How do you set up a conversion in Google Analytics 4 (GA4)?
You set up a conversion in GA4 by marking an event as a key event: Admin → Data display → Events and the "Mark…
Google Ads, Meta, GA4: none of them can read your intentions.
They learn from what you show them, and you only show them the final result. That's why they optimize for randomness, not for your business.
Zarezerwuj termin na bezpłatne naprawcze konsultacje.
Book a free callIf you measure visits to the site but not the moments when a user actually moves closer to a decision, you’re losing time. Not money. Time. Because you’re spending the money anyway, on campaigns, on website changes, on tests that „didn’t work out”. Metrics like these can make the marketing team feel good, but they give you nothing to base a decision on: they don’t show whether you’re heading toward more revenue or more conversions 1. Without events you lose decision time, the stretch between the question „what’s wrong?” and the answer „we know what to change”. Two short cases from practice
A home accessories store was getting steady traffic from its campaigns. The conversion rate was low, so the obvious verdict was: low-quality traffic. Once events were added, it turned out that:
Instead of cutting the campaigns, the team changed one part of the process: the copy on the problem page and the layout of its elements. The effect on the process:
Without events, the decision would have been logical. And wrong.
A web design company saw that one channel brought in the cheapest leads. With results that steady, scaling it was the natural move. But the events showed that:
The turning point was switching the optimization priority from „cost per lead” to „quality of intent”. The business effect:
There’s one pattern I see most often:
The result:
Here’s my position: if you measure the final result but not the steps that lead to it, the algorithms learn randomness, not intent. It doesn’t matter whether we’re talking about Google Ads, GA4 or Meta. They all work on the same principle: we optimize for what you report to us as important. The WordStream blog makes the point vividly: if your goal is conversions but you haven’t set any up, don’t expect smart bidding to deliver results, because the algorithm doesn’t know what to look for 2
This needs saying plainly. Events are not a priority if:
In that case events won’t „fix” anything. They’ll just show the chaos in higher resolution. Adding dozens of events to the code „because GA4 recommends it” won’t help if the company isn’t clear on why it collects the data. As practitioners point out, you shouldn’t track everything in sight, because that’s the shortest route to drowning in a sea of data you don’t need 3
I’m not writing this as a „GA4 specialist”. I’m writing it as someone who:
Events were never a technical problem. They were always a matter of answering one question: what really matters in this business.
Before you go back to your campaigns, ask yourself one question: Can I point to one specific moment on the site after which I know the user is genuinely closer to a decision, or that they’ve just abandoned it? If not, you don’t have an analytics problem. You have a problem with the language your business uses to talk to decisions and algorithms.
How an event differs from a conversion, and why adding to cart isn’t a sale yet, is something I explain in the article what an event is in web analytics. If you want to work out which moments to measure in your store or business, here’s what analytics and tracking setup looks like.
Yes, as long as the events are limited to a few key decision moments rather than dozens of technical interactions. For most online stores and lead-gen businesses, 5–8 well-chosen events are enough, each answering the question „did the user take a step closer to a decision”. The rest of the data very often just distracts and slows decisions down.
The simplest test: would losing this event change a business decision?
If its absence doesn’t affect what you do with your budget, offer or sales process, it probably isn’t a decision event. Decision events are the ones after which you change direction, not the ones that just add a line to a report.
No. The event structure always follows the sales model, not the tool. Two stores in the same industry can need completely different events if they differ in how long the decision takes, product price, the role of content or contact with a salesperson. Universal event checklists are technically convenient, but they rarely hit the real decision points of a business.
Footnotes:
Author - Michał Wojciechowski
I have run Google Ads accounts myself since 2019 and I create the ads and copy for them. I have run campaigns on 9 markets and worked on projects on 8 more.
I started in graphic design and film, so I look at an ad the way a viewer does. I work directly with business owners: regular calls, decisions based on data, and work on the offer and the landing page, not only on the ads.
One hour on a call is usually enough to see why the sales are not coming in.