Website analytics has been in a slightly strange place for quite some time. At least in my opinion.
Almost everyone who runs a website wants to know how many people visit it, where they come from, what content they read, and whether they find what they came for. In many cases, you also want to know whether a particular ad led to a purchase and pass that information back to the advertising platform.
At the same time, this fairly simple need has grown into an enormous ecosystem of tracking, identifiers, cookies, consent banners, and interconnected systems. Understanding the whole setup can sometimes require more analytics expertise than even the marketing agency that sold you the entire stack actually has.
When I looked at my own websites, I started wondering: does it really need to be this complicated?
I began building simpler analytics for my sites, focused on answering the questions I actually cared about. I experimented with different approaches across different websites and eventually combined them into something that worked for me.
That turned into something more. I’ll come back to that a little later, but first I want to explain what I was actually trying to solve.
Does Anyone Visit My Website?
Sure, you can stare at the server logs scrolling by and conclude that something is happening. But that doesn’t exactly answer the question in the most useful way. I needed something a little more than logs.
So why not just use Google Analytics and call it a day?
User privacy matters to me. Every time I install a comprehensive analytics system on a website, it bothers me a little knowing that I don’t have a complete picture of exactly what data it collects and processes without sitting down for a lengthy session with its privacy documentation.
There’s nothing inherently wrong with analytics. Quite the opposite. If I build a website or an online service, I want to know whether people actually use it.
I want to know where visitors come from. Which pages interest them. Whether people find the site through Google, social media, or AI services. Whether a campaign is working. What kinds of devices people use to access the site.
And if nobody is interested in a particular page, that’s useful information too.
But I don’t need to know who an individual visitor is, whether they’re male or female, or other details like that.
Gender can, of course, be useful information when marketing certain products. But is it really as important as we often assume? You could debate that endlessly, and there probably isn’t a single right answer. That’s a whole different discussion.
How Much Do You Really Need to Know About a Visitor?
Over the years, analytics has evolved into increasingly detailed tracking. Rules and laws are introduced to protect privacy, while websites add banners, consent forms, and lengthy explanations about where visitors’ data may be sent and how many “partners” it may be shared with.
That doesn’t sound like the experience I want for people using my services. In today’s world, it’s impossible to avoid all forms of tracking, but we can at least think about it and try to make our own small corner of the internet a little more controlled.
Over the years, web analytics has grown into something much larger than its original purpose.
Cookies, user identifiers, browser storage, IP addresses, and various identification techniques make extremely detailed tracking possible. At the same time, this has created a new problem for the average website owner.
If all I want to know is whether 50 or 5,000 people read my article today, why should I build a system around that which tries to identify them beyond simply knowing that someone visited the page?
Why should the first thing a visitor has to do when arriving on a website be to decide whether I’m allowed to track them even further? In many cases, wouldn’t it be enough to know that people arrived, stayed for more than five seconds, and perhaps scrolled all the way to the bottom of the page?
And how many website owners actually need all the data that modern analytics systems are capable of collecting?
I think most of us can get by with much less, as long as what we do collect is accurate and genuinely useful.
How I Solved My Problem
I looked at various tools for analyzing how my websites were being used. Many of them came very close to what I was looking for, but none quite matched my needs.
A few years ago, I built the first version of my own solution. I connected some of my websites to it and started collecting only the data I actually needed. Does anyone visit the site? Where does the traffic come from? What do people read? What works?
Over time, I gradually removed the analytics solutions I had previously used on my websites.
Eventually, the service grew into something of its own, and the various silly project names gave way to a permanent one. It became Metrimato.
Metrimato was built from the ground up with respect for individual privacy.
Metrimato does not store a visitor’s IP address or browser user agent. It does not set cookies, build browser fingerprints, or store persistent user identifiers.
Visitors can be distinguished from one another within the same day using an identifier based on a random value that changes daily. The next day, the same person appears to Metrimato as a new visitor.
This means visits cannot be linked from one day to the next or across different websites. That is an intentional limitation.
Metrimato isn’t supposed to know who you are or where else you’ve been. It’s supposed to know what happened on the website.
I can already see a finger going up: doesn’t this completely break the ability to track whether an ad actually worked?
Yes. At least some of it.
If an ad is supposed to cling to a user’s browser like a burr and still be able, weeks later, to tell various parties that this person previously clicked an ad and has now made a purchase, I think it’s fair to ask how necessary that really is.
If an ad is genuinely good, it brings someone to the website. If the site and the offer are a good match, that person may also end up making a purchase during the same visit. I can measure that without tracking someone for weeks.
If the purchase decision takes days or weeks of consideration, however, the question becomes more interesting. Was it really that particular ad from weeks ago that ultimately led to the sale? Maybe it started the journey, but many other things could have influenced the decision along the way.
Attribution is useful. I’m just not convinced that it requires tracking every visitor for as long as possible.
You Can Still Get a Surprising Amount Out of Analytics
Protecting privacy doesn’t mean analytics has to be nothing more than a visitor counter.
Metrimato can show you things like pages and navigation paths, traffic sources, search engines, social media, UTM campaigns, devices, browsers, and operating systems.
It also includes reading time, scroll depth, exit pages, and bounce rate.
I also wanted to include something that has become increasingly relevant on today’s web: AI traffic.
Metrimato distinguishes between traffic arriving from AI services and AI crawlers that visit and read your website. This is becoming increasingly interesting as people discover content not only through Google, but also through ChatGPT and other AI services.
When needed, you can also send custom events from your website, such as form submissions, downloads, or clicks that lead to a purchase.
So the idea isn’t to collect as little data as possible.
The idea is to collect as much useful data as possible without needing to track people.
When It Comes to Privacy, the Best Personal Data Is Data That Was Never Collected
This was one of the most important reasons I built Metrimato.
When we talk about data security, the focus is often on how well the data we collect is protected. And of course, that matters.
But an even more effective way to protect data is not to collect it in the first place.
If a system doesn’t store visitors’ IP addresses, user identifiers, or persistent identifiers, that data can’t later be leaked, misused, or combined with other datasets.
I Also Wanted Analytics I’d Actually Use
The other reason Metrimato came to be is much more practical.
Analytics software is often far too complicated.
There are endless reports, views, settings, and concepts to deal with. It’s easy to end up in a situation where your analytics system collects an enormous amount of data, but nobody actually looks at it.
At that point, it isn’t particularly useful.
With Metrimato, I wanted to approach things differently.
I open the service and see what’s happening across my websites.
If I have an online store, a company website, a blog, and a campaign landing page, I want to see them all in one place. If something catches my attention, I click on it and dig deeper.
Analytics should help you understand your website, not require you to study analytics.
I Wasn’t Trying to Rebuild Google Analytics
This is perhaps an important distinction. The goal of Metrimato isn’t to replicate Google Analytics feature by feature, or any other analytics service for that matter.
If you need user-level tracking connected to advertising networks and highly advanced marketing automation, Metrimato may not be the right tool for you.
But a huge number of businesses, organizations, developers, content creators, and online stores ultimately just want answers to much simpler questions:
Do people visit my website?
Where do they come from?
What do they look at?
What works?
What doesn’t?
These questions can be answered without building a profile of the person on the other side of the screen.
Now I Need Real Users to Put Metrimato to the Test
Metrimato has now reached the point where I want to see it being used on real websites.
My own use and development environments can only reveal so much. Real-world websites, WordPress installations, online stores, browsers, and different setups introduce new challenges and uncover things I simply can’t test at the same scale on my own.
I’m now looking for the first users to join the Metrimato testing group.
If you run a website, blog, online store, or several websites and you’re interested in privacy-respecting analytics, you can apply to join.
The purpose of the testing group isn’t simply to confirm that everything works.
I want to hear what’s unclear, what’s missing, what’s annoying, and where Metrimato still doesn’t do what it should.
You can learn more about Metrimato and apply to join the testing group at metrimato.com.
At its core, Metrimato is built around one simple idea:
Measure your website, not your visitors.