Microsoft Azure Solutions Architect Expert [AZ-305] Exam Guide (2026)

If you can say whether a workload should sit on a virtual machine, a container platform, or Functions, whether the data is a relational ledger or a document store, and whether the stem is buying backup, high availability, or Site Recovery, you are reading the right outline.

AZ-305 is the exam behind Microsoft Certified: Azure Solutions Architect Expert. The credential requires Azure Administrator Associate first. The exam page names the sitting Designing Microsoft Azure Infrastructure Solutions, marks the English language version updated on 17 April 2026, and publishes a passing score of 700. The study guide publishes the matching skills outline under the heading Skills measured as of April 17, 2026. That outline is the whole scope of the current sitting.

The sitting is live. Retirement date is none. The exam page publishes a Pearson VUE schedule path and a free practice assessment. Expert certifications expire annually and renew on a free, unproctored Microsoft Learn assessment. This guide is for people scheduling that April 2026 blueprint, who already hold the title and need the current task list, or who are moving from Azure administration into the architect role.

Who this exam is for

Take it as a map of the Azure solutions-architect job Microsoft scores under this code. The audience profile asks for subject matter expertise designing cloud and hybrid solutions that run on Azure. The five named areas are compute, network, storage, monitoring, and security.

The responsibility list is concrete. Advise stakeholders. Translate business requirements into designs that align with the Azure Well-Architected Framework and the Cloud Adoption Framework for Azure. Implement those designs by partnering with developers, administrators, security engineers, and data engineers. Microsoft is explicit that the candidate should have advanced experience in networking, virtualization, identity, security, business continuity, disaster recovery, data platforms, and governance, and should manage how a decision in one area changes the whole solution. Azure administration, Azure development, and DevOps processes are named as expected experience, not as optional reading.

The prerequisite is a hard gate. Microsoft Certified: Azure Solutions Architect Expert requires Microsoft Certified: Azure Administrator Associate. If subscriptions, role-based access, Policy, virtual machines, storage accounts, and virtual networks are still a catalog rather than weekly work, sit AZ-104 first. The architect sitting assumes those controls and then asks which design to recommend.

Skip it if the goal is the language of the cloud rather than the design of it. Service definitions, pricing models, and shared responsibility belong to AZ-900. Skip it if the work you actually want is running someone else's subscription day to day without owning the compute shape, the data model, or the recovery objective, because that job is AZ-104. Skip it if the work you actually want is hardening identity, compute, and posture without owning the solution tradeoff across pillars, because that job is AZ-500. Skip it if the work you actually want is picking the VPN SKU, the Front Door origin, or the Private Link alias, because that job is AZ-700. Those controls appear on this outline as recommendations. They are scored there as the whole job.

The current expert architect credential Microsoft publishes is this one. Use this page for the AZ-305 task list. Use the AZ-104 exam guide when the stem is an administrator control that happens to sit on a workload. Use the AZ-500 exam guide when the stem is a security control that happens to sit on a design. Use the AZ-700 exam guide when the stem is a network-engineer control that happens to sit on a recommendation.

Exam shape and how the score works

The April 2026 outline has four functional groups, and Microsoft publishes weight ranges rather than fixed percentages.

Do the arithmetic on those ranges before building a plan. The lower bounds total 90%. The upper bounds total 110%. The midpoints total 100%. No exact split exists to memorize, and the bands are the only ordering Microsoft gives you. Infrastructure sits alone at the top. Identity, governance, and monitoring sit alone in the upper middle. Data storage sits alone in the lower middle. Business continuity sits alone at the bottom. Any study plan that gives four equal weeks overweights backup and underweights compute, application architecture, migrations, and network design.

The logistics come off the credential page, the exam page, and the duration table. The sitting is scheduled through Pearson VUE. Price depends on the country or region where the exam is proctored, so no single global figure is published. Languages are English, Japanese, Chinese (Simplified), Korean, German, French, Spanish, Portuguese (Brazil), Chinese (Traditional), and Italian. A failed attempt can be retaken 24 hours later, and the wait varies for later retakes. Microsoft tells candidates to register with a personal MSA account, because an organizational account loses the exam record if the person leaves that organization.

The AZ-305 credential page and exam page opened for this guide do not publish a minutes figure. Microsoft's duration table lists associate and expert role-based exams without labs at 100 minutes of exam time and 120 minutes of seat time, and lists exams that may contain labs at 120 minutes of exam time and 140 minutes of seat time. Microsoft states plainly that it does not publish a list of exams with labs, because labs can be pulled at any time for an outage or a bandwidth problem, and that the real exam time is confirmed at registration and on the launch screens. Plan from the duration table. Do not invent a timer the exam page did not give you.

Scoring is the part most candidates get wrong. Technical exam scores are reported on a scale of 1 to 1,000 and 700 or greater passes. Microsoft says the scaled score may not equal 70% of the points. A passing score is based on the knowledge and skills needed to demonstrate competence as well as the difficulty of the questions. Multi-part questions usually award one point per correctly answered component, so all, some, or none of the points on a question are possible. There is no penalty for guessing. Some questions are unscored and used to collect data, and candidates are never told which, so every question deserves an answer.

The score report gives one overall number, a pass or fail status, and a bar chart per skill area. It does not give a numeric score per section, and Microsoft warns that the bars cannot be added up to reconstruct the result. A short bar in a small area can mean a handful of questions went wrong, and scoring zero in an area that carries only a few questions is documented as normal. Business continuity at 15 to 20% is the area where a short bar is easiest to misread as a catastrophe.

Two more details change how the sitting is taken. Microsoft Learn is available in a split screen during role-based exams, covering everything on learn.microsoft.com except Q&A, practice assessments, and your profile, with no extra time added and no navigation outside the domain. Five minutes of break time are built into the clock, questions were removed to make room for it, and any question you have already seen is gone once the break starts.

The credential itself follows the expert renewal rule. Expert certifications expire annually. Renewal is free, unproctored, open book, and shorter than the original exam, inside a six month window before expiry. Passing extends the certification one year from the expiration date.

What changed on 17 April 2026

The study guide publishes a change log comparing the previous outline with the current one. The full table is one row.

Three things are true of that table, and each one changes what you should do with older study material.

The only row is Audience profile, and it is marked Minor. Nothing on the table is marked Major. The exam code did not move, the four functional groups did not move, and the published weight ranges are the ones above. AZ-305 material written against the previous outline is still structurally correct.

The table carries no row at all for identity, governance, and monitoring, for data storage, for business continuity, or for infrastructure. Those are the areas that decide the compute shape, the data model, the recovery objective, and the landing-zone structure. They behave on the current exam the way they behaved on the previous one, which makes those areas the cheapest place to bank points and the least excusable place to lose them.

The English-language update date on the credential page and the exam page is 17 April 2026, the same date the study guide uses for skills measured. Use that date for the outline. Localized versions update about eight weeks later. If the exam is not available in a preferred language, Microsoft lets the candidate request an additional 30 minutes.

How a compute shape gets picked

Infrastructure is 30 to 35%, the change log skipped it, and the same discriminator shows up in migration stems, application-architecture stems, and high-availability stems as often as in compute ones.

Microsoft's compute guide sorts the services on the hosting model and on the skill the team already has. Read the control the stem is buying before reading the options.

Azure Virtual Machines are infrastructure as a service. Microsoft manages the fabric. The customer manages the guest, the patches, and the network path inside the virtual network. The compute decision tree names lift and shift as the like-for-like move that defers optimization, and it lands many of those workloads on virtual machines or on Azure VMware Solution when the estate is already VMware. Operational overhead is high. A stem that wants full control of the operating system, a line-of-business installer that cannot be containerized, or a VMware private cloud that must keep vCenter, is a virtual-machine stem.

Azure App Service is platform as a service for web apps, mobile back ends, REST APIs, and WebJobs. The platform owns the instances. Deployment slots are the built-in application-update path. Operational overhead is low. A stem about a regional web app whose team writes code and does not want to patch VMs is App Service.

Azure Functions is functions as a service. Triggers run the code. Consumption and Flex Consumption are the serverless plans. Durable Functions are how that code keeps state. Operational overhead is very low. A stem about an event, a queue message, or a timer that should cost nothing when idle is Functions. A stem about a long-running web request that needs a warm process is not.

Containers split three ways on this outline, and mixing them is how the question is lost. Azure Container Instances run a single container or a group, with no orchestrator and no dedicated nodes. Azure Container Apps run containers on Kubernetes without handing over the Kubernetes API. Azure Kubernetes Service hands over the API and the control plane, recommends six nodes in production, and expects a team that already operates Kubernetes. Azure Red Hat OpenShift is the answer when the investment is already OpenShift. A stem that asks for kubectl and cluster add-ons is AKS. A stem that wants scale-to-zero containers without cluster operations is Container Apps. A stem that wants one container for a short job is Container Instances.

Azure Batch is the named batch-processing recommendation. It runs large parallel and high-performance jobs and can scale the pool to zero when the job finishes. A stem about nightly rendering, financial simulation, or a thousand tasks that should vanish afterward is Batch. Putting that work on an always-on virtual-machine scale set is the trap.

Two more facts change how those answers are used. A complete solution can include two compute services. Microsoft says so on the decision tree. All of the platforms in that guide are regional, so a multi-region design needs an external router. Front Door, Traffic Manager, or a load balancer in front of two regional stamps is a second recommendation, not a reason to treat the compute services as interchangeable.

Read the constraint the stem is buying. An existing server that must move with the guest intact is a virtual machine. A web app the team will only deploy is App Service. An event that should run and disappear is Functions. A container whose team must keep the Kubernetes API is AKS. A container whose team must not keep that API is Container Apps. A thousand parallel jobs that should scale to zero are Batch.

How a data store gets picked

Data storage is 20 to 25%, the change log skipped it, and most of its questions reduce to one skill. Name whether the stem is about a relational ledger, a flexible document, or a binary object, then name the service that implements that model.

Microsoft's data-store guide says a single store rarely satisfies every access pattern, and that most production systems pick more than one model. Choose the model first. Shortlist services second.

Azure SQL Database is the managed SQL Server engine for cloud-native applications. Elastic pools, Hyperscale, and built-in high availability are the reasons Microsoft lists it. Azure SQL Managed Instance is the near-complete SQL Server environment for a lift-and-shift that still needs SQL Agent, cross-database queries, or a virtual-network injection. A stem about a new SaaS catalog on SQL is SQL Database. A stem about an on-premises SQL Server that must move with minimal code change is Managed Instance. Azure Database for PostgreSQL and Azure Database for MySQL are the answers when the engine itself is the constraint, PostGIS for the first and a MySQL application for the second.

Service tier is its own question on the relational half of the outline. General Purpose separates compute from storage and can tolerate a cold cache after failover. Business Critical keeps data on locally attached SSD and uses a replica set similar to Always On, with Read Scale-Out as the free read offload. Hyperscale is the distributed tier for very large databases and rapid backups. Basic and Standard, on the DTU model, have local redundancy and do not offer zone redundancy. A stem that names zone redundancy on a DTU Basic database is testing that table.

Semi-structured data is a Cosmos DB or DocumentDB stem more often than a SQL stem. Azure Cosmos DB for NoSQL is the schema-less, multi-region document store with a SQL-like query language, TTL, and change feed. Azure DocumentDB is the MongoDB wire-protocol path. Azure Managed Redis is the in-memory key-value path for cache, session state, and publish-subscribe. A stem about a product catalog whose schema changes per tenant is Cosmos DB. A stem about sub-millisecond cache in front of that catalog is Redis. Putting the catalog in Blob Storage because the documents are JSON is the trap.

Unstructured data is Blob Storage, Data Lake Storage, or Azure Files. Blobs hold objects. Data Lake Storage is the analytics lake on top of that object store. Files are the managed SMB share. The outline then asks for the balance of features, performance, and cost, which is hot against cool against archive on the object side, and for protection and durability, which is LRS against ZRS against GRS on the account. A stem about images that are read once a year is archive. A stem about a lift-and-shift file share is Azure Files.

Data integration and data analysis are the last two bullets in the storage group. They are how the architect moves data between those stores and how a team reads it afterward. Microsoft Fabric, Azure Data Explorer, and Azure Databricks sit on the analytics side of the data-store guide. A stem about near-real-time telemetry with Kusto is Data Explorer. A stem about a lakehouse that other Fabric workloads will share is Fabric. They are not the answer when the stem is still trying to pick the system of record.

How a continuity control gets picked

Business continuity is 15 to 20%, the lightest band, the change log skipped it, and the question still shows up in compute stems, SQL stems, and storage stems.

Microsoft files Azure Backup and Azure Site Recovery under Azure Recovery Services, and each one has its own job. Read the recovery objective the stem names before reading the options.

Azure Backup protects data and restores it. Microsoft states that it focuses on data protection, and that disaster recovery and failover orchestration belong to Site Recovery. Use Backup when the stem wants protection against deletion, corruption, or ransomware, short-term or long-term retention, a central policy across workloads, or a restore of an item or a workload. Recovery Services vaults and Backup vaults store the recovery points. Policies set schedule and retention. Soft delete, immutable backup, and role-based access are the named controls that keep those points from disappearing. Backup storage itself is LRS, ZRS, or GRS. GRS is the default Microsoft recommends. LRS is the low-cost copy inside one datacenter. ZRS keeps the copies in availability zones in the same region.

Azure Site Recovery keeps the application running by replicating a machine from a primary site to a secondary location, failing it over, and failing it back. It replicates Azure virtual machines between regions, machines from Azure Extended Zones into the connected region, and on-premises VMware, Hyper-V, Azure Stack, and physical servers into Azure. Continuous replication covers Azure VMs and VMware VMs. Hyper-V can replicate as often as every 30 seconds. Recovery plans sequence a multi-tier application. Test failovers do not break ongoing replication. A planned failover is the zero-data-loss path for a scheduled outage. An unplanned failover accepts the data loss that the replication frequency already implied.

High availability is the third control, and it is not a synonym for either of the first two. Availability Zones, zone-redundant SQL, and a load balancer in front of two virtual machines keep the service up inside the failure domain the stem named. Azure SQL Database defaults to local redundancy inside one datacenter, with an RPO of zero for committed data. Zone redundancy is the extra switch that survives a zone outage. Without that switch, a zone outage becomes a disaster-recovery problem, and the recover path is active geo-replication, a failover group, or a geo-restore of a geo-redundant backup. Business Critical adds locally attached SSD and readable secondaries. General Purpose fails over onto a new compute node and can start cold.

The official split is short. A deleted file, a corrupted database, or a ransomware event that needs last night's copy is Backup. A region that went dark and must run the same virtual machines somewhere else is Site Recovery. A zone that failed and must not take the application down is high availability. RPO is how much data the business can lose. RTO is how long the business can wait. A stem that names both numbers is asking you to pick the control that can meet both, not the brand you studied last.

How to study the April 2026 blueprint

Order the weeks by the weight bands, not by the order the areas are printed in.

Week 1. Infrastructure, the 30 to 35% area. Build one workload three ways. Put the same API on a virtual machine, on App Service, and on Functions, and write down which operating-system task, which deployment slot, and which idle-cost behavior each one removed. Deploy one container on Container Instances, one on Container Apps, and a three-node AKS cluster, and confirm that only AKS gives you the Kubernetes API. Create a Batch pool, submit a job, and watch the pool scale to zero. Then design the application around those hosts. Put a queue between a web front end and a worker. Put an Event Grid topic in front of the Functions app. Put API Management in front of the API only when the stem is an API gateway, not when it is a general load balancer. Put Azure Managed Redis in front of a read-heavy page. Store app settings in App Configuration or Key Vault, and deploy the app from a pipeline rather than from a laptop. Finish the week on the network recommendation. Draw VPN against ExpressRoute against Virtual WAN for the hybrid path, and Load Balancer against Application Gateway against Front Door against Traffic Manager for delivery. Use the AZ-700 exam guide when a stem goes one level deeper into SKUs.

Week 2. Identity, governance, and monitoring, the 25 to 30% area. Create a management group tree under the tenant root, hang two subscriptions under it, and assign one Policy and one role at the management group so both subscriptions inherit. Write down that the root cannot be moved or deleted, that new subscriptions land there by default, and that a Policy on the root hits every resource in the directory. Tag a resource group and use that tag in a Policy. Then pick an identity path. Create a cloud-only user. Walk password hash synchronization, pass-through authentication, and federation until you can name the extra infrastructure each one adds. Microsoft names Cloud Sync as the future of synchronization and the replacement for Connect. Connect V1 is retired. Finish the week in Key Vault and Monitor. Create a vault, store a secret, a key, and a certificate, and assign access with Azure RBAC rather than with a vault access policy. Create a Log Analytics workspace and an Azure Monitor workspace and write down that they are different resources with different query languages, KQL against PromQL. Turn on Application Insights on the App Service from week 1. Write one metric alert and one autoscale rule.

Week 3. Data storage, the 20 to 25% area. Create Azure SQL Database on General Purpose and on Business Critical, enable zone redundancy where the table allows it, and fail over a failover group to a second region. Create a Managed Instance in a virtual network and write down the SQL Server feature that forced that choice. Create a Cosmos DB for NoSQL account with a second region, set a TTL, and read the change feed. Create a storage account and put the same payload in hot, cool, and archive. Create an Azure Files share and a Data Lake container. Apply a service endpoint to one account and a private endpoint to the other, then use the AZ-104 exam guide for the administrator form of that pair and the AZ-700 exam guide for the provider side. Open Microsoft Fabric or Azure Data Explorer only long enough to say which one is the analysis answer for a lakehouse and which one is the analysis answer for high-ingest telemetry.

Week 4. Business continuity, the 15 to 20% area, and a full review. Create a Recovery Services vault. Back up a virtual machine, a SQL database, and a file share. Set soft delete and an immutable policy. Create a Site Recovery vault, enable replication of a test virtual machine to a second region, run a test failover, then a planned failover. Enable zone redundancy on the Business Critical database from week 3 and confirm that Basic DTU still cannot take that switch. Spend the rest of the week rereading the study guide task statements and naming, for each bullet, the Well-Architected pillar the recommendation serves and the Cloud Adoption Framework phase that would have produced the landing zone around it.

Take the free practice assessment linked from the exam page and run the exam sandbox at least once before using this outline against a clock. The sandbox exists so the question formats cost you nothing on the timer.

Traps that look like easy elimination

Expert items rarely give one plausible answer and three absurd ones. They give two controls that both sound correct and one detail that picks between them.

• Azure Administrator Associate is required for the Expert credential. Passing AZ-305 without that associate certification does not complete the title.
• The Well-Architected Framework is the workload-quality guide. The Cloud Adoption Framework is the portfolio and landing-zone guide. A stem about pillar tradeoffs is Well-Architected. A stem about a platform landing zone is Cloud Adoption Framework.
• A platform landing zone is the shared foundation. An application landing zone is where one workload lives. Most organizations have one platform landing zone per Microsoft Entra tenant.
• Management groups sit above subscriptions. Policy and role assignments inherit down. The tenant root cannot be moved or deleted, and a Policy there hits the whole directory.
• Password hash synchronization copies a hash to Microsoft Entra ID. Pass-through authentication checks the on-premises password without a federation farm. Federation is AD FS. Cloud Sync is the path Microsoft names as the replacement for Connect.
• Key Vault Standard is software-protected keys. Premium is HSM-protected keys at FIPS 140-3 Level 3. Authentication is Entra ID. Azure RBAC can govern the vault and the data. A vault access policy can govern only the data.
• A Log Analytics workspace speaks KQL. An Azure Monitor workspace speaks PromQL. They are not the same resource.
• A lift-and-shift that must keep the guest is a virtual machine. A web app the team will only deploy is App Service. An event that should go idle at zero cost is Functions.
• Container Instances have no orchestrator. Container Apps hide the Kubernetes API. AKS exposes it. OpenShift is the answer only when the estate is already OpenShift.
• Azure Batch is the parallel-job answer. An always-on scale set is the wrong place to park a thousand tasks that should disappear.
• All of the compute platforms in Microsoft's decision guide are regional. Multi-region reach needs Front Door, Traffic Manager, or another external router.
• Azure SQL Database is the cloud-native engine. SQL Managed Instance is the lift-and-shift engine. PostgreSQL and MySQL are answers when the stem names that engine.
• General Purpose SQL can start cold after failover. Business Critical keeps replicas on local SSD and can offload reads. Basic and Standard DTU cannot take zone redundancy.
• Cosmos DB is the document model. Blob Storage is the object model. JSON in a blob does not make that blob a document database.
• Azure Backup restores data. Azure Site Recovery fails a machine over. High availability keeps the service up inside one region or zone. They are three answers.
• GRS is the default backup-storage recommendation. LRS is one datacenter. ZRS is one region across zones.
• A planned Site Recovery failover is the zero-data-loss path. An unplanned failover accepts the lag the replication frequency already set.
• Azure Load Balancer is Layer 4. Application Gateway is regional Layer 7. Front Door is global Layer 7. Traffic Manager is DNS and does not see the packet.
• API Management is an API gateway. It is not the AZ-305 answer for a general load-balancing stem.
• A service endpoint covers every instance of a service and keeps public DNS. A private endpoint covers one instance and needs a private DNS record. The AZ-104 and AZ-700 exam guides teach the same split.

If deleting the scenario still lets you pick the answer from the service name, the question is easier than the live exam.

How this maps to CloudFluently

Start with the official material. The Microsoft Certified: Azure Solutions Architect Expert credential page carries the audience profile, the AZ-104 prerequisite, and the renewal path. The Exam AZ-305 page carries the 17 April 2026 English update, the 700 passing score, the practice assessment, and the exam sandbox. The study guide for Exam AZ-305 carries the task statements, the weight ranges, and the change log quoted above. Exam duration and exam experience explains the expert timer, the built-in break, and the Microsoft Learn split screen. Exam scoring and score reports explains the 700 scaled pass mark and why that scaled score may not equal 70% of the points. Microsoft Certification renewal explains the annual free assessment.

AZ-305 sits on top of Azure fundamentals and Azure administration, and that ground is already live here. Shared responsibility, regions, and the management-group idea are the AZ-900 Azure Fundamentals study notes and the AZ-900 Azure Fundamentals practice exam sets. The Azure Fundamentals roadmap sequences that ground, and the AZ-900 exam guide is the first companion page. Subscriptions, RBAC, Policy, virtual machines, storage accounts, and the administrator form of service endpoints against private endpoints are the AZ-104 exam guide. Relational against non-relational stores, and the vocabulary behind Blob, SQL, and Cosmos DB, are the DP-900 Azure Data Fundamentals study notes and the DP-900 Azure Data Fundamentals practice exam sets. Workload vocabulary that shows up in application-architecture stems is the AI-900 Azure AI Fundamentals study notes and the AI-900 Azure AI Fundamentals practice exam sets.

The security-engineer and network-engineer readings of the same controls are already live as well. Key Vault, identity, and posture are the AZ-500 exam guide. VPN against ExpressRoute against Virtual WAN, and Load Balancer against Application Gateway against Front Door against Traffic Manager, are the AZ-700 exam guide.

Work those until the platform, the administrator controls, the data vocabulary, the security controls, and the network recommendations are automatic, then use this page and the official study guide for the architect-only skills: the compute shape, the data model, the continuity control, and the landing-zone structure that holds them.

Frequently Asked Questions

What is the passing score for AZ-305? 700 or greater on a scale of 1 to 1,000. Microsoft states that this is a scaled score and that it may not equal 70% of the points, because a passing score is based on the knowledge and skills needed to demonstrate competence as well as the difficulty of the questions.

How long is the exam and how many questions are there? The AZ-305 credential page and exam page do not publish a minutes figure. Microsoft's duration table lists expert role-based exams without labs at 100 minutes, and exams that may contain labs at 120 minutes. The exam time is confirmed at registration and on the launch screens. Microsoft does not publish a question count for any individual exam, and says most certification exams typically contain between 40 and 60 questions.

Does AZ-305 have labs? Microsoft does not publish a list of exams with labs, because labs can be removed at any time for outages or bandwidth issues. The exam time is confirmed at registration and on the launch screens.

Does AZ-305 complete the Expert credential on its own? No. Microsoft Certified: Azure Solutions Architect Expert requires Microsoft Certified: Azure Administrator Associate and Exam AZ-305.

What changed on 17 April 2026? The change log has one row. Audience profile is marked Minor. The four functional groups and the published weight ranges did not move.

Which skill area is heaviest? Design infrastructure solutions is the top band at 30 to 35%. Design identity, governance, and monitoring solutions sits at 25 to 30%. Design data storage solutions sits at 20 to 25%. Design business continuity solutions is lowest at 15 to 20%.

Why do the percentages not add up to 100? Because Microsoft publishes ranges. The lower bounds total 90% and the upper bounds total 110%. The midpoints happen to total 100%, and that still does not pin an exact split. Use the bands to order study time.

Can I use Microsoft Learn during the exam? Yes, on role-based exams. The split screen covers learn.microsoft.com, minus Q&A, practice assessments, and your profile. No extra time is added, the clock keeps running, and navigation to other domains is blocked.

How long does the certification last? Expert certifications expire annually. Renewal is free, online, unproctored, and open book, with a six month window before expiry. Passing extends the certification one year from the expiration date.

Can I retake it if I fail? A failed attempt can be retaken 24 hours after the first attempt. Waiting periods for later retakes vary.

How does AZ-305 relate to AZ-104, AZ-500, and AZ-700? AZ-104 is the required administrator credential and scores the controls the architect later recommends. AZ-500 scores the security controls that sit on that design. AZ-700 scores the network-engineer design. AZ-305 scores the solution recommendation: the compute shape, the data model, the continuity control, and the landing zone. The official outline is the study guide for Exam AZ-305. Use Microsoft for the task statements and the weight ranges. Use this page for what changed, what the discriminators are, and how to sequence the April 2026 blueprint.