Using Azure Advisor for cost recommendations means treating it as a triage tool: it scans your resources against usage telemetry and surfaces a ranked list of cost actions, such as shutting down or resizing idle VMs and buying reservations for steady usage. The value is that it is free and already running. The limit is that it works resource by resource and cannot see your roadmap, so every recommendation needs validating before you act on it.
This article is part of our Azure cluster. Start with the complete guide to Azure cost optimization, the pillar this piece links up to. Advisor supports the See step in our See, Cut, Lock, Run method: it helps you find candidates, and the Cut decisions still belong to you.
Advisor is a recommendation engine, not an optimizer. It tells you a VM looks idle; it does not know that the VM is a quarterly batch host that is supposed to look idle most of the time. It tells you reserved capacity would save money on current usage; it does not know you are migrating that workload off Azure next quarter. The recommendations are a strong starting list. The judgment about which to act on is the work.
The cost recommendations Advisor surfaces
Advisor's cost category clusters around a recognizable set of actions. Knowing the categories helps you read the list quickly and route each item to the right validation.
| Recommendation | What it flags | How to validate |
|---|---|---|
| Shut down idle VMs | VMs with very low utilization | Confirm it is not a periodic or standby host |
| Resize / rightsize VMs | VMs oversized for their load | Check peak usage, not just average |
| Buy reservations | Steady usage that suits committed pricing | Confirm the workload will persist for the term |
| Buy a savings plan | Consistent compute spend across types | Compare against reservation coverage |
| Delete unused resources | Idle public IPs, unattached disks | Confirm no owner still needs them |
Step 1: Read the potential savings, then sort by effort
Each recommendation comes with an estimated saving, which makes prioritization easy: sort by impact and start at the top. But pair the saving with the effort and risk of acting. Deleting an unattached disk is high-saving and near-zero-risk, so it goes first. Buying a three-year reservation is high-saving but high-commitment, so it goes through a fuller analysis before you click buy. The estimated saving is a guide to where to look, not an instruction to act blindly.
Step 2: Validate every recommendation against real usage
The single most important habit is to never act on a recommendation without confirming the usage behind it. For a rightsize suggestion, look at peak load over a representative window, not just the average Advisor used, because a VM that averages low but spikes hard at month-end will struggle on a smaller size. We walk through that validation in how to rightsize Azure virtual machines. For an idle-VM flag, confirm it is genuinely unused and not a standby or periodic host before shutting it down, using the broader sweep in how to find idle and orphaned Azure resources.
A long Advisor list and no time to validate it?
Our Azure cost audit takes the Advisor recommendations as a starting point, validates each against real usage and your roadmap, and turns the list into a prioritized, safe action plan with the savings quantified. On the performance model, you pay only from realized savings. No savings, no fee.
Book an Azure cost audit →Step 3: Be especially careful with commitment recommendations
Advisor's reservation and savings-plan recommendations deserve extra scrutiny because they lock in spend. Advisor sizes the recommendation on recent usage, which is reasonable, but it cannot know whether that usage is about to change, whether the workload is moving, or whether you should rightsize first so you do not commit to an oversized baseline. The order matters: rightsize, then commit, so you never reserve waste. For how the two commitment instruments compare, see Azure Reservations vs Azure Savings Plan for compute.
Step 4: Use Advisor continuously, not once
Advisor refreshes as usage changes, which makes it useful as an ongoing signal rather than a one-time report. Fold a regular review of the cost recommendations into your operating rhythm so new idle resources and rightsizing opportunities are caught as they appear, rather than accumulating until the next big cleanup. That continuous use is the Run step in our method. Pair it with budgets and alerts so cost movement is caught proactively, as set out in Azure budgets and cost alerts: a setup guide.
The recommendation categories and behavior described above reflect Azure Advisor as of May 2026. Verify the current recommendation set and how savings estimates are calculated in Azure documentation before acting on high-value items, as the tooling evolves.
The Azure Cost Optimization Field Guide includes the Advisor triage workflow and the validation checklist we run before acting on each recommendation type. It is the downloadable companion to this article.
The short version
Azure Advisor is a free, built-in recommendation engine that surfaces idle, rightsize, and commitment opportunities ranked by estimated saving. Use it as triage: sort by impact, validate every recommendation against real peak usage and your roadmap, take the low-risk wins immediately, scrutinize commitment recommendations hard, and review the list continuously. To place Advisor inside a full pass, follow how to run an Azure cost optimization assessment. When you want the recommendations validated and turned into a safe action plan, that is exactly what our Azure cost optimization service delivers.