November 4, 2020

Adding analytics with Plausible Analytics

I've added Plausible Analytics to the main disciple.dev site; here's what it is and why.

Adding analytics with Plausible Analytics

Public site statistics for this website are now available at stats.disciple.dev.

Indroducing Plausible

A few months ago, I came across Plausible Analytics as a powerfully simple alternative to Matomo and Google Analytics. I was hosting my blog on WordPress.com Personal, and WordPress' rudimentary Jetpack got me by... basic stats, although still not great to work with.

Now I've moved this site to Ghost, a powerful headless CMS built on NodeJS which I have loved for years. It's made me want to write again, and a chance to try out Plausible, as well. So I took the time today to set up docker-compose in my DigitalOcean droplet where this site lives, followed the self-hosting setup, and after a bit of juggling ports with a nginx reverse proxy and setting up SSL with Let's Encrypt, we're in business!

Why Plausible?

This is a good question, and I want to be thorough.

Website analytics is a primary marketing tool in today's world: it is a strong indicator of how effective your content is in accomplishing the goals you have for it. If you're selling Alpaca hand-knit glittens on Etsy, you want to know how many people are viewing your mittens and how many poeple are buying them, which colors they like the most, and whether people are sharing your product on Facebook or Pinterest. That's analytics.

#1: Possible to not Google

For most people, if they have their own website, Google Analytics is the de facto standard when it comes to measuring web sites. It is a powerful tool and tighly integrated with Google's other popular services. I won't lie, I was tempted, a little.

But I have issues with Google, and that is primarily their main business model: selling people's data. This is why all of Google is "free": it tracks everything it can about what you do on the internet, all your habits of sites you visit and things you do, people you talk to, things you shop for. All of Google's product exist as giant funnels of your data in to the Big Black Box, from which Google tailors ad services to you; advertisers pay Google to use that tailored data to specifically target you with ads.

I'd rather pay for a service like ProtonMail outright than the pseudo-free Gmail, knowing that Gmail is trying to push me and my data more and more into Goolge's maw.

#2: Perfectly Lightweight

Sure, I'm injecting a little JavaScript in my site to run Plausible—a whopping 1.28kb of it. That is a fraction of the size of Google Analytics, and I'm loading it off a domain that I control.

It runs unobtrusively, it tells me when you stopped by, and it doesn't tell Google or Facbook or the Chinese Communist Party your were here.

#3: Plausibly Privacy Focused

I don't need to know your mother's maiden name (for analytics purpose, anyways). I don't need to know your thneed shopping habits. I don't need to know your IP address. There are no cookies. No cross-site tracking.

I know you were here, that's it. I don't know anything about you. Welcome!

#4: Pleasantly Open-Source

Go ahead, check it out. Contribute. Make it better! Nothing to hide.

#5: Pay or not—your choice

Choice is important. I like the challenge of self-hosting and let's be honest, I don't pay much at all to keep this blog running. That's the beautify of this combination, I don't need to sell my digital soul for "free" Google analytics.

I have the means to host Plausible myself, but many may not. That's fine! I highly encourage those with the need for solid, simple analytics to start a free trial, and check it out, and do support the growth of a new, powerful platform.If my site magically explodes in longevous popularity, I'll pay, too. Because it's good to pay money for quality services.