Exam format: 50 multiple-choice questions · 75 minutes · 80% to pass (40/50 correct) · Free on skillshop.google.com · Valid 12 months · You CAN have GA4 open in another tab during the exam.
1
Setup your learning environment
Day 1 · 1–2 hours
Create a free Google account and access skillshop.google.com — sign in and locate the GA4 Certification module
Must
Request access to the Google Analytics Demo Account (Google Merchandise Store) — this is your practice sandbox
Must
Add your own GA4 property to a site or use Google Tag Manager's preview mode to get hands-on with the interface
Key
2
Master the GA4 data model (most critical)
Day 2–3 · 3–4 hours
Understand the event-based model vs UA session-based. Every interaction is an event with parameters. This is the foundation of every exam topic.
Must
Memorize the 4 automatic events:
page_view, session_start, first_visit, user_engagementMust
Learn Enhanced Measurement events: scroll, click, video_start/progress/complete, file_download, form_submit, site_search — enabled in settings, no code needed
Must
Know the 3 event types: Automatic → Enhanced Measurement → Recommended (e-commerce) → Custom in terms of implementation effort
Key
Understand event parameters vs user properties — what's the difference, scope, and character limits
Key
3
Property setup & configuration
Day 3–4 · 2–3 hours
Know the GA4 account structure: Account → Property → Data Streams (Web / iOS App / Android App)
Must
Understand Key Events (formerly "Conversions" — renamed in 2024). How to mark an event as a Key Event, and how this differs from UA Goals.
Must
Learn Data Filters (internal traffic filter) — how to create, test, and activate them. Know filter states: Testing vs Active.
Key
Understand Custom Dimensions & Metrics — event-scoped vs user-scoped, how to register them in Admin before they appear in reports
Key
Know how to use DebugView for real-time event verification during implementation
Nice
4
Reporting & Explorations (heavily tested)
Day 4–5 · 3–4 hours
Know all 6 Exploration types: Free-form, Funnel, Path, Segment Overlap, User Lifetime, Cohort. Know WHEN to use each one.
Must
Navigate the Standard Reports: Acquisition (User vs Traffic), Engagement, Monetization, Retention, Demographics, Tech, Realtime
Must
Build a Funnel Exploration in the demo account — know open vs closed funnel, and what "elapsed time" between steps means
Must
Understand that Realtime Report shows last 30 minutes of data (exam trick question)
Key
Understand how to customize reports and use the Library section to publish/organize report collections
Nice
5
Attribution, Audiences & Integrations
Day 5–6 · 2–3 hours
Default attribution model = Data-Driven (DDA). Uses ML to distribute credit. Know the alternatives: Last click, First click, Linear, Time decay, Position-based.
Must
Know the Conversion Paths report (in Advertising section) and Model Comparison report
Must
Understand Audience Builder: conditions, duration (max 540 days), and Audience Triggers (events that fire when a user joins an audience)
Key
Know Predictive Audiences: "Likely 7-day purchasers", "Likely 7-day churning users", "Predicted 28-day top spenders" — powered by Google ML
Key
Integrations to know: Google Ads (remarketing + auto-import), BigQuery (raw SQL access), Search Console, Looker Studio, Firebase
Key
6
Privacy, Compliance & Final Review
Day 6–7 · 1–2 hours
Consent Mode v2: what it does (adjusts data collection based on user consent), behavioral modeling, and when to use it
Must
Data Retention: default is 2 months; can be extended to 14 months. Affects Exploration reports (not standard reports).
Must
Google Signals: enables cross-device reporting and demographic data by linking GA4 to signed-in Google users who opted into personalized ads
Key
Property deletion: 35 days before permanent deletion (exam trick question)
Key
Do at least 2 full practice tests from open-exam-prep.com and practicetestgeeks.com before sitting the real exam
Must
Click a topic card to expand key concepts. Weight = estimated % of exam questions.
Event-Based Data Model
~25%
- Everything is an event — replaces UA's session+pageview model
- Automatic events: page_view, session_start, first_visit, user_engagement
- Enhanced Measurement (toggle in Admin, no code): scroll, click, video_*, file_download, form_submit, site_search
- Recommended events (e-commerce): purchase, add_to_cart, begin_checkout, view_item
- Custom events: coded manually, max 500 per property
- Event parameters: max 25 per event, 100 chars each
- User properties: up to 25, persisted across sessions
Explorations
~20%
- Free-form: flexible pivot/table, like a custom report
- Funnel: step-by-step conversion flow (open = any entry point, closed = must start at step 1)
- Path: tree graph of user navigation sequences
- Segment Overlap: Venn diagram of up to 3 segments
- User Lifetime: LTV analysis by acquisition cohort
- Cohort: retention by acquisition group over time
- Explorations are private by default, can be shared
Key Events & Attribution
~18%
- Key Events = what was "Conversions" before 2024. Any event can be marked.
- Default attribution: Data-Driven (DDA) — ML assigns credit across touchpoints
- Other models: Last click, First click, Linear, Time decay, Position-based
- Conversion Paths report: under Advertising section
- Model Comparison report: compare attribution models side by side
- Default lookback window: 30 days (key events), 90 days (non-key event)
- Attribution applies to channel/campaign credit, not to event counting
Audiences
~15%
- Build on: events, event parameters, user properties, demographics
- Max membership duration: 540 days
- Audience Triggers: log a custom event when user joins/leaves audience
- Export to Google Ads for remarketing campaigns
- Predictive audiences: Likely 7-day purchasers, Likely 7-day churners, Top 28-day spenders (requires ML eligibility)
- Audiences are not retroactive — only include future users
Property Setup & Config
~12%
- Structure: Account → Property → Data Streams
- Data Streams: Web (Measurement ID: G-XXXXXXX), iOS, Android
- Data Filters: Internal Traffic filter (testing → active states)
- Custom Dimensions: event-scoped or user-scoped, must register before use in reports
- DebugView: real-time event stream for testing (enable via GTM preview or ?gtm_debug)
- Property deletion: 35 days recovery window
- User roles: Administrator, Editor, Analyst, Viewer
Privacy & Compliance
~10%
- Consent Mode v2: GA4 adjusts collection when users decline consent; uses behavioral modeling to fill gaps
- Data Retention: default 2 months, can extend to 14 months — affects Exploration reports only
- Google Signals: cross-device + demographics; requires users signed into Google who consented to ad personalization
- User-ID: assigns a consistent ID across devices for logged-in users
- Data thresholding: applied when Google Signals active to protect small user groups
- User data deletion: request via Admin for specific users
All
Events
Reporting
Attribution
Privacy
📋 Exam Fast Facts
Questions
50 multiple-choice
Time
75 minutes (~90 sec/q)
Pass score
80% = 40/50 correct
Cost
Free · Retake after 1 day
Valid
12 months
Open book?
YES — GA4 allowed in another tab
⚡ Automatic Events (memorize)
page_view
Every page load / history change
session_start
New session begins
first_visit
First time on site/app
user_engagement
10+ sec active on page
🔍 Exploration Types — When to Use
Free-form
Custom pivot tables, any dimensions/metrics combination
Funnel
Step-by-step drop-off analysis. Open = any entry; Closed = must start at step 1
Path
What did users do before/after a specific action? Tree visualization
Segment Overlap
How much do 2–3 audience segments intersect? (Venn diagram)
User Lifetime
LTV by acquisition channel, retention over user's lifetime
Cohort
Group users by when they first visited — track behavior over weeks
🎯 Attribution Models
Data-Driven (default)
ML-based, distributes credit based on actual conversion patterns across touchpoints
Last Click
100% credit to the last channel before conversion
First Click
100% credit to the first channel in the path
Linear
Equal credit to all touchpoints
Time Decay
More credit to touchpoints closer to conversion
Position-Based
40% first, 40% last, 20% distributed to middle touchpoints
⚠️ Classic Exam Trick Questions
Realtime shows
last 30 minutes (not 60 min, not 8h)
Property deletion
35 days recovery window (not 60, not 30)
Data retention affects
Explorations only (not standard reports)
Default retention
2 months (extendable to 14 months)
Conversions renamed to
Key Events (2024 update)
Audience max duration
540 days
Default attribution model
Data-Driven (not Last Click)
Conversion Paths location
Under Advertising section (not Reports)
🔗 Key Integrations
Google Ads
Import Key Events, share audiences for remarketing
BigQuery
Raw event export → SQL queries for advanced analysis
Search Console
Organic search queries in GA4 reports
Looker Studio
Custom dashboards and data blending
Firebase
App + web data unified (same GA4 property)
Merchant Center
E-commerce product performance data