GA4 Setup Nobody Taught You, From Events to Conversions Without the Headache

Why GA4 Confuses Everyone

Universal Analytics made sense to most marketers. You had sessions, pageviews, bounce rate, goals. It wasn't perfect, but the mental model was clear: a user visits your site, views pages, maybe completes a goal. Done.

GA4 threw that mental model out the window and replaced it with something fundamentally different. Everything is an event. A pageview is an event. A scroll is an event. A purchase is an event. A user clicking a button, watching a video, downloading a file — all events. There are no goals anymore, just events that you mark as conversions (now called "key events").

This sounds elegant in a product announcement. In practice, it means that everyone who spent years building intuition around Universal Analytics had to start over. And Google's documentation, while extensive, reads like it was written by engineers for engineers. The result: most GA4 implementations I audit are either half-configured or actively collecting bad data.

Let me walk through how to set this up properly, without the jargon and theory that usually makes these guides useless.

The Event Model Explained Simply

In Universal Analytics, the data model was hit-based. You had different hit types: pageviews, events, transactions, social interactions. Each hit type had its own structure and its own reporting.

GA4 unified everything into events. Every interaction is an event, and every event can carry parameters (extra information attached to it).

Here's the mental shift: instead of thinking about categories, actions, and labels (the old UA event structure), think about event names and parameters.

Example — old UA approach:

• Category: Video
• Action: Play
• Label: Product Demo Video

• Event name: video_play
• Parameters: video_title = "Product Demo Video", video_duration = 120, video_provider = "YouTube"

Events That Actually Matter vs The Noise

GA4 automatically collects a bunch of events out of the box. Some are useful. Some are noise. Knowing the difference saves you from drowning in data.

Automatically collected events worth paying attention to:

• user_engagement (fires when the page is in focus — too broad to be actionable)
• form_start (sounds useful but fires on any field focus, including accidental ones)

Setting Up Custom Events Properly

The default events won't cover your specific business actions. You need custom events for things like:

• Form submissions (the real ones, not just field focus)
• Add to cart
• Checkout steps
• Account creation
• Contact button clicks
• Specific page views that indicate intent (pricing page, demo page)

Method 1: GA4 Interface (Simple Events)

Go to Admin > Events > Create Event. This works for simple events based on existing events. For example, you can create a "pricing_page_view" event that fires when page_view fires and the page_location contains "/pricing".

This method is good for: creating events based on page views of specific URLs, modifying existing events, simple conditions.

This method is bad for: tracking complex interactions, button clicks, form submissions, anything that requires JavaScript logic.

Method 2: Google Tag Manager (Everything Else)

For anything beyond simple URL-based events, use Google Tag Manager (GTM). This is where most of your custom tracking will live.

Setting up a form submission event in GTM:

Naming conventions matter enormously. Establish rules before you start creating events:

• Use snake_case (form_submit, not formSubmit or Form Submit)
• Be specific but not too long (contact_form_submit, not contact_us_page_form_submission_complete)
• Use consistent prefixes for related events (checkout_begin, checkout_shipping, checkout_payment, checkout_complete)
• Document every custom event in a spreadsheet with the event name, parameters, trigger conditions, and business purpose

Marking Key Events as Conversions

In GA4, any event can be marked as a key event (conversion). But just because you can doesn't mean you should. Marking too many events as conversions dilutes the signal and confuses your optimization.

Events that should typically be key events:

• Purchase / transaction complete
• Lead form submission (the thank-you page or actual submission, not form start)
• Phone call click (if calls are valuable to your business)
• Demo or trial signup
• Account creation

• Page views (even important pages — use these in audiences instead)
• Scroll events
• Video views (unless video completion is your actual business goal)
• File downloads (unless your business model is content downloads)

Go to Admin > Events. Find your event in the list. Toggle the "Mark as key event" switch. That's it. But remember: it only counts events going forward, not historically.

Important detail: key events in GA4 and conversion actions in Google Ads are now separate concepts. If you want your GA4 key events to be used for Google Ads optimization, you need to import them into Google Ads separately. Go to Google Ads > Goals > Conversions > Import > Google Analytics 4 properties.

Linking to Google Ads and DV360

This is where most of the value lives for advertisers. Without proper linking, your ad platforms and analytics are speaking different languages.

Google Ads Linking

Go to Admin > Product Links > Google Ads. Click "Link" and select your Google Ads account. Enable auto-tagging in Google Ads (should already be on).

What linking gives you:

• Google Ads campaign data appears in GA4 reports
• You can import GA4 key events as Google Ads conversions
• You can build GA4 audiences and share them to Google Ads for targeting
• You can see the full user journey from ad click to conversion

DV360 Linking

Admin > Product Links > Display & Video 360. Link your DV360 partner or advertiser. This requires admin access on both platforms.

What DV360 linking gives you:

• DV360 cost data in GA4 reports
• Audience sharing to DV360
• Cross-channel analysis between DV360 and other channels in GA4

Building Useful Explorations

GA4's standard reports are... fine. They give you the basics. But the real analytical power lives in Explorations (formerly Analysis Hub). This is where GA4 actually becomes useful for making decisions.

Exploration types worth knowing:

Free-form exploration is your workhorse. It's basically a custom pivot table builder. Use it for:

• Breaking down conversions by source/medium and landing page
• Analyzing user behavior by device category
• Comparing conversion rates across different user segments

• Checkout funnel analysis (how many people add to cart vs begin checkout vs purchase)
• Signup flow optimization
• Content consumption patterns

• Understanding navigation patterns
• Finding unexpected paths to conversion
• Identifying pages that cause exits

Building a basic conversion analysis exploration:

This single exploration will tell you more about your campaign performance than GA4's default reports ever will.

Audiences for Remarketing

GA4's audience builder is powerful but underused. Build audiences here, share them to Google Ads and DV360, and use them for remarketing.

High-value audiences to build:

Cart abandoners: Users who triggered add_to_cart but not purchase within the same session. This is remarketing 101 but you'd be surprised how many GA4 setups don't have this audience configured.

Engaged visitors: Users who visited 3+ pages, spent more than 2 minutes on site, and scrolled past 50% on at least one page. These are people who showed real interest but didn't convert.

High-intent page visitors: Users who visited your pricing page, demo page, or contact page but didn't convert. These audiences tend to have the highest remarketing ROI because the intent signal is strong.

Past converters (for exclusion or upsell): Users who completed a purchase in the last 30 days. Use this for exclusion (don't waste retargeting budget on people who just bought) or for upsell campaigns with different messaging.

Important: Audiences only populate going forward from when you create them. Create your audiences early, even before you need them for campaigns. They need time to build up user volume.

Common GA4 Mistakes That Corrupt Your Data

I audit GA4 implementations regularly and see the same mistakes repeatedly. Here are the ones that cause the most damage:

Duplicate event tracking. The most common culprit: having enhanced measurement enabled for page_view AND a separate GTM tag also firing page_view. Every pageview gets counted twice. Your traffic appears to double overnight. Check for this by looking at the page_view count in Realtime — if it seems suspiciously high, you probably have duplicate tracking.

Not filtering internal traffic. Your team visits your site hundreds of times a month. If you're not filtering that out, your analytics include fake sessions, artificially inflated pageviews, and test conversions. Set up internal traffic filtering: Admin > Data Streams > your stream > Configure tag settings > Define internal traffic. Then activate the filter in Admin > Data Settings > Data Filters.

Collecting PII in event parameters. GA4's terms of service prohibit collecting personally identifiable information (email addresses, phone numbers, names) in event parameters. It's tempting to pass form field values as parameters, but if those fields contain PII, you're violating the terms and risk having your property shut down.

Not setting up cross-domain tracking. If your marketing site is on www.example.com and your checkout is on checkout.example.com, GA4 treats these as separate sites by default. Users who click from your site to checkout look like new referral traffic, and their conversion journey gets broken. Set up cross-domain tracking in your data stream settings.

Ignoring data retention settings. GA4's default data retention is 2 months for event data in explorations. That means your explorations can only look back 2 months unless you change this. Go to Admin > Data Settings > Data Retention and set it to 14 months (the maximum for free GA4).

Using too many custom dimensions. GA4 free allows 50 event-scoped custom dimensions and 25 user-scoped ones. That sounds like a lot until you start using them for everything. Plan your custom dimensions carefully and don't create ones you won't analyze.

Making GA4 Work for You

GA4 isn't going anywhere, and once you get past the initial confusion, it's genuinely more capable than Universal Analytics was. The event-based model gives you flexibility that sessions and pageviews never could. The audience builder is excellent. Explorations are powerful analytical tools once you learn them.

The key is investing the setup time upfront. A well-configured GA4 property with clean event tracking, proper conversion marking, platform linking, and thoughtful audiences will serve your business well. A hastily set up property with default settings and no custom events will give you numbers that look like data but don't actually help you make decisions.

If your GA4 setup feels half-baked, or if you're looking at numbers you don't trust, you're not alone — and it's worth fixing before you make another budget decision based on unreliable data. AdCharta helps brands set up and audit their analytics properly, from event architecture to conversion tracking to platform integrations. If you want someone to look at your GA4 setup with fresh eyes, get in touch.

Ready to Grow Your Ad Performance?

Get a free audit of your current advertising setup and discover untapped growth opportunities.