar-figure savings estimate

Note

Cost figures produced by Nerdio Advisor, Auto-Scale Savings, and User Cost Attribution are estimates and projections based on Microsoft's published pricing and observed environment usage. They are intended to guide optimization decisions and demonstrate value, and may differ from reconciled Azure billing. Executive Insights uses actual cost data from the Azure Cost Management API where access has been granted.

Was this article helpful?

0 out of 0 found this helpful
Have more questions? Submit a request

Comments (0 comments)

Article is closed for comments.