A few days back, we removed website analytics from WaniKani. You know, the one made by the “Big G.” You know, Google Analytics (honestly, pretty afraid to call out the name in case an automated smite from “G” befalls us). This is all part of a broader effort to remove third-party applications and trackers from our websites. We’re still working on it, but this was one of the big ones.
Why do this? It’s a couple things:
One is we (the people at WaniKani) don’t like to be tracked by these kinds of things ourselves. Quite a few of us are the type who actively take steps to prevent that sort of thing. So, why are we doing something to you that we don’t want for ourselves? Let’s have some integrity, here, and stop. There’s no reason we need to see your age, location, clicks, conversions, and the like.
The second reason is a business / philosophical one.
The quick version is that these web analytics weren’t doing anything for us. Over the years I actually suggested we remove them. Every time, a few people would come in and say “Well, it’d be nice if…” or “I might want to see if…” Then every year, it’d turn out nobody was doing those things.
Which is good, I suppose! The metrics available in these types of things are not metrics we even care about. It’s like a social media “Like” button. A lot of “Likes” means a post was “good.” Not a lot of “Likes” means a post was “bad.” So, when we write a post, the goal inevitably becomes “how do I make a post that gets thousands of Likes?” It just changes how you do things. You no longer do it for its intrinsic purpose; you do it with the goal of “getting more Likes,” which are two very different things. Perhaps there’s some correlation between “good quality” and “lots of Likes”? I don’t know and I don’t care, but it certainly doesn’t feel like “quality” is the driver of social media engagement in the year of our Crabigator and Savior, 2026.
Grading (e.g. in school) is a pretty good metaphor for what we’re doing here as well. The moment you start assigning grades to something, you are quantifying the quality of it. And so, the goal of the student shifts from “attempting to learn” to “figuring out how to get a good grade for as little work as possible.” That’s what I certainly did the moment I realized our geometry teacher in 8th grade wasn’t actually looking at our homework and would give full marks if you did something that looked vaguely like math. Humans, I think, feel intrinsically rewarded from learning and puzzle solving, but grades slowly strip that joy away until there’s nothing left.
I could go on and on. The point of all this is: metrics are an incredibly forceful and persuasive guide.
The technology known as “G’s Website Analytics” encourages certain things over others. It’s not doing it on purpose. It’s just a thing, after all. But, in using the tool we are steered in a direction that it encourages. Neil Postman once said: “embedded in every tool is an ideological bias, a predisposition to construct the world as one thing rather than another… To a man with a computer, everything looks like data.” In this case, I say “To a man with Google Analytics, everything looks like page views and visitors and clicks and conversion rates.”
The problem is, those are not things we care about in the least. We haven’t cared about those in a long time. I certainly haven’t been like “I need to see how many people are visiting this page in order to make a decision about it.” That said, I do vividly remember Google Analytics when it first came out, thinking how amazing it was. Data data data! You installed it, you used it, and everyone else did too. And if you didn’t, you were bad at business. It was an edge you needed to survive. Hustle, baby. Even to this day, I imagine the decision to remove analytics would be shocking to some. Why would you even do that? If it was just about privacy, why not just switch to a more privacy-focused service? They are out there.
Well, call me a bad business boy, I suppose, because I can’t think of a single big-picture edge that web analytics has ever given us. It’s fun to look at, I guess? But it’s never going to tell me if someone’s learning more or transferring their knowledge (taking what you’ve learned and being able to use it in a realistic setting). Which brings me to my next point: we do have metrics we care about. They just aren’t page views or number of visitors or how much money we made on any given day or week or month or year. We just care if learners are learning. We care about items guru’d and items burned. And, we really care about transfer. We could always do a (way) better job of those things, but when making decisions, those are pretty much the only things we care about, talk about, and argue over.
I’m almost certain we (often?) make decisions that, short-term, hurt revenue. We make a decision that makes something feel uncomfortable, or makes learning more difficult. But, and hopefully you’ve realized this, the act of learning (if you’re actually learning, anyways) does not feel good. It’s uncomfortable.
When learning feels easy, you’re probably not learning in a way that will lead to transfer. That’s always been my criticism of a lot of learning systems out there—they prioritize the learner’s comfort over their learning. Looking way, way back, when we started prototyping WaniKani, the internal tagline for the entire system was “Dark Souls, but for learning kanji.” If you don’t know what that means, the basic idea was to make a kanji learning app that was exceedingly difficult, that really punished people’s mistakes. And, when you succeeded, you’d feel like you accomplished something, because you did! We certainly backed off a couple steps from that philosophy, but you can still feel some of that philosophy even today. Learning has to be hard, at least in the right ways, for learning to be effective.
Anyways, I digress. My point here is that the metrics we use for decision making are usually in conflict with the metrics that web analytics provide. At the same time, the data that exists in web analytics feels so safe and comfy, and so scientific! It’s so easy to get into decision paralysis by hiding behind it. It discourages us from thinking and imagining and using our knowledge. So, why not get rid of it, you know? Let’s live and have some fun. It removes friction from product development and makes it easier to think through the important metrics while making it more difficult to look at the irrelevant ones, focusing how we make our decisions.
And, cherry on top, I believe it’s the morally right thing to do.
So, duh. It’s gone. See yah! Please don’t smite us for our insolence, Goog. 🙇🏻