How to learn the AWS Solutions Architect exam
This path assumes you can already navigate the AWS console and know what an EC2 instance is. It gets you to a pass on the Solutions Architect Associate exam by testing your ability to choose between imperfect options under cost, security and performance constraints.
Hours
70-90 hours
Calendar
about 12-15 weeks at 6 hours a week
Stages
6
Start from
You need basic networking (CIDR blocks, subnets, ports) and hands-on AWS console familiarity. You do not need to be an active developer or have prior cloud certification — the common belief that you must be a practising engineer is false.
Why the usual order fails
The AWS Solutions Architect exam is not a service tour. Knowing what Route 53, SQS and KMS do will not get you a pass. The exam presents scenarios that are deliberately ambiguous: a workload that is almost stateless, a budget that is nearly enough, a compliance requirement that conflicts with performance. You are asked to pick the least wrong architecture. Every question tests whether you can weigh cost against durability, latency against consistency, operational burden against control. Service knowledge is the vocabulary, but the exam marks your reasoning.
Most study plans start with a broad sweep of every service in the AWS catalogue, then add practice exams at the end. That order fails because you spend early weeks memorising what services exist rather than learning how they interact under pressure. This path starts in week one with a three-tier web app and forces you to break it: swap EFS for S3, force a failover, remove a NAT gateway and watch the costs change. By the time you learn the full service list, you already know why the services exist.
The path, in the order it has to be walked.
Each stage assumes the one above it and nothing below it. The check at the foot of each is the point — you are meant to fail it, go back, and try again before moving on.
Core compute and storage trade-offs
weeks 1-2You build a three-tier app on EC2, EBS, S3 and EFS and deliberately break it. This stage sits first because every later decision assumes you know what a compute-hour costs, what a snapshot restores, and why an EBS volume is not a file share.
- EC2 instance families and pricing models (On-Demand, Reserved, Spot, Savings Plans)
- EBS volume types: gp3, io2, st1, sc1
- S3 storage classes and lifecycle rules
- EFS vs FSx for Lustre vs EBS
- AMI and instance store vs persistent storage
- Elastic Load Balancing: ALB vs NLB vs GWLB
- Auto Scaling groups, launch templates, scaling policies
You can now: You can take an existing single-instance web app and split it into a load-balanced, auto-scaled architecture, stating the monthly cost within 25%.
Networking and hybrid connectivity
weeks 4-5VPCs, subnets, and routing are where most candidates lose marks. This stage sits after compute because you now have a real app to connect, and before security because you cannot talk about NACLs or security groups without first knowing what a route table does.
- VPC, subnets, CIDR, route tables, NAT gateways
- VPC peering, Transit Gateway, PrivateLink
- Security groups vs NACLs
- Site-to-Site VPN and Direct Connect
- Route 53: hosted zones, routing policies, health checks
- CloudFront and Global Accelerator
You can now: You can design a VPC for a multi-tier app with public, private and isolated subnets, and explain why a NAT gateway in one AZ is not a single point of failure.
Security and identity
weeks 6-7IAM, KMS, and logging underpin every architecture decision, but they are hard to learn in the abstract. This stage comes after networking because you now have resources to protect and a VPC to lock down. It sits before databases and serverless because those services have their own security models that assume IAM fluency.
- IAM users, roles, policies, and permission boundaries
- KMS keys, envelope encryption, and key rotation
- S3 bucket policies vs IAM policies vs ACLs
- CloudTrail, CloudWatch Logs, and VPC Flow Logs
- AWS Shield, WAF, and DDoS basics
- Secrets Manager and Parameter Store
You can now: You can write an IAM policy from scratch that grants least privilege for a Lambda function to read from one DynamoDB table, and explain why a resource-based policy would be wrong.
Databases and decoupling
weeks 8-9This stage introduces the services that most often appear as wrong answers: DynamoDB, RDS, SQS, SNS, and EventBridge. It sits after security because most database failures on this exam are IAM failures in disguise, and before serverless because SQS is the integration point for Lambda.
- RDS: Multi-AZ, read replicas, and backup windows
- DynamoDB: tables, indexes, streams, and capacity modes
- ElastiCache: replication and failover
- SQS: queues, dead-letter queues, and visibility timeout
- SNS, EventBridge, and Step Functions
You can now: You can take a monolithic app with an hourly batch job and redesign it into an event-driven pipeline using SQS and DynamoDB, then explain why you did not use RDS.
Serverless and containers
weeks 10-11Lambda, ECS, and EKS are the last service-heavy stage because the exam expects you to know when not to use them. By now you have enough context to see serverless not as a new paradigm but as a different cost and latency profile for components you already understand.
- Lambda: execution limits, concurrency, and cold starts
- API Gateway: REST, HTTP, and WebSocket APIs
- ECS with Fargate vs EC2 launch types
- EKS and Kubernetes basics
- Amazon Aurora Serverless v2
You can now: You can take the same web app from stage one and decide which parts to move to Lambda, which to ECS, and which must stay on EC2, with cost and latency as the only constraints.
Cost optimisation and exam practice
weeks 12-14The last stage is not a review. It is a series of timed practice exams with a focus on cost and trade-off questions. This is where you learn to read a scenario and identify the constraint that matters, which is the skill the exam actually measures.
- AWS Budgets, Cost Explorer, and Trusted Advisor
- Right-sizing, and Savings Plans
- Practice exams: timed, 65 questions, 130 minutes
- Review of wrong answers classified by service area
You can now: You can complete a full 65-question practice exam in under 130 minutes and score above 75%, with every wrong answer classified as either a knowledge gap or a reading error.
The hours are fixed. The calendar is yours.
The same path at three speeds, and what actually goes wrong at each.
3 hours a week
At 3 hours a week you finish in 23-30 weeks, or roughly six months. This pace is sustainable if you already work with AWS daily, but the gap between sessions means you will spend 20-30% of each session just recalling what you did last week.
8 hours a week
At 8 hours a week you finish in 9-11 weeks. This is the sweet spot for people with a full-time job: enough time to build and break things, not so much that the exam becomes a second job. You can do a full practice exam every weekend.
20 hours a week
At 20 hours a week you finish in 4-5 weeks. This pace only works if you are not working full-time. The risk is that you move through stages before the trade-offs sink in, and you end up memorising services instead of reasoning about them.
Four ways people lose a year on this.
Each of these is appealing for a real reason, which is why they are common. What they cost is the part nobody mentions.
Watching service tour videos
People spend 40 hours on video courses that walk through every AWS service in isolation. This feels productive, but the exam never asks 'What is DynamoDB?' It asks 'Which database meets these three conflicting requirements?' A service tour gives you a menu, not a recipe.
Doing practice exams only
People cram with practice exams and memorise the answers. This works for a week and then fails on test day because the exam has a huge question pool. You are learning the shape of the question, not the architecture that produced it. You need to build before you drill.
Ignoring the cost dimension
Many candidates treat cost as an afterthought, a few easy marks at the end. But cost is a first-class constraint in nearly every Solutions Architect question. If you cannot compare the monthly cost of two architectures within 25%, you are guessing on half the exam.
Assuming you need developer experience
People delay starting the exam because they think you must be a software engineer. The exam expects you to read JSON and understand basic web architecture, not write production code. The real barrier is not programming skill, it is the ability to reason about systems under constraint.
How to know it worked.
Not a syllabus you tick off — six things to attempt. If one of them is still hard, the stage it belongs to is the one to go back to.
- Draw a VPC with public and private subnets and label every route table entry.
- Explain why a NAT gateway is in a public subnet but a database is not.
- Write an IAM policy that allows a Lambda to read from one S3 bucket and nothing else.
- Design a multi-AZ RDS deployment and estimate its monthly cost within 25%.
- Take a single EC2 web app and split it into an auto-scaled, load-balanced architecture.
- Complete a 65-question practice exam in under 130 minutes and score above 75%.
Questions about learning the AWS Solutions Architect exam.
The five we hear most. The intro film answers most of them in a minute and a half.
how long does it take to prepare for the AWS Solutions Architect exam?
For most people with basic AWS experience, it takes 70 to 90 hours over 12 to 15 weeks. That is not time spent watching videos; it is time spent building, breaking, and fixing architectures. The exam is not a memory test, so the hours you spend in the console count as much as the hours you spend reading.
can I pass the AWS Solutions Architect exam without hands-on experience?
No, not reliably. The exam tests your ability to choose between architectures under cost and performance constraints, and that skill only comes from building. You can memorise service names and still fail every trade-off question. A free-tier account and a few hours of console work are enough.
is the AWS Solutions Architect exam hard?
It is hard in a particular way. There is little obscure trivia; the difficulty comes from questions where every option is technically plausible and you must choose the one that best satisfies cost, security and availability at once. Candidates who come from a pure development background often find this harder than they expect.
do I need to know how to code for the aws solutions architect exam?
No. The exam does not require you to write code. It does require you to read short code snippets in the context of IAM policies or Lambda functions, and to understand basic concepts like JSON and environment variables. If you can read a config file, you are ready on that front.
is the AWS Solutions Architect exam worth it?
For most people working with AWS, yes. The certification is widely recognised and the preparation itself changes how you design systems: you start thinking about failure modes, cost curves, and the operational burden of every component. Even if you never sit the exam, the study process makes you a better architect.
A path is not a course.
This is the shape of the AWS Solutions Architect exam, not the teaching of it. Hand the same path to BiliMind and it writes the lessons onto it — in the form each one calls for, with practice that gets marked and a schedule that bends when your week does. Mapping a course is free and unlimited.
This roadmap was drafted on 18 September 2026 by the same engine that builds BiliMind courses, and reviewed before publishing. Hours are honest estimates for working competence, not for a demo — treat them as a starting point and adjust them against your own first fortnight.