Marketing analytics: GA4, attribution and testing

Set goals that matter, track them in GA4, read attribution honestly, test changes properly, and report results each month.

Most small businesses have more marketing data than they can use and still cannot answer the simple question: which of this is working. GA4 was installed by someone who left, the ad platforms each claim credit for the same sale, and the monthly report is a screenshot of follower counts. This course starts with the decisions you need to make, then builds the measurement to support them: clean UTM tags, GA4 events you have tested, honest attribution, a dashboard people read, tests large enough to trust, and the unit economics that say whether a channel deserves more money. It suits owners, freelancers and junior marketers who have to explain results.

What you'll be able to do

• Write a measurement plan that links business goals to a short list of KPIs
• Build tagged links with a UTM naming convention a team can follow
• Install GA4 and set up key events tested in DebugView
• Answer marketing questions from GA4 reports and explorations
• Explain why GA4 and ad platforms report different numbers and pick a source of truth
• Build and share a Looker Studio dashboard from your own data
• Plan an A/B test with a realistic sample size and write up the result
• Calculate acquisition cost, customer value and payback, and present a monthly report

Syllabus

Module 1: Decide what to measure before you measure anything

How to turn a business goal into a short list of KPIs, tell useful numbers from vanity metrics, and write a measurement plan before you touch any tool.

• 1.1 A number is only worth tracking if it changes a decision
• 1.2 From business goal to marketing KPI
• 1.3 Write a measurement plan on one page

Module 2: Tag every link so you know where visitors came from

How UTM parameters work, how to name them consistently, and how to keep a shared tagging sheet so campaign data stays clean in GA4.

• 2.1 What UTM parameters are and why analytics needs them
• 2.2 A naming convention your whole team can follow
• 2.3 Tagging mistakes that break your reports
• 2.4 Build your UTM tagging sheet

Module 3: Set up GA4 and track the events that matter

How GA4's events-based model works, how to install it, which events it records automatically, and how to mark the actions that matter as key events.

• 3.1 GA4 counts events, not page views
• 3.2 Install GA4 with the Google tag or Tag Manager
• 3.3 Mark the actions that matter as key events
• 3.4 Set up and test your key events

Module 4: Read GA4 reports without getting lost

Which GA4 reports answer which questions, what the main metrics mean, and how to use explorations for questions the standard reports cannot answer.

• 4.1 The five questions the standard reports answer
• 4.2 Users, sessions and engagement in plain words
• 4.3 Explorations for questions reports cannot answer
• 4.4 Answer five real questions from your GA4 data

Module 5: Give credit to channels fairly, and know the limits

How attribution models split credit for a sale, why GA4 uses data-driven attribution by default, why ad platforms overclaim, and what attribution cannot tell you.

• 5.1 Attribution models and what each one rewards
• 5.2 Why Meta, Google Ads and GA4 never agree
• 5.3 What attribution cannot see
• 5.4 Write your attribution notes for the business

Module 6: Build a dashboard in Looker Studio that people read

How to plan a dashboard around decisions, connect GA4 and spreadsheet data to Looker Studio, and lay out charts that a busy owner understands in a minute.

• 6.1 Start the dashboard from the decisions it supports
• 6.2 Connect GA4 and your other data in Looker Studio
• 6.3 Charts and layout a busy owner understands
• 6.4 Build and share your marketing dashboard

Module 7: Run A/B tests that tell you something real

How to write a testable hypothesis, why sample size decides what you can learn, what statistical significance means, and how to test when your traffic is small.

• 7.1 A test starts with a hypothesis, not a hunch
• 7.2 Sample size and significance without the maths
• 7.3 Testing when you do not have much traffic
• 7.4 Plan and run one A/B test

Module 8: Know what a customer costs and what one is worth

How to calculate customer acquisition cost, customer lifetime value and payback period from your own numbers, and use them to judge whether a channel is worth more money.

• 8.1 Customer acquisition cost, honestly counted
• 8.2 Lifetime value starts from margin, not revenue
• 8.3 Payback period: how long until a customer pays you back
• 8.4 Build a unit economics sheet

Module 9: Turn the numbers into decisions every month

A monthly reporting routine that takes an hour or two: check the data, read the dashboard, explain what changed, and agree a short list of actions.

• 9.1 Check the data before you trust it
• 9.2 Write a report that explains, not just lists
• 9.3 Close the loop on last month's actions
• 9.4 Run your first monthly marketing review

Frequently asked questions

How long does the course take?

About eleven hours across nine modules, including the exercises. Plan on five to six weeks, since you need a few weeks of GA4 data and a full test cycle before the final review.

Do I need to know statistics or code?

No. You need basic spreadsheet formulas. Sample size and significance are explained in plain words with a free calculator, and GA4 setup uses the Google tag or your website builder's integration.

Does it cover Universal Analytics?

No. Standard Universal Analytics properties stopped processing data in July 2023 and the course teaches GA4 only. If you have old reports, the lessons on events and engagement explain what changed.

How is this different from Paid ads: Meta, Google and TikTok?

Paid ads covers installing pixels and running campaigns on each platform. This course covers measurement across every channel: GA4, attribution, dashboards, testing and whether the spend pays back.

Explore more

• All Academy courses