How Azure Monitor Advanced Metrics Reduce Risk and Cloud Costs

In this blog post How Azure Monitor Advanced Metrics Reduce Risk and Cloud Costs we will explain how deeper resource-level monitoring can help you find unexpected growth, investigate problems sooner and make better cloud spending decisions.

Most Azure cost problems do not begin with a dramatic outage. They start quietly: a storage container grows each week, an application keeps data longer than expected, or an abandoned project continues consuming capacity. Standard reports may show that the overall bill is rising without showing exactly what is responsible.

Azure Monitor advanced platform metrics are designed to close that visibility gap. They add more detailed measurements to selected Azure services, allowing your team to see what is happening inside a resource rather than treating it as one large, unexplained cost.

What advanced platform metrics actually do

Azure Monitor is Microsoft’s service for collecting measurements about the health, performance and usage of Azure resources. A metric is simply a number recorded over time, such as storage consumed, requests processed or processor usage.

Azure already collects standard platform metrics automatically for most resources. Advanced platform metrics are an optional, paid tier that provides more detailed measurements for supported services. They use the same Azure Monitor tools, dashboards and alerting processes your team may already know.

At the time of writing, advanced platform metrics are in public preview. Azure Blob Storage, which businesses commonly use for documents, backups, application files and data archives, provides a useful example of what the technology can do.

Instead of seeing only the total capacity of a storage account, you can measure the capacity and number of files in individual containers. A container is a logical area used to organise stored data, similar to a top-level folder.

• Container Blob Capacity shows how much storage each container is consuming.
• Container Blob Count shows how many stored objects, or files, each container contains.
• ContainerName is a dimension, meaning a label that lets you filter or compare the results by container.

The measurements include current files, previous file versions and soft-deleted files that Azure is temporarily retaining. That detail matters because retained and deleted data can still contribute to storage growth.

Why account-level monitoring is often not enough

Imagine a storage account increasing from 24 terabytes to 31 terabytes over two months. The finance team sees the higher invoice, but the account contains production documents, backups, application exports and an AI testing project.

Standard account-level monitoring confirms that storage has grown. It does not necessarily make the responsible workload obvious. The IT team may spend hours asking application owners for information or running separate reports.

With container-level advanced metrics, the team could discover that an archive container added 5.5 terabytes while the other containers remained stable. Further investigation might reveal that an application update stopped deleting temporary export files.

The business outcome is not merely a better graph. It is a shorter investigation, clearer ownership and the chance to correct the issue before another billing cycle passes.

Use advanced metrics for earlier cost detection

Advanced platform metrics should not be confused with real-time application monitoring. Changes to Azure Blob Storage’s container-level capacity measurements can take several hours to appear, particularly in large storage accounts.

That means they are not the right signal for detecting a website outage in seconds. Their strength is detecting capacity growth and cost patterns much earlier than a monthly invoice or quarterly cloud review.

A practical monitoring setup might include:

1. A dashboard showing the largest storage containers.
2. A weekly view showing which containers are growing fastest.
3. An alert when a high-value container passes an agreed capacity threshold.
4. An owner for each monitored workload who receives the alert.
5. A regular review of old versions, soft-deleted files and retention settings.

Metrics tell you where growth is occurring. Ownership information tells you who needs to act. This is why advanced monitoring works best alongside Azure tags and budgets, which connect cloud resources to departments, projects and cost centres.

Enable only the detail your business needs

Because advanced platform metrics are paid and must be enabled deliberately, switching them on everywhere is rarely the best starting point. Begin with storage accounts where unexplained growth, shared ownership or compliance requirements create a genuine business risk.

Azure allows the monitoring rule to cover all containers, a list of named containers or containers with a particular naming prefix. A prefix is simply the shared beginning of a name, such as finance- or production-.

The following Bicep example shows how an Azure team could enable the feature for containers beginning with production-. Bicep is a Microsoft configuration language used to deploy Azure settings consistently rather than configuring every resource manually.

resource storageAccount 'Microsoft.Storage/storageAccounts@2026-01-01' existing = {
 name: 'yourstorageaccount'
}

resource advancedMetrics 'Microsoft.Storage/storageAccounts/advancedPlatformMetrics@2026-09-01' = {
 parent: storageAccount
 name: 'ContainerLevelCapacityMetrics'
 properties: {
 enabled: true
 ruleConfig: {
 filterType: 'ContainerPrefixFilter'
 filterValues: [
 'production-'
 ]
 }
 }
}

This example limits the additional monitoring to containers that matter to production operations. Your Azure team should test the configuration and confirm current preview availability, pricing and deployment requirements before using it broadly.

Turn measurements into decisions

A common monitoring mistake is collecting more data without deciding what anyone should do with it. A dashboard that nobody reviews simply creates another cost.

For every advanced metric you enable, define the business question it is meant to answer. Useful questions include:

• Which application or department is driving storage growth?
• Are backup, retention or file-versioning settings keeping unnecessary data?
• Is an AI or analytics project producing data faster than planned?
• Should older data move to a lower-cost storage tier?
• Is the organisation approaching a capacity, budget or contractual limit?

Alerts should also lead to a clear action. A storage growth alert might create a service ticket, notify the resource owner and request a review of retention settings. If an alert merely sends another email to an overloaded shared inbox, it is unlikely to reduce cost or risk.

For AI workloads, combine infrastructure monitoring with the usage, quota and model-level controls described in our guide to monitoring Azure AI services. Organisations operating multiple AI deployments may also benefit from a tenant-wide Microsoft Foundry operations dashboard.

Do not treat monitoring as a security control

Detailed metrics can expose unusual growth or activity, but they do not replace cybersecurity monitoring. Microsoft Defender helps identify threats across Microsoft environments, while Wiz provides visibility into cloud risks, exposed resources and dangerous configuration combinations.

Advanced metrics also do not make a business compliant with Essential 8, the Australian Government’s cybersecurity framework that many organisations use to reduce common attack risks. They can, however, support stronger operational accountability by showing that critical cloud resources are monitored, reviewed and assigned to responsible owners.

The most effective approach connects Azure Monitor with cost management, Defender, Wiz and documented response processes. That gives decision-makers one operating model for reliability, spending and security rather than several disconnected dashboards.

A sensible place to begin

Start with one storage account that is expensive, growing quickly or shared by several teams. Enable metrics only for the containers that matter, establish a baseline and review the results for several weeks.

Then decide whether the extra visibility is producing a measurable outcome: faster investigation, deleted waste, improved forecasting or clearer accountability. Our earlier article on Azure Monitor advanced metrics for faster detection and cost control provides additional guidance on connecting these signals to broader cost decisions.

CloudProInc is a Melbourne-based Microsoft Partner and Wiz Security Integrator with more than 20 years of enterprise IT experience. We help organisations across Australia and internationally build practical monitoring, cost management and security controls without turning the project into a giant consulting exercise.

If you are unsure what is driving your Azure growth, or whether additional monitoring will pay for itself, we are happy to take a practical look at your current setup — no strings attached.