AZ-104 eight-week study schedule, weighted by official exam domains

Last reviewed: November 21, 2026 · 12 min read

An AZ-104 plan that gives every domain the same number of weeks under-prepares identities, governance and compute, and over-prepares monitoring. This article turns the official domain weights into an example eight-week schedule. Adjust the dates to your own weekly hours; the method for doing that is in our companion guide, how to split your AZ-104 hours by exam weight.

Use the official outline, not a third-party rumor

The skills and weights below come from Microsoft's AZ-104 study guide (skills measured as of 26 October 2023, still the list published on that page when we reviewed it in November 2026). Microsoft updates outlines periodically. Check the study guide before you start, and again a week before you sit the exam.

DomainPublished weight
Manage Azure identities and governance20–25%
Implement and manage storage15–20%
Deploy and manage Azure compute resources20–25%
Implement and manage virtual networking15–20%
Monitor and maintain Azure resources10–15%

A score of 700 or greater is required to pass, per the same study guide. The exam environment can be explored in Microsoft's exam sandbox, linked from that page. Microsoft also provides a practice assessment; the study guide currently points to a newer experience on AI Skills Navigator and notes that the legacy practice assessment on Microsoft Learn is no longer being updated.

How this eight-week example is built

The example assumes eight weeks at about eight hours a week (64 hours), with the last week reserved for review and timed practice. Study hours in weeks 1–7 are split using the midpoints of the official weight ranges, scaled so they add to 100%. That is the same method as our free AZ-104 study hours calculator. If you have six weeks or twelve, or only five hours a week, use the calculator and keep the same topic order.

This is a planning example, not a promise that eight weeks is enough for every candidate. Your starting point matters: people who already administer Entra ID and virtual networks need fewer hours in those domains and more in their weak spots.

Week-by-week plan

For each skill: read the official Learn material, then build it in a lab (portal, Azure CLI or PowerShell). Tick the skill only when you have done both.

Weeks 1–2 — Manage Azure identities and governance (20–25%)

This is one of the two heaviest domains. Cover the study guide items, including:

Lab ideas: create a resource group, assign a Reader and a Contributor to different scopes, apply a built-in Azure Policy, lock a resource group, and set a budget alert on the subscription you use for labs.

Week 3 — Implement and manage storage (15–20%)

Lab ideas: create a storage account, lock it down to a virtual network or private endpoint, write a lifecycle rule, and mount an Azure file share. Delete the account when you finish if you do not need it the following week.

Weeks 4–5 — Deploy and manage Azure compute resources (20–25%)

This is the other heavy domain. It usually needs two weeks.

Lab ideas: deploy a VM, attach a data disk, place two VMs behind a public load balancer (you will reuse this in the networking week), and create an App Service with a deployment slot. Shut down or delete VMs when you are not using them.

Week 6 — Implement and manage virtual networking (15–20%)

Lab ideas: peer two virtual networks, prove name resolution, add an NSG rule that breaks and then fixes SSH or RDP, and deploy Bastion in a test network. VPN gateways and Bastion can incur charges if left running — delete the resource group when the lab is done.

Week 7 — Monitor and maintain Azure resources (10–15%), plus catch-up

This is the lightest domain by weight, but it is easy to treat as reading-only. Do the labs.

Use leftover hours this week to retake weak topics from weeks 1–6. Take Microsoft's official practice assessment at around this point, note the weakest domain, and move review time toward it.

Week 8 — Review and exam technique

Revisit every skill you rated below "confident," especially anything you read but never built. Do a timed practice assessment. Walk through the exam sandbox so the interface is not a surprise. Details are in our exam-day and practice assessment guide. Book or confirm your exam only when coverage and practice scores meet the target you set for yourself. A 700 passing score is the official bar; your practice target should be comfortably above that, because practice tests are not the real exam.

Keep lab costs under control. Use an Azure free account or an employer sandbox, put each lab in its own resource group, set a budget alert, and delete the resource group when you finish. Virtual machines, public IP addresses, load balancers, VPN gateways, Bastion and Site Recovery can generate charges if they are left running. Microsoft's current free-account offer is described at azure.microsoft.com/en-us/free; terms change, so read the offer before you rely on credits.

Do not use "exam dumps." They violate Microsoft's exam agreement and teach memorized answers instead of the skills the exam (and the job) measure.

Track coverage, not just hours

A schedule only works if you record what you actually finished. Rate each skill from 1 to 5 at the start and as you go. You are closer to ready when nearly every skill has a lab, your confidence averages 4 or more, and official practice scores are at your target.

The AZ-104 Study Planner & Tracker turns this method into a workbook: an automatic weekly schedule from your dates and hours, an 81-skill tracker aligned to the official outline, a study log, a practice-test log and a readiness dashboard.

Always re-check the current AZ-104 study guide. If Microsoft changes a domain weight or adds a skill, update the schedule rather than following this example blindly.

Sources (reviewed November 2026): Microsoft Learn, Study guide for Exam AZ-104: Microsoft Azure Administrator; Microsoft Certified: Azure Administrator Associate; Cost Management alerts. Always follow the current version of those articles.

This guide was drafted with AI assistance and checked against public Microsoft documentation. How we use AI. Spotted an error? Email [email protected].