Why I Stopped Installing Google Analytics 4 on Client Sites
I have spent fifteen years building things for the web. I have written thousands of lines of tracking code, configured hundreds of dashboards, and spent way too many late nights trying to make sense of marketing data.
For most of that time, google analytics was the default. You created an account, grabbed the tracking code, pasted it into the header, and walked away. It worked. You got a clean list of pages, referral sources, and active users.
Then Google killed Universal Analytics and forced everyone onto google analytics 4.
I remember the exact moment I realized the industry had lost its mind. It was a Tuesday afternoon last autumn. I was sitting on a Zoom call with Dave, a long-time client who runs a regional logistics company. Dave is a smart guy, but he is not a data scientist. He just wanted to know how many people visited his "Get a Quote" page from a LinkedIn post we ran the week before.
Dave opened his browser, went to the google analytics login, and stared at the screen. He clicked. He scrolled. He looked at a blank "Explorations" canvas that looked like an empty Excel spreadsheet on a bad day.
"Where is the traffic?" he asked.
I had to explain to him that the data was delayed by up to 48 hours. I had to explain that we needed to build a custom report, register custom dimensions, and configure a custom path exploration just to see where his users came from. Dave had already spent $3,400 on an external consultant who promised to set up his google analytics ga4 property correctly, yet he still could not answer a basic business question.
That was the day I stopped recommending it.
The Stockholm Syndrome of Modern Web Tracking
We have been conditioned to believe that if a tool is difficult to use, it must be powerful.
Google has built an entire ecosystem around this complexity. People spend weeks studying for a google analytics 4 certification. They pay hundreds of dollars for a specialized google analytics course. They spend hours playing around in a google analytics demo account, trying to figure out why their conversion numbers do not match their actual stripe backend.
Think about how insane this is. You are taking a google analytics certification just to learn how to read a chart on your own website. You are wasting precious cognitive load trying to decipher technical documentation, searching for old google analytics mcp legacy partner terms, or debugging tag managers, instead of talking to your customers.
The truth is simple: google analytics 4 (ga4) was not built for you.
It was built for enterprise corporations with dedicated data engineering teams. It was built to feed Google's machine-learning ad engines. It shifts the tracking model from simple pageviews to complex, nested events. That is great if you are a multi-national retail giant tracking a complex cross-platform mobile app funnel. It is absolute overkill if you are running a service business, a SaaS, or a local shop.
The Cost of Bloat
When you load GA4 on your website, you are not just adding a script. You are adding weight.
You are loading a heavy JavaScript library that slows down your mobile page speed. You are forcing your visitors to agree to intrusive, ugly cookie banners just to comply with GDPR, because Google tracks users across the entire web to build advertising profiles.
And what do you get in return?
- Data sampling that guesses your traffic instead of counting it.
- Data retention limits that delete your historical records after 14 months unless you pay to export them to BigQuery.
- A UI that changes so often you need a new google analytics 4 certification every year just to find the exit rate metric.
You do not need a jet engine to power a scooter. You just need to know what is working and what is not.
A Cleaner, Honest Way Forward
After the disaster with Dave, my team and I started doing things differently. We stripped GA4 out of our client builds entirely.
Instead, we started writing simple, lightweight analytics scripts that run directly on our clients' servers.
We built a system that tracks exactly what matters: pageviews, referrers, device types, and custom conversion events (like a button click or a form submission). The script is tiny—under 3 kilobytes. It loads in milliseconds. It does not collect personal identifiable information, which means no annoying cookie consent banners are required for basic tracking. Best of all, the data belongs entirely to the business owner. No data sharing with ad networks. No delayed reporting.
When Dave logs in now, he sees a single page. He sees his visitor count, his top-performing blog posts, and his lead conversion rate. It takes him three seconds to understand his marketing performance. He hasn't looked at GA4 since.
If you are tired of fighting with bloated dashboards, losing your historical data, and spending your life trying to figure out basic web metrics, you do not have to accept Google's defaults. You can own your data again. We can build and integrate a clean, custom web analytics dashboard directly into your existing setup, giving you clear insights without the tracking headache.