Why I don't use Google Analytics
Respecting your visitor's privacy
There are too many trackers on the modern web. Every website you visit is collecting information about you and your browsing habits. Where you go. What you look at. How long you spend looking at it. What device you're using. This has become a hidden part of the web which we accept as the price of free. I'm not against tracking, I'm a software engineer who works on a website where we track our users to some extent too. I am against excessive tracking that you have little control over, both as a user and a site owner. So why don't I use Google Analytics?
Google Analytics used to be my go-to tool for tracking and collecting data about people visiting my website. It's free, easy to add to your site and provides a lot of information about your visitors. Over time, I've moved towards alternative analytics providers. This is driven by my growing concern over privacy and my use cases for analytics and tracking.
Privacy is a major concern in the modern web. Trust in platforms like Facebook has been hurt1, due in part to data leaks and large-scale scandals like Cambridge Analytica2. By offering free analytics services, companies like Google and Facebook can gain even more information about individuals. More data that can be sued to sell advertising and more data that can be leaked.
Google Analytics offers even greater precision with Cross Device tracking with Google signals. Allowing tacking of users between devices and sessions. Following users wherever they are signed in to a Google account. This is why visiting a website using Google Analytics or AdWords can often lead to advertising following you everywhere you go. Advertising and capture of user data to power even more personalised advertising, is rampant and we should take a stand.
By opting out of Google Analytics for my blog and other websites I own, I can minimise the amount of data I send Google about people I serve. I do track users on Benevolent Bytes but I do so with a self-hosted Fathom instance.
Google Analytics is popular because it's comprehensive. You can see how many people from each demographic. Their age, they gender, their interests. See how they interact with every link and image on your site. Find out which devices they use and where they've come from. It's a lot. It can be really helpful to some of this information, as a developer it's useful to see how someone interacts with parts of your application.
As a blogger, I don't really care. I don't need that information. I maintain a simple blog with links between pages. I don't care if a visitor clicked a button or navigated to a page from a particular link. I want to create content for other's to consume at their leisure. I don't want to optimise the click through rate. I don't need to know what percentage of my visitors are male or female or if they're interested in hiking.
If they're interested in what I have to say, they'll visit a page. They might even share it online. All I care about is roughly how many visits I've had on any particular page. It's useful for me to know an approximation of how popular certain topics and posts are so I can tailor my content to the audience. It's also nice to see where people came from, if I repost to the DEV community and I get a more traffic from there. It informs my strategy. That's it.
I'm not advocating for removing analytics from your website all together. The insight you can gain from analytics is invaluable. I am advocating for taking some time to reflect on what you want to gain. If you're like me and just want some basic information about visitors like which pages are the most popular and therefore what you should write more about, see if there are some better alternatives out there.
I also recommend pairing down analytics if you're using more than one, it's not only taking more of your visitors data that you might not need, extra trackers add bloat to your site. Figure out what you want to know and the bare minimum you can get by with. Respect your visitor's privacy and don't give Google any more data.