Walk both certification ladders rung by rung — every tier, exam style and fee — to see whether the AWS or Google Cloud path suits your career plan.
Picture a mid-career engineer with a free evening, two browser tabs open — aws.amazon.com/certification in one, cloud.google.com/learn/certification in the other — and a genuine intention to commit to a multi-year plan tonight. One tab shows four tiers, exam codes like SAA-C03, and prices climbing from $100 to $300. The other shows three tiers, no exam codes anywhere, and a $99 to $200 range. The tabs look similar enough to feel interchangeable. They are not, and the differences run deeper than price.
Short answer: commit to the AWS path if you want maximum employer reach or you have no platform constraint — its ecosystem is the largest, and its ladder offers role-specific rungs from Associate through Specialty. Commit to the Google Cloud path if your work, employer or ambitions centre on Google Cloud, particularly around data and machine learning, where its Professional tier is unusually deep. This article walks both ladders rung by rung, then compares exam style, cost and the market, so the engineer with those two tabs open can close one of them.
(Choosing between three providers rather than two? Microsoft's track is out of scope here by design — the three-way verdict lives in AWS vs Azure vs Google Cloud certification.)
AWS structures its portfolio, as of August 2026, into four tiers. No exam has prerequisites — recommended experience is guidance only — and every certification is valid for 3 years.
Two entry credentials: AWS Certified Cloud Practitioner (CLF-C02) — 65 questions, 90 minutes, aimed at people with up to six months of general AWS exposure — and AWS Certified AI Practitioner (AIF-C01), the same length and price, aimed at people who work with AI/ML solutions without necessarily building them. AIF-C01 was among the first AWS exams to introduce ordering, matching and case-study question types. Foundational passing score: 700 scaled.
Five role-based exams, each 65 questions and 130 minutes with a 720 scaled passing score: Solutions Architect (SAA-C03) — the classic first technical cert, with a recommended year of hands-on design experience; Developer (DVA-C02); CloudOps Engineer (SOA-C03) — note this is the renamed successor to SysOps Administrator, not a discontinued cert (SOA-C02's last day was 29 September 2025); Data Engineer (DEA-C01); and Machine Learning Engineer (MLA-C01), whose MLA-C02 update opened for registration on 1 September 2026.
Three exams at 75 questions and 180 minutes with a 750 scaled passing score: Solutions Architect Professional (SAP-C02), DevOps Engineer Professional (DOP-C02), and the new Generative AI Developer Professional (AIP-C01), launched via a 2025 beta and standard from 2026. A useful renewal quirk: passing SAP-C02 also renews Solutions Architect Associate and Cloud Practitioner, and DOP-C02 renews Developer and CloudOps.
Security (SCS-C03) is the flagship here — 65 questions in 170 minutes. Advanced Networking (ANS-C01) is on its way out, with a final exam date of 31 December 2026 and no announced successor. Several other specialties (Database, Data Analytics, Machine Learning) have already been retired, their ground absorbed by the Associate and Professional AI/data certs — a reminder that AWS actively prunes this tier.
Google Cloud's portfolio, also as of August 2026, spans three tiers and 13 active certifications. Nothing carries an exam code — credentials go by name — and no exam has prerequisites. Foundational and Associate certs last 3 years; Professional certs last only 2. Google publishes no passing scores: results are pass/fail, provisional on screen, confirmed within 7–10 days.
Cloud Digital Leader (50–60 questions, 90 minutes) covers cloud concepts for business-facing roles, with a version refreshed recently — work from the current exam guide. Generative AI Leader (launched 2025, same format and price) covers generative-AI fundamentals and business strategy. Both renew via a short $60-class renewal exam or Google Skills coursework.
Three exams, each 50–60 questions over 2 hours: Associate Cloud Engineer, the standard first technical rung, with six-plus months of hands-on experience recommended; Associate Data Practitioner (launched late 2024), the new entry point for data careers; and Associate Google Workspace Administrator, which replaced the retired Professional-level Workspace cert at the end of 2024.
Eight role-based exams, and this tier is the Google path's centre of gravity: Cloud Architect (the only exam with a published case-study structure — two case studies from a pool of four, 20–30% of questions), Data Engineer (40–50 questions, unusually), DevOps Engineer, Security Engineer, Network Engineer, Machine Learning Engineer (recently updated toward the Gemini Enterprise Agent Platform rather than a purely Vertex-AI focus), Database Engineer, and Cloud Developer. All run 2 hours at $200 plus tax, and all expire after 2 years.
The eight-way Professional spread is broader than AWS's three Professional exams — Google pushed its role specialisation to the top of the ladder, where AWS splits it between Associate, Professional and Specialty tiers.
Neither path includes hands-on labs — everything is question-based, proctored through Pearson VUE online or at a test centre. Within that, four style differences are worth weighing.
Transparency. AWS states the bar (scaled scores of 700/720/750 by tier) and reports your score. Google reports pass or fail and nothing else. If knowing your margin matters to how you study, AWS's approach is the friendlier one.
Question formats. AWS's classic exams are multiple choice and multiple response; its newest AI exams (AIF-C01, MLA-C01) add ordering, matching and case-study types. Google uses multiple choice and multiple select throughout, with the Cloud Architect case studies as the standout format feature — and because the case-study pool is published, you can read them before exam day.
Pacing. AWS Associates run 65 questions in 130 minutes; Professionals 75 in 180; the Security Specialty 65 in 170. Google exams run 50–60 questions (40–50 for Data Engineer) in 120 minutes. Both leave roughly two minutes per question; stamina, not speed, is the differentiator at Professional level.
Retakes. Google's policy is explicit: 14 days after a first fail, 60 after a second, 365 after a third, capped at four attempts in two years for Associate and Professional exams. Confirm AWS's current retake terms on its policies page when you book.
One thing neither provider publishes: pass rates. Any source quoting "the GCP Architect has a 40% pass rate" (or similar) is fabricating. Compare formats and experience expectations instead.
Add up a realistic multi-year journey rather than comparing single fees.
| Factor | AWS path | Google Cloud path |
|---|---|---|
| Difficulty signals | Published scaled passing scores; score reports; no labs | Pass/fail only; case studies on Cloud Architect; no labs |
| Prerequisites | None at any tier | None at any tier |
| Cost (2026, USD) | $100 Foundational / $150 Associate / $300 Professional & Specialty | $99 Foundational / $125 Associate / $200 Professional (+tax) |
| Validity | 3 years, all tiers | 3 years Foundational/Associate; 2 years Professional |
| Renewal levers | Higher exam renews lower certs; 50% discount voucher for active holders; free Cloud Quest path for Cloud Practitioner; 1-year Skill Builder maintenance extension | Short renewal exams ($60–$100 where offered, e.g. $100/1 hr for Cloud Architect); Google Skills course path adds 1 year only |
| Best for | Reach across the largest employer base; role depth at several tiers | Google Cloud shops; data/ML-centred careers |
| Career path | CLF/AIF → role Associate → Professional/Specialty | CDL → ACE or Data Practitioner → role Professional |
Worked example over six years, list prices: an AWS climber taking Cloud Practitioner, Solutions Architect Associate and Solutions Architect Professional pays $550 in first-sitting fees; because SAP-C02 renews the lower certs and holders carry a 50% voucher, the renewal burden stays light. A Google climber taking Cloud Digital Leader, Associate Cloud Engineer and Professional Cloud Architect pays $424 up front — but the Architect cert expires every 2 years, at $100 per renewal exam (a 25-question, 1-hour sitting currently focused heavily on generative-AI solutions), and letting it lapse more than 30 days means a full-price retake. Cheaper entry, busier maintenance. Neither total should decide your career, but if you plan to hold three or four certifications at once, AWS's renewal mechanics scale better.
Workload share is the demand proxy that matters, and the most recent verified figures — Synergy Research Group, Q3 2025 — put AWS at 29% of cloud infrastructure services and Google Cloud at 13%, within a big three holding 63% of a $106.9 billion quarterly market. Synergy also noted the direction of travel: AWS averaging just under 30% over the prior year, down from about 32% in 2021, with Google among the gainers. AWS separately reports 1.42 million active certifications and 1.05 million certified individuals as of January 2025 — evidence of enormous recognition, and of a crowded field.
Translated into careers: the AWS path is relevant to more employers in almost every market, while the Google path concentrates its value where Google Cloud is deployed — disproportionately data-, analytics- and ML-heavy organisations — and benefits from a thinner competing talent pool there. On pay, the one independently verified figure in this matchup favours Google: Skillsoft's 2025 top-paying certifications list reports an average of about $190,000 for US-based Professional Cloud Architect holders (its 2024 edition had the same credential top-ranked at $200,960). Skillsoft regularly ranks AWS architect and security certs highly too; pull exact AWS numbers from skillsoft.com rather than second-hand quotes, and remember every salary figure varies by location, experience and role.
If security engineering specifically is your destination, the dedicated head-to-head between AWS's Security Specialty and Google's Professional Cloud Security Engineer is compared in Google Cloud vs AWS security certifications rather than here.
Three sketches, because "better" only resolves against a situation:
The employer-anchored engineer. Works at a company already committed to one provider. Path chosen: whatever the company runs, at Associate level first — Associate Cloud Engineer in a Google shop, Solutions Architect Associate in an AWS one. Certifying against the platform you use daily halves the study burden, because the exam scenarios describe your Tuesday.
The data specialist. Building a career in pipelines, warehousing or ML. The Google ladder offers an unusually coherent sequence — Associate Data Practitioner, then Professional Data Engineer, then Professional Machine Learning Engineer — mapped to an ecosystem employers in this niche genuinely run. AWS answers with Data Engineer Associate and its new AI credentials, so this is a lean toward Google, not a rout; let your target employers' stacks cast the deciding vote.
The unconstrained career changer. No employer, no niche, choosing cold. Take the AWS path: the widest set of doors, the most recognisable credentials, and a forgiving renewal regime — then add a Google cert later if a role demands it. Cloud fundamentals transfer well; a second provider's Associate exam is far cheaper in study hours than the first.
That last point is where structured practice fits naturally: once you have worked through the official exam guide and built against each objective, a timed simulation shows whether you are ready to book. ExamPractice offers free sample questions across both ladders — from Google Professional Cloud Architect and Professional Data Engineer to the AWS tiers — with fuller question sets and practice-test simulation for subscribers.
You can — no rule prevents it — but sequencing beats parallelism. Reach Associate or Professional depth on one platform, let real work consolidate it, then cross over. Two shallow Foundational badges impress nobody.
Indirectly, yes. Identity, networking, storage classes and managed compute concepts rhyme across clouds, so your second provider costs fewer study hours. But each exam tests its own service catalogue and design idioms, and cross-cloud intuition supplies confident wrong answers — verify everything against the target platform.
Comparable. Both entry exams (Cloud Practitioner; Cloud Digital Leader or Associate Cloud Engineer) are bookable with no prerequisites, sit in 90–120 minutes, and are available online-proctored. Speed depends on your starting knowledge, not the provider.
Neither, in their old forms. AWS renamed SysOps Administrator to CloudOps Engineer (SOA-C03) in 2025; Google retired the Professional Google Workspace Administrator at the end of 2024 in favour of an Associate-level replacement. Existing holders' certs remained valid for their normal terms.
The engineer with two tabs open should close Google's if reach is the goal and no platform commitment exists — the AWS ladder's market weight, transparent scoring and three-year, voucher-assisted renewal regime make it the stronger general-purpose path. They should close Amazon's if their work or ambitions live where Google Cloud is strong — above all in data and machine learning — and they can accept two-year Professional validity as the toll for a scarcer, well-regarded credential. What neither engineer should do is leave both tabs open indefinitely: the value in either path comes from climbing it, and the best-evidenced rung to start on is the Associate level of whichever platform your next job will actually run.
Exam facts in this guide were checked against official certification-provider pages on . Fees, exam codes and policies change — confirm on the provider’s own site before you book.
Put it into practice
Reading about an exam only takes you so far. Work through practice questions for your certification and find the gaps before exam day does.
·10 min read
Certification or degree? Compare cost, time, employability and career ceiling honestly, with a decision framework for students and career changers.
Continue reading·10 min read
Network+ or CCNA? Compare difficulty, depth, cost, renewal and employer recognition, then pick the networking certification that fits your career plan.
Continue reading·9 min read
CCNA or CCNP? Since 2020 there is no prerequisite, so the choice is yours. Compare cost, difficulty, salary data and who should skip straight to CCNP.
Continue reading
