Do we need cost analysis turned on for every client cluster?

Yes. Kubernetes cost views in Microsoft Cost Management only show clusters with the AKS cost analysis add-on enabled. The views work at subscription scope, support Enterprise Agreement and Microsoft Customer Agreement subscriptions, and need a role such as Reader or Cost Management Reader.

Can we stop clients' dev and test clusters out of hours?

Yes. Stopping an AKS cluster shuts down its control plane and agent nodes, so that compute stops being billed, and AKS keeps the cluster state for up to 12 months. Standalone pods are deleted. Microsoft advises against stopping mission-critical clusters, because a restart can fail in capacity-constrained regions.

Is there a quick first check for a new client's clusters?

Azure Advisor includes AKS cost recommendations by default, filtered to each cluster's configuration and usage. They cover things like Vertical Pod Autoscaler recommendation mode and faster autoscaler scale-down. Viewing them needs Owner, Contributor or Reader access.

Coverage

AKS cost optimisation for MSPs

AKS spend usually leaks in two places: pod CPU and memory requests set well above real usage, which leaves nodes overprovisioned, and small dev and test clusters left running unattended. Without clear visibility, that waste is hard to spot and harder to assign to the team that owns it.

42

automated finders for AKS

Read-only, across every connected client tenant, tied to real billing data.

Modernise
Rightsizing
Scaling
Off-hours schedules

Where AKS spend goes to waste

The main cost levers Microsoft documents for AKS, and the ones an MSP should check across every client.

Rightsize pod requests

When requests and limits sit above actual usage, nodes fill with capacity that is reserved but never used. The Vertical Pod Autoscaler suggests CPU and memory values from each container's usage history, and its recommendation-only mode lets you review them before applying them in production. Don't pair it with the Horizontal Pod Autoscaler on the same CPU or memory metrics. ^[1]

Let the cluster autoscaler remove idle nodes

The cluster autoscaler adds nodes when pods cannot be scheduled and removes underutilised nodes once their workloads can move elsewhere. Where cost matters most, Microsoft recommends tuning the autoscaler profile to scale down faster. ^[2]

Use Spot node pools for interruptible work

Spot virtual machines run on spare Azure capacity at discounts of up to 90% on pay-as-you-go prices. Azure can evict them whenever it needs the capacity back, so they suit workloads that cope with interruption. ^[3]

See costs by cluster and namespace

The AKS cost analysis add-on breaks spend down by clusters and namespaces alongside Azure compute, network and storage, and reconciles the figures with the Azure invoice. It also reports idle charges for capacity no workload is using. It needs the Standard or Premium pricing tier. ^[4]

Commit for steady workloads

For predictable, long-running workloads, Azure Reservations on one- or three-year terms can cut compute costs by up to 72% compared with pay-as-you-go. Where usage is steady but spread across VM sizes and regions, an Azure savings plan for compute commits to a fixed hourly spend instead. ^[1]

Cut monitoring and logging costs

Microsoft recommends moving metrics from Container insights to Azure Monitor managed service for Prometheus to reduce cost. Switching off control plane log categories you don't need, using Basic Logs where suitable and filtering logs before they reach Log Analytics all help too. ^[1]

How Cerberus Cloud covers AKS

Cerberus Cloud runs 42 automated finders for AKS across every connected client tenant. Each one ties a technical signal to real billing data, so findings arrive with an estimated recoverable value, ranked by impact, effort and risk, with evidence and remediation scripts for your engineers to review.

Access is read-only, so nothing changes in a client environment until your team decides it should. See how it works or the multi-tenant portfolio view.

Modernise
Rightsizing
Scaling
Off-hours schedules

Frequently asked questions

Do we need cost analysis turned on for every client cluster?: Yes. Kubernetes cost views in Microsoft Cost Management only show clusters with the AKS cost analysis add-on enabled. The views work at subscription scope, support Enterprise Agreement and Microsoft Customer Agreement subscriptions, and need a role such as Reader or Cost Management Reader.
Can we stop clients' dev and test clusters out of hours?: Yes. Stopping an AKS cluster shuts down its control plane and agent nodes, so that compute stops being billed, and AKS keeps the cluster state for up to 12 months. Standalone pods are deleted. Microsoft advises against stopping mission-critical clusters, because a restart can fail in capacity-constrained regions.
Is there a quick first check for a new client's clusters?: Azure Advisor includes AKS cost recommendations by default, filtered to each cluster's configuration and usage. They cover things like Vertical Pod Autoscaler recommendation mode and faster autoscaler scale-down. Viewing them needs Owner, Contributor or Reader access. Where Advisor fits for MSPs

See AKS findings on a sample portfolio

A 15-minute walkthrough: tenant connection, prioritised findings, remediation scripts and savings history.

Book a Walkthrough All coverage

Sources

Last checked 15 October 2026. Product details change; tell us if something here is out of date.

All 15 coverage families