PPC Unit
All articles
Analytics · 5 min read

Conversions in GA4: track money, not clicks

Why most analytics conversions are useless, which events truly matter and how to pass them into ad systems so campaigns earn profit instead of draining budget.

Yevhen Chesniuk
Conversions in GA4: track money, not clicks

Advertising without proper analytics is spending blind. You see clicks and impressions but miss the main thing: how much money each campaign brought. In this article we break down which conversions truly matter and how to set them up so advertising earns profit.

First — see the customer funnel

Before talking about conversions, you need to understand the customer journey. People don’t buy straight away: they visit the site, look at a product, compare, add to cart, hesitate, come back — and only then buy or leave a request.

Analytics should show this whole path. Then you see exactly where you lose people: on the product page, the cart, the form.

i The funnel is a diagnosis

Without a full funnel you optimise at random. Seeing the customer journey is half the job. The second, more important half — correctly defining what in this funnel is a real conversion.

The main mistake: counting everything

Most accounts track dozens of “conversions”: a button click, scrolling 100% of the page, a video view, a price-list download, time on site over a minute.

This isn’t bad in itself. Such micro-events are useful for behaviour analysis. But there’s a problem: when you mark them as conversions and pass them into advertising — you deceive your own campaigns.

Imagine: the Google Ads algorithm sees that a “conversion” is a page scroll. It starts looking for people who love scrolling pages. Budget goes to traffic that scrolls the site and leaves. Formally there are many conversions. In reality — no sales.

💡

The algorithm optimises exactly towards what you called a conversion. Call it a scroll — you get people who scroll. Call it a purchase — you get buyers.

What to actually track — revenue-generating actions

Advertising should optimise towards events that bring money. For most businesses there are only three types:

1
Purchases
For e-commerce — the purchase event with a mandatory order amount. The cleanest signal: there's money — there's a conversion.
2
Calls
A large share of sales in services and B2B happens by phone. Track clicks on the site phone number and set up call tracking for ad calls. A call is a hot lead.
3
Lead forms
A request form submission, a messenger enquiry, a callback request. The generate_lead event.

These three — purchases, calls, lead forms — are your real conversions. The rest is supporting analytics.

✕ Useless for advertising
  • —Scrolling 100% of the page
  • —Watching a video on the site
  • —Time on site over a minute
  • —A click on any button
✓ Works for profit
  • +A purchase with an order amount
  • +A call (phone-number click / call tracking)
  • +A lead form submission
  • +A callback request

Pass the value, not just the fact

Even a correct conversion without value works at half power. The generate_lead event without an amount tells the algorithm only “there was a request”. It doesn’t distinguish a €20 request from a €2,000 one — and chases quantity.

Pass the value of every conversion:

  • E-commerce — the real order amount
  • Services — the average lead value (average order × lead-to-sale conversion)

This turns “optimise by quantity” into “optimise by money”.

How data reaches advertising

GA4 on its own is just a mirror. The power begins when conversions are passed onward, into the systems where advertising runs:

1
Site event → GA4
A purchase, call or request is recorded as an event in Google Analytics 4.
2
GA4 → Google Ads
Linked accounts — conversions become available to campaigns for optimisation.
3
Enhanced Conversions + offline data
For completeness — passing CRM data when a lead actually became a sale.
4
Meta and TikTok via Conversions API
The same for social — server-side event delivery.

The logic is simple: the fuller the data the algorithm sees, the more precisely it finds paying customers.

Sharper
targeting — the algorithm sees real buyers
Less
spend on irrelevant traffic
More
profit from the same budget

Incomplete data works the opposite way: the algorithm “under-sees” results, undervalues good campaigns and drains budget on weak ones.

Google itself rewards correct setup

This isn’t only your interest — Google is invested in advertisers setting up conversions correctly. The better advertising works, the more the client buys it.

✓ What correct setup gives

Google raises the account’s Optimization Score for correct conversions and Enhanced Conversions, regularly runs promo programmes with ad credits for quality tracking, and unlocks smart bidding strategies (Target ROAS) that simply don’t work without correct conversions.

Briefly on technical cleanliness

! To keep data accurate

Pass a unique transaction_id — GA4 will discard duplicate purchases. Check every event in GTM Preview and GA4 DebugView before publishing. Don’t mix micro-events with real conversions in one list for advertising.

Conclusion

Tracking everything is easy and almost useless. Value comes from something else: clearly defining revenue-generating conversions (purchases, calls, lead forms), giving them real value, and sending that data into ad systems.

The fuller and cleaner the picture — the smarter the advertising, the less budget waste and the higher the profit. And as a bonus — a higher Optimization Score and ad credits from Google itself.

Want us to check what counts as a conversion for you and whether value is being passed? Get in touch — we’ll audit your analytics.

Need help with advertising or analytics?

We will audit your account and show growth opportunities. The first assessment is free.

Discuss a project