Cloud Digital Leader (CDL) Complete Guide: Exam Scope, Study Plan & Fastest Path to Passing

Cloud Digital Leader (CDL) is the entry-level Foundational-tier certification in Google Cloud's lineup, designed for a broad range of roles beyond just engineers. Together with AWS's Cloud Practitioner (CLF) and Microsoft Azure's AZ-900, it's considered one of the three foundational cloud certifications worldwide — no coding required, intentionally accessible to business roles. This article starts from the 2026 revision exam scope and walks through language availability, a free study route, a real pass report from an IT beginner who made it through in 25 hours, and post-certification career paths — a one-stop overview of CDL.

Google Cloud lags slightly behind AWS and Azure in Japan in terms of brand recognition, but it holds technical advantages in AI/ML, data analytics, and containers, and leading Japanese companies like Mercari, CyberAgent, SmartHR, LayerX, LINE, and freee have adopted GCP as a primary platform. For anyone targeting these companies, or for non-technical staff driving GCP adoption internally, CDL is the ideal "first card" to put on the table.

CDL Exam Specifications

Here are the exam basics. CDL is a 90-minute, 50-60 question multiple-choice/multiple-select exam at $99 USD (roughly 14,000-16,000 JPY), with a 3-year validity. You can sit it either online (OnVUE) or at a Pearson VUE test center. Google Cloud completed its migration from Kryterion (Webassessor) to Pearson VUE on February 22, 2026, so the registration flow is now unified with AWS/Azure.

The passing score is undisclosed, but the industry consensus is that around 70% is the passing line. Japanese pass reports converge on "70-80% gives you a safe margin", and the rule of thumb is that if you can consistently hit 80% on the Practice Exam, you're ready for the real thing.

The 4 Exam Domains

CDL is organized around 4 domains. The core is the three pillars of Data, AI, and Modernization, with foundational Digital transformation concepts layered on top. Domain weights aren't explicitly published, but Google's official Exam Guide implies roughly 10% / 30% / 28% / 32%.

Domain 1: Digital transformation with Google Cloud (~10%)

The first domain covers the concepts of cloud itself and digital transformation. Expect questions on the essential characteristics of cloud computing (the NIST five: on-demand self-service, broad network access, resource pooling, rapid elasticity, measured service), the differences between public/private/hybrid/multi-cloud, CapEx vs OpEx, and how Google Cloud "accelerates business transformation". Much of this overlaps with AZ-900 Domain 1, so studying for both compounds nicely.

Domain 2: Data transformation with Google Cloud (~30%)

This domain covers data analytics, where Google Cloud has its strongest hand.BigQuery dominates the question set: characteristics as a serverless analytical data warehouse ("instantly query petabyte-scale data via SQL"), pricing models (on-demand vs capacity-based), BigQuery ML for SQL-based machine learning, and BigQuery Omni for multi-cloud analytics.

Beyond BigQuery, the key is knowing when to pick each service: Dataflow (Apache Beam-based stream/batch processing) for pipelines, Pub/Sub (globally scalable messaging) for messaging, Dataproc (managed Hadoop/Spark) for data integration, Dataprep for data preparation, Dataplex for data governance, and Looker / Looker Studio for visualization.

The relationship between Cloud Storage and BigQuery comes up often. Start with the simple distinction — "BigQuery is for analytics, Cloud Storage is general-purpose storage" — then layer on knowing that BigQuery can directly query files in Cloud Storage via external tables, and how to choose among storage classes (Standard / Nearline / Coldline / Archive).

Domain 3: AI and machine learning with Google Cloud (~28%)

This is the domain whose weight has grown the most in recent years. The center is the Vertex AI platform (Google Cloud Next in April 2026 announced a rebrand to the Gemini Enterprise Agent Platform, but the current CDL scope still uses the Vertex AI naming) — expect questions on AutoML, Custom Training, Model Garden, Vertex AI Studio, and Agent Builder.

On the generative AI side, you'll see the characteristics of the Gemini family (Pro / Flash / Ultra / Nano), enterprise search via Vertex AI Search, document processing via Document AI, and the broad concepts of pre-existing ML APIs like Speech-to-Text / Text-to-Speech / Translation API. You won't be asked granular API names at the level of individual Cognitive Services, but you do need enough understanding to pick Vertex AI Agent Builder or Dialogflow CX in scenario questions like "Which service should you use to build a customer support chatbot?".

Domain 4: Infrastructure and application modernization (~32%)

The largest-weighted domain. On compute, expect frequent questions on choosing among Compute Engine (IaaS VMs), Google Kubernetes Engine (GKE, managed Kubernetes), Cloud Run (serverless containers), Cloud Functions (FaaS), and App Engine (PaaS). The key is being able to articulate decision rules in your own words: "Compute Engine when you need full control, Cloud Run for serverless containers, GKE for managed Kubernetes".

On networking, expect the fact that VPC (Virtual Private Cloud) is a global resource (unlike the region-scoped VPCs in AWS or Azure), global distribution via Cloud Load Balancing, and choosing between Cloud Interconnect and Cloud VPN. On security, the exam covers the positioning of IAM, Identity-Aware Proxy (IAP), Cloud Armor, Cloud KMS, VPC Service Controls, and Security Command Center at an introductory level.

For migration strategy, it's worth memorizing Google Cloud's official "6 Rs" framework (Rehost / Replatform / Repurchase / Refactor / Retire / Retain), as well as the existence of migration tools like Migrate to Virtual Machines (formerly Migrate for Compute Engine) and Database Migration Service.

Can You Pass Using Only Official Resources?

The short answer: yes, CDL is an exam you can pass using Google's official resources alone. Paid Udemy courses and books help as supplements, but they're not required. Work through the following four resources in order, and even IT beginners can reach passing territory in four weeks.

The Google Cloud Skills Boost Learning Path is offered completely free, and alongside the equivalent Coursera courses it's the standard route for GCP study. You can create a Skills Boost account instantly with a Google account, which lowers the barrier to getting started.

A 4-Week Roadmap for IT Beginners

Looking across Japanese pass reports, study hours vary significantly by background. The average ranges are 20-40 hours for IT beginners, 15-25 hours with 1-3 years of IT experience, and 5-15 hours for those experienced with another cloud. The plan below assumes an IT beginner over four weeks.

Week 1: Complete Modules 1 and 2 of the Cloud Digital Leader Learning Path on Google Cloud Skills Boost (Digital transformation and Data transformation). Open the BigQuery sandbox (a free tier you can use without registering a credit card) and actually run Google's sample queries — it cements understanding quickly.

Week 2: Module 3 (AI and machine learning). Trying out Gemini's free trial in Vertex AI Studio or stepping through the AutoML experience in the Cloud Console makes the otherwise abstract AI terminology suddenly tangible. At this stage, check whether you can explain in your own words questions like "What's the difference between Vertex AI and BigQuery ML?" and "When should you use Gemini Pro vs Gemini Flash?".

Week 3: Module 4 (Infrastructure and application modernization), the most heavily weighted module. Going through at least one cycle of spinning up a VM in Compute Engine, deploying a sample app on GKE Autopilot, and deploying a container on Cloud Run leaves the compute service distinctions firmly in your head. New Google Cloud signups get $300 in free credits over 90 days, so the hands-on practice is essentially free.

Week 4: Finishing touches. Iterate on the official Practice Exam until you can score 80% or better, and review the relevant Skills Boost modules for any weak areas. The day before, watch Pearson VUE's tutorial video to get familiar with the test UI — the standard pattern to avoid panicking on exam day.

Insights from Japanese Pass Reports

Reading across Japanese CDL pass reports on Qiita, Zenn, note, and Hatena Blog, several common patterns emerge.

First, the time to pass ranges from two weeks to two months. The shortest case is an engineer with 5+ years of IT experience passing after a one-week sprint; the longest is a non-technical sales rep with no IT background taking two months. What everyone has in common is a simple winning formula: "do one pass through Google Cloud Skills Boost, then run the Practice Exam at least three times". Conversely, people who leaned heavily on paid materials like Udemy tend to report regrets like "the terminology was different from official sources and confused me" or "the content was outdated and didn't match the actual exam". The shared takeaway: official resources first is the most cost-effective strategy.

A common failure pattern is memorizing service positioning in isolation and failing to handle context-driven questions on exam day. For example, memorizing "BigQuery is an analytical data warehouse" isn't enough — you need to be able to pick BigQuery in a scenario like "What's the best service for ad-hoc analytics on 10TB of web logs?". Make a deliberate habit of working backwards from business scenarios to service selection, rather than studying services in isolation.

What to Aim for After CDL

CDL isn't a destination — it's the entrance to the broader Google Cloud certification family. Where you go next depends on the role you're aiming at.

If you're aiming to be an engineer, the next step is Associate Cloud Engineer (ACE). It centers on hands-on operations via the Cloud Console and gcloud CLI: 120 minutes, $125. It tests operations of Compute Engine, GKE, Cloud Storage, IAM, and Cloud SQL — effectively the required certification for GCP engineers.

If you're aiming for generative AI strategy, take Generative AI Leader (GAIL). Launched in May 2025, this new exam covers the Gemini family, Vertex AI Agent Builder, RAG, prompt engineering, and other generative AI strategy topics for business roles. The CDL + GAIL combination is a strong positioning as a business professional who can speak fluently about cloud and generative AI.

If you're aiming for data analytics, the choices are Associate Data Practitioner (ADP) or Professional Data Engineer (PDE). ADP is the entry level (currently English-only); PDE is the flagship data engineer certification, centered on data pipeline design combining BigQuery, Dataflow, and Pub/Sub. Whether to go through ADP first or attempt PDE directly depends on your hands-on data experience.

Frequently Asked Questions

Exam information in this article is based on the official Google Cloud Cloud Digital Leader certification page and the official Google Cloud documentation (CC BY 4.0). This article is not an official Google LLC product and has no partnership or sponsorship relationship. Google, Google Cloud, Google Cloud Platform, BigQuery, Vertex AI, Gemini, Looker, and Dataflow are trademarks or registered trademarks of Google LLC. Information is current as of May 24, 2026, based on publicly available official materials. For the latest details, always check theofficial Exam Guide.