Preparing For and Passing The Google Professional Cloud Architect Exam (2025)

In this post, I share my journey of becoming a Google Cloud Certified Professional Architect to help our company achieve Google Partner status. I compare the partnership requirements between Microsoft and Google, explaining why Google's path was more accessible for our immediate goals. I detail my study approach using Google Cloud Skills Boost and the Professional Cloud Architect Study Guide, breaking down what actually appeared on the exam. Finally, I walk through the exam experience itself—from the intense security checks to that moment of relief when "PASSED" appeared on my screen. Whether you're pursuing certification for career growth or to help your company become a cloud partner, this guide provides a roadmap based on real experience. All you can do is try!
Google Cloud Architect Certification Guide, Image of the Google Cloud Certified Professional Cloud Architect Badge
What Partner Program Should We Pick
In the summer of 2024, one of our founder's had tasked me with identifying what it takes to become a cloud partner with at least one of the big three cloud providers; Google, Microsoft, and Amazon.
I've been with the company since the year after its founding in 2016 and nearly all of our clients were using, or pivoting towards, either Microsoft Azure or Google Cloud Platform (GCP). That left Amazon out of the equation. Receiving a Microsoft Solutions Partner designation, focusing on the Data & AI area, was a non-trivial task. However, becoming a certified Google Partner was much more approachable; both from a business and technical perspective.
Requirement | Google Partner Advantage | Microsoft Solutions Partner (Data & AI) |
---|---|---|
Overall Structure | Three tiers: Member, Partner, Premier Partner Three functions: Sell, Service, Build |
Solutions Partner designation with 70+ points across five metrics |
Technical Credentials | Two individuals with professional certifications (e.g., Cloud Architect, Data Engineer) |
Prerequisites: Two Azure Administrator Associate AND two Azure Solutions Architect Expert certifications before other certs count |
Points System | N/A - Based on meeting specific requirements | 70 points minimum needed across all categories |
Customer Requirements | Business plan recommended but not mandatory | Net Customer Adds (10 points per customer, max 30 points) Customer Success: Usage Growth (1 point per 1% ACR growth, max 20 points) Customer Success: Deployments (2 points per advanced service, max 10 points) |
Partner Directory | Create profile to appear in directory upon approval | Listed after achieving Solutions Partner designation |
Time to Achievement | Achievable within 2-3 months with focused effort | Typically 6+ months due to certification prerequisites and customer metrics |
Certification Flexibility | Choose from various certification paths based on company strengths | Rigid certification path with strict prerequisites |
Microsoft Solutions Partner (Data & AI)
While Google's partnership path seemed straightforward, Microsoft's Solutions Partner designation is like climbing Everest with a backpack full of requirements. The process requires earning at least 70 points across five distinct metrics, and you need to score in each category—no cherry-picking your strengths!
First, Microsoft wants to see "Performance" through net customer adds. Each new customer nets you 10 points (up to 30 total), but they need to be properly tracked in Partner Center with linked PartnerIDs—administrative hoops that require jumping through.
The "Skilling" requirements are where things get truly daunting. Before you even start accumulating points, you need specific prerequisites: two team members with Azure Administrator Associate certifications AND two with Azure Solutions Architect Expert certifications. Only after clearing these hurdles can additional certifications like Azure Data Engineer Associate or Fabric Analytics Engineer start counting toward your 40 possible points in this category.
The "Customer Success" metrics focus on usage growth and deployments—essentially requiring you to prove you're not just acquiring customers but helping them succeed and expand their Azure footprint. Each 1% growth in Azure consumed revenue (ACR) earns a point (up to 20), while deploying advanced Azure services earns you 2 points per service (up to 10 total).
Faced with this mountain of requirements versus Google's more approachable path, our choice became clear. We could start building our Google partnership credentials relatively quickly while keeping Microsoft as a longer-term goal.
Google Partner Advantage
Google Cloud Partnership Tiers
Google Partner Advantage offers three tiers; Member, Partner and Premier Partner. There are also three types of functions a Partner can perform; Sell, Service, and Build. Each function has different requirements that need to be met to be granted Partner status. Service refers to partners who primarily focus on providing consulting, implementation, and support services for Google Cloud products (that's us), Build refers to partners who create custom solutions or applications that integrate with Google Cloud, and Sell refers to partners who primarily focus on selling Google Cloud products and managing customer accounts.
Since our company specializes in professional services (consulting, implementation, etc.), we followed the Service level, Partner-tier partnership requirements to start. These are as follows:
Google Partner Advantage Requirements
Step 1: Complete Authorization request
To achieve Partner Level, organizations must meet the requirements per product family. Organizations must choose one or more combinations of products and regional Authorizations.
Step 2: Earn your credentials
Attain required credentials and certifications required for your organization to achieve desired product Authorizations, allowing you to sell them.
- Two individuals w/ professional certs/qualifying creds
- Technical credentials/certifications:
- Associate Cloud Engineer Certification
- Professional Cloud Architect Certification (I picked this one!)
- Professional Cloud Database Engineer Certification
- Professional Cloud Developer Certification
- Professional Cloud DevOps Engineer Certification
- Professional Cloud Network Engineer Certification
- Professional Cloud Security Engineer Certification
- Professional Data Engineer Certification
- Professional Machine Learning Engineer Certification
- Maps Technical Fundamentals Credential
- Technical credentials/certifications:
Step 3: Start your Partner Directory profile
Add information, offices, and any links or documents to your profile. When you become a partner you will show up in our Partner Directory.
Step 4: Submit your business plan (recommended)
The Annual Business Plan contains your go-to-market plan, associated resources, and expected performance metrics for a specified period of time.
For the purposes of this blog post, I'll be largely skipping Steps 1, 3, and 4. A follow-up post will go in more-depth on how these problems can be tackled.
Now that we understood the partnership requirements, it was time to choose which certification would best serve both my career and our company's goals.
I'm a Data Engineer but...
Our company employs a vast array of fantastical talent. Folks span every facet of human life. Even so, we largely have three main roles; Data Analyst, Data Engineer, Developer. However, our titles, up until very recently, have been vague; Analytics Consultant. This has allowed me the freedom to explore what interests me and what I'm good at. It's given a well-roundedness in this space.
In the span of 8 years working in the tech space, I've held the role of junior data analyst, A/B and MVT analyst, technical program manager, data architect, but one role usually sticks out primarily for me; senior data engineer. And yet, I picked the Professional Cloud Architect exam... for a much more strategic reason.
During my research, I noticed that one of the requirements to becoming a Premier Partner with Google required more than only Data Engineer certs, or only Data Analyst certs. Google was looking to see if your company had the flexibility to service clients across a wide array of issues, be it machine learning, AI tooling, architecture, analytics, etc. So, looking to the future I decided, the hell with it, how much more difficult can this Professional Architect cert be? Time to study! [insert maniacal laughter]
With my certification target locked in, I needed a solid game plan to prepare for this challenging exam. Here's what worked for me.
Cloud Certification Study Resources
Study Resource 1: Google Cloud Skills Boost
Two perks of having your company sign up for Google Partner Advantage are:
- Free access to all learning paths on Google Cloud Skills Boost
- Free exam voucher when all course related material is completed within 10 weeks of course start through their Kickstart program
Google Cloud Skills Boost is an invaluable tool that will help prepare you for the exam. It covers all resources, processes, and the entire GCP ecosystem as it relates to your certification path. Along with detailed videos and quizzes, where Skills Boost really shines is the hands-on labs. Google has student sandboxes to help you practice in GCP and there is a fair mix of terminal/CLI and UI work.
For all the good that Skills Boost has there is one major downfall. It is very generalized. As an example, the labs don't explain that in the real world based on requirements that you would use Compute Engine when application admins need root access and their workloads use a specially hardened operating system. Rather you'll learn how to spin up and shit down VMs, how to select boot disks, and memory. While Skills Boost provided excellent hands-on experience, I needed something more to bridge the gap between technical knowledge and real-world application.
This is where the Professional Cloud Architect Study Guide book comes in.
Study Resource 2: Professional Cloud Architect Study Guide
This book helped to solidify all the knowledge I gained from Skills Boost. It provides a brief overview of all four case studies the exam may test you on. For each case study it provides a thought framework for how to navigate business and technical requirements and move forward with recommendations on the various services and products available.
Each chapter has a set of review questions which help apply the learnings via real world problems. Additionally, with the purchase of the book, you'll gain access to the Sybex Online Test Bank for a full year. These questions are more difficult but also more similar to the ones found in the actual exam, further preparing you. After months of preparation, the day of reckoning finally arrived. Here's what the exam experience was really like.
The Exam Experience
Cloud Architect Exam Topics
The online courses set a basic foundation for how GCP products and services are built and what their core functions are. The labs served as class room activities to understand how each resource worked and was configured.
The study guide provided a more refined approach to each GCP resource. It went into greater depth of these products and services and provide many more real world examples. Additionally, the four case studies were included in much of the book's content.
My exam breakdown was, to the best of knowledge:

- Case Study-related questions (~11 questions) - Most of the case study questions could have been answered without reading the case studies. That said, if you are stumped, read the case study (it's included with every relevant exam question) as you may find a clue that helps you.
- Networking (~17 questions) - Questions about HTTP, TCP/UDP, Global Load Balancers, SSL Proxy, Cloud CDN, Cloud DNS, VPN, VPC Peering, Interconnect (Dedicated, Partner)
- Compute (~13 questions) - Questions were heavy on GKE and Kubernetes, especially nodes, pods, replicaSets, etc. There was also mention of App Engine and it's capabilities Almost no questions about machine types e.g. E2, A1, N2, GPU/TPU
- Storage (~9 questions) - Questions asked about BigTable, BigQuery, Cloud Storage, FireStore
- There was little to no mention about Data Flow, Cloud Composer, Data Fusion, etc.
Pre-exam security checks
TL;DR Take the exam in a completely empty room with only a desk, chair, laptop or PC/Monitor combo. That's it!
This was a little gnarly. I've been through a handful of proctored exams but this was the most intensive. This is probably due to my home office; which is mostly organized but not uncluttered.
If you're unfamiliar with proctored exams there a few common practices and events that take place:
- Downloading a secure web browser that takes over your entire screen, including your taskbar.
- Submit photos of your testing area using your mobile phone
- Verifying your identification by uploading a selfie and photo of a government-issued ID
- No food/drinks allowed
- Bathroom breaks vary by proctor but, generally, any questions that were viewed prior to the break are not allowed to be changed or reviewed.
I cleared my other work PC, three monitors, all peripherals, and extra cords. And I was still sent to their AI-assisted chat support. This involved walking around with my laptop and showing all the nooks and crannies of my office, under the desk, each individual wall, then the left wall again (I don't know why). I was able to keep my closet doors and filing cabinet drawers closed but everything else seemed to be under a microscope. I did make it through and was placed back in the waiting area.
Exam time scaries
Terrifying! Absolutely terrifying! And then, after nearly two hours, slightly less terrifying. The allotted exam time is two hours. I took one hour and fifty minutes. There are fifty questions. After reading the first question, I felt a heaviness in my gut. It was like everything I had study the past few months was completely erased from my mind. After a few deep breaths, I picked my strategy:
- Select every question for review - As a safeguard, I knew this would allow me to revisit previous questions to double-/triple-check my answers.
- Read every question twice, especially the first time around - Many questions are worded in a way to make you think critically and potentially stump you. However, there are usually clues that will help eliminate at least 1 or 2 of the answers.
- Don't spend too much time on any one question - Rather than focusing on whether or not I truly believed I chose the right answer, I would may my best educated guess and move on. This gives me ample time to...
- Run through all questions at least a second time - It took a full hour to get through all fifty questions. The last hour I felt more relaxed, less pressured because I knew I had an answer for every question.
After clicking submit, your taken to a very underwhelming completion page. After some quick searching up and down the page, I saw that I had PASSED. No confetti, no big font, not even an exclamation point! But it didn't matter, all that hard work, studying, practicing, it had all paid off!
A few days later, I was emailed an official email from Google stating "Congratulations: You are now Google Cloud Certified". In the email you'll find access to your certification credentials, instructions to add your badge to LinkedIn, a Google Cloud Certified digital toolkit, access to the Benefits page where you can find items like mugs, shirts, and hoodies to show off your new certification, and access to the Google Cloud Certified Directory.
So, for any of you contemplating tackling this exam, I say "GO FOR IT!", you have all the resources you need to succeed. I'm rooting for you. All you can do is try!