A while ago, after years of being "analytics-free", I decided to try Plausible Analytics and I want to share what led me to start thinking differently.

The original provocation was learning that Photopea, despite financial success as a one-person operation, earns most of its revenue from advertising and only four percent of its revenue from subscriptions or memberships. Considering that I am trying to finance my own sustainability directly via the people using the app, I thought it curious that this income equalled a sort of 'minimum wage' despite being a well-known, high-traffic project. I don't intend or know how to make something as complex and deep as a Photoshop clone, and so I wondered if I would have fewer opportunities than this—perhaps I need to be open to selling ads at some point in the future, if that's what it takes to keep content freely accessible on the web. (On the other hand, I believe it's better to avoid comparing yourself to others and complexity is not synonymous with income, but I'm not sure how to think about all this at the moment).

Another reason is that I had a hard time turning garbage numbers like 'requests' (which include bots and counts multiple files for each pageload) into something that gives me an idea of 'how many people are actually looking at this?'. I believe in talking to the people who use what you make, but I think many (or most?) people don't have time to write in their impressions, and so this will always be smaller by a magnitude you cannot know; there is value in passive feedback and I think most people would prefer this if it's done with purpose.

So far it feels good. I aggregate visitors to various projects into a single picture. It's nice to have more meaningful numbers and to find out about interesting places that link to you, like this Brazilian guy's tech blog. One of my 100 steps to success, still in-progress, is to track your growth over time, and this is another way to do that.

I hesitated all these years because Google Analytics has become a form of surveillance capitalism, and this taints most other approaches to analytics (or at least our perceptions of it). How can this be remedied? Does it help to use open-source technology? Or if the company hosting the technology aligns with your values? Collecting data generally requires trust because one cannot verify beyond the 'privacy policy'—what about being transparent and just showing what one collects?

Similar to Buttondown, which I use to send mailing lists, I dig the spirit behind the project and its team and would like to lend my support. All this feels holistic so far, but I'm open to having my mind changed again. What do you think?

Follow my journey on Twitter (or via the mailing list)